Wf)t Htbrarp ofttje Umber£ttp of J?orti) Carolina Collection of Motti) Caroltmana Cnbobjeb bp HFofcn g>prunt W\\ '- of the Class of 1889 -■W.mkfrr 4- This book must not be taken from the Library building. Digitized by the Internet Archive in 2012 with funding from University of North Carolina at Chapel Hill http://archive.org/details/riseofcottonmillmitchell X31 3 THE RISE OF COTTON MILLS IN THE SOUTH SERIES XXXIX NO. 2 JOHNS HOPKINS UNIVERSITY STUDIES IN Historical and Political Science Under the Direction of the Departments of History, Political Economy, and Political Science THE RISE OF COTTON MILLS IN THE SOUTH BY BROADUS MITCHELL, PH.D. Instructor in Political Economy BALTIMORE THE JOHNS HOPKINS PRESS Copyright 1921 THE JOHNS HOPKINS PRESS MESS OF THE NEW ERA PRINTINQ COMPANY LANCASTER, PA. CONTENTS Pagi Preface vii Chapter I. The Background 9 Chapter II. The Rise of the Mills 77 Chapter III. The Labor Factor 160 Chapter IV. The Role of Capital 232 0- PREFACE In prefacing some observations on the history of the South a writer has said: "It will be something if these papers shall make it plain that my subject is a true body of human life — a thing, and not a mass of facts, a topic in political science, an object lesson in large moralities. To know the thing itself should be our study ; and the right study of it is thought and passion, not research alone." 1 The same is true of the present story of the South's espousal of manufactures in place of whole devotion to agriculture. Rightly set forth, it is not only an industrial chronicle, but a romance, a drama as well. One who himself bore a part in the events here described, at the outset of my project hoped that I would grasp both the economic and the spiritual aspects of the period under review. 2 This I have tried to transmit to the reader, and I have found that the fuller the account of ma- terial circumstance, just so much the clearer becomes the spiritual significance. In point of view I owe most to my Father, accepting his concise explanation that the South was overcome at Appo- mattox because it placed itself in opposition to the compell- ing forces of the age — by agency of the invention of the cotton gin held to slavery instead of liberty, insisted upon States' rights in place of nationality, and chose agriculture alone rather than embracing the rising industrialism. As a result, the task since 1865 has been to liberalize the South in thought, nationalize it in politics, and industrialize it in production. " Would we make cotton king ? Let us aspire to spin every fibre of our exhaustless fields. By such align- ments with this wondrous mother-age, we shall enable the South to take her rightful part in determining the national iWilliam Garrott Brown, The Lower South in. American History. 2 Mr. J. C. Hemphill. Vlll THE RISE OF COTTON MILLS IN THE SOUTH destiny.'' 3 My study is little more than illustration of this analysis of the past, this interpretation of the present and future. Formerly, a landed aristocracy shut out the average man from economic participation ; but with the rise of cotton mills, the poor whites were welcomed back into the service of the South. As a conclusion from my survey I cannot but express the anxiety that through lessons of the old mistake we shall avoid the new error, insuring that an aristocracy of capital shall not now preclude industrial democracy. My purpose has been to describe the birth of the industry in the South rather than its development. In only a small number of instances has this pian~been departed from; many topics rich in interest have not been broached. I regret that two books did not come into my hands in time to be used in this study. Holland Thompson's " The New South," and George T. Winston's "A Builder of the New South'" (the story of the life work of D. A. Tomp- kins), are contributions which will! be found valuable. I owe thanks for special assistance to Professor Jacob H. Hollander and Professor George E. Barnett, of The Johns Hopkins University, who guided the investigation; the pro- prietors of the Manufacturers' Record, who permitted me to use the early files of the paper; Mr. T. S. Raworth, of Augusta; Mr. William M. Bird, of Charleston; Professor Yates Snowden and Mr. August Kohn, of Columbia, all of whom made documentary material available to me. Others have given me hardly less generously of their time and thought; footnote references to interviews and correspond- ence with these must serve as acknowledgment in each case. B. M. 3 Samuel Chiles Mitchell, " Educational Needs of the South," in The Outlook, N. Y., vol. lxxvi, no. 7, p. 415 ff. THE RISE OF COTTON MILLS IN THE SOUTH CHAPTER I The Background This opening chapter undertakes a brief survey of the historical and economic background out of which the cotton manufacturing industry of the South, as a distinct develop- ment, emerged. It may be said that thus to begin the story of the rise of the mills with discussion of a period which lies a century in advance, is not unlike the production of a play hopeful in conception, robust in theme and rapid in action, but in which the curtain first lifts to show a stage which, except for a few unrelated characters, remains empty throughout an entire act. It is a purpose here to refer to the views of some observers who believe they have caught glimpses of men and facts in these prior years not only presaging but causally related to the main action later. The total of this chapter will show, however, that the development, as such, first substantially showed itself and had its complete genesis about the year 1880. In the neglect of Southern economic history, information of the early period is not abundant, yet there is less dispute as to findings of fact than as to right interpretation of ma- terial evidences agreed upon. In bringing the several beliefs into parallel presentation it will be seen that concerning the rise of cotton mills in the South a little body of theory ex- ists. Several of the statements that will be given are not well-informed, and others are almost too studied, so that they lose perspective. Interpretations will be cited in con- nection with the different stages under discussion, so that the relative weighting of these stages, as intended by writers, will appear. IO THE RISE OF COTTON MILLS IN THE SOUTH [i l6 It is first useful to notice the limits of divergence of views. One who wrote with empirical purpose and may be believed to have been not deeply interested in the historical setting of the mills, has said of one State, taken by him as typical : " The story of the development of the cotton manu- facturing industry in South Carolina is not wanting in im- pressive elements. From the beginning in 1790 till 1900 it was a struggle of gradually increasing intensity and exten- sion." 1 This conception of continuity is in marked contrast with a representative expression of another Southerner likewise for some time a resident of the North. After re- ferring to promising industrial beginnings it is declared that : "... a manufacturing development throughout the Piedmont region of the South might have continued parallel with that which has taken place in Pennsylvania, except for the . . . combined influence of the invention of the cotton gin, the institution of slavery, and the checking of . . . im- migration. As late as 1810 the manufactured products of Virginia, the Carolinas and Georgia exceeded in value those of the entire New England states. By Whitney's invention . . . cotton planting became so profitable, that for a period of forty years the price remained above twenty>five cents a puuiuJL — Par^oTie^wereabandoned. ... As cotton and slavery advanced, the population of free white work people were driven further and further into the mountain country, and thus many of the white industrial workers of 1800 be- came the poor mountain farmers of 1850 . . . the owners of factories who operated with free white labor in 1800 had become in 1850 the cotton planters operating with black slave labor. . . . When the abolition of slavery re- moved one great difficulty of industries and the white peo- ple who had formerly deserted manufactures for agriculture went back to the pursuits of their fathers, these mountain- eers formed the labor supply." 2 1 P. H. Goldsmith, The Cotton Mill South, p. 4. 2 D. A. Tompkins, in The South in the Building of the Nation, vol. ii, p. 58. For a more summary statement, cf. ibid., Cotton Mill, Commercial Features, pp. 108-109. Cf. also ibid., History of Meek- / 117] THE BACKGROUND II Not so categorical as one opinion that " from 1810 to 1880 the South was industrially a desert of Sahara," this view still makes it clear that from a point early in the cen- tury until a date subsequent to the Civil War, absorption in cotton culture threw manufacturing of all sorts into the discard. There is sufficient evidence that in what may be roughly called the Revolutionary Period, the South was well started toward a balanced economic development, with manu- factures as well as agriculture. 3 In South Carolina early encouragement was given to the manufacture of cotton spe- cifically ; one Hugh Templeton, seeking inventor's privileges, in 1789 deposited with State authorities a plan for a carding machine and " a complete draft of a spinning machine, with eighty-four spindles, that will spin with one man's attendh ance ten pounds of good cotton yarn per day." 4 In 1795 the legislature authorized commissioners to project a lottery for the benefit of William McClure in his effort to establish a cotton manufactory to make " Manchester wares." 5 The lenburg County, vol. i, pp. 133-137; The Tariff and Reciprocity; Road Building and Repairs, p. 24; W. L. Trenholm, The Southern States, quoted in C. D. Wright, Industrial Evolution of the United States, pp. 145-146; J. A. B. Scherer, Cotton as a World Power, p. 168 ff.; Walter H. Page, The Rebuilding of Old Commonwealths, P- 139. 3 " Upon the whole, the last half of the Eighteenth century, before the influence of the cotton gin and Arkwright's inventions were fully felt in the South, was a period when agriculture yielded some ground in primary manufactures and household industries." (V. S. Clark, in South in Building of Nation, vol. v, p. 308). Cf. Tompkins, The South's Position in American Affairs, p. 1. Of North Carolina a careful student has said : " Though there were no towns of any size, the number and skill of the artisans was such that, in 1800, it seemed probable that the logical development would be into a frugal manu- facturing community, rather than into an agricultural state" (Hol- land Thompson, From the Cotton Field to the Cotton Mill, p. 25). See, especially with reference to iron making in this period, Richard H. Edmonds, Facts About the South (ed. 1894), p. 3 ff . There is importance in the founding of the Manumission Society, with 1600 active members as late as 1826 (ibid., pp. 26-27). 4 August Kohn, The Cotton Mills of South Carolina, pp. 10-11. 5 Ibid., pp. 9-10. In an appropriation bill of 1809, the sum of $1000 was advanced to Ephraim McBride " to enable him to con- struct a spinning machine on the principles mentioned in a patent he holds from the United States" (ibid., pp. 10-11). In the same 12 THE RISE OF COTTON MILLS IN THE SOUTH [ll8 South shared in the national impulse toward economic self sufficiency consequent upon the stoppage of colonial com- merce with England and the Revolution. Proceedings- of. (the. Safety Committee in Chowan County,. North Carolina, for March 4, 1775, show that "the committee met at the house of Captain James Sumner and the gentlemen ap~ pointed at a former meeting of directors to promote sub- scriptions for the encouragement of manufactures, informed the committee that the sum of eighty pounds sterling was subscribed by the inhabitants of this county for that laud- able purpose." The chairman offered ten pounds to the first producer in a certain time of fulled woolen cloth. The provincial congress took steps the same year to stimulate, by bounties, the manufacture of gunpowder, rolling and slitting mill products, cotton cards, steel, paper, woolen cloth and pig iron. 6 Although their objects were possibly political as well as industrial, mechanics'' societies existed at Charleston and Augusta before and about the year 1810; in Augusta were made some of the earliest attempts in this country to im- prove the steam engine. 7 As early as 1770 there was formed in South Carolina a committee to establish and promote manufactures, with Henry Laurens as chairman. 8 The pur- chase by Southern States of the patent rights of Whitney's cotton gin is to be interpreted not as a design to leave off cotton manufacturing, but rather as evidence of a prevalent spirit for mechanical improvement. Glimpses at individual establishments show the textile industry of the South in this Revolutionary Period to have year the request of the president of the Homespun Company of South Carolina for a loan on account of a patent was unfavorably received by a legislative committee, but it was recommended that he be allowed until the next meeting of the legislature "to report on the utility of the machine called the Columbia Spinster, so as to entitle, in case the same be approved, the inventor of the same to the sum provided by law for his benefit" (ibid., p. n). Cf. ibid., pp. n-13. 6 For these facts the writer is indebted to a MS. of M. R. Pleas- ants, " Manufacturing in North Carolina before i860." 7 Clark, in South in Building of Nation, vol. v, p. 310. 8 Kohn, Cotton Mills of South Carolina, p. 7. II9] THE BACKGROUND 13 been generally of the domestic character. Manufacturing was conducted by individuals rather than corporations, and was usually directly connected with plantations. Daniel Hey ward, a planter, in a letter in 1777, declared with refer- ence to his " manufactory " that if cards were to be had "there is not the least doubt but that we could make six thousand yards of good cloth in the year from the time we began." 9 Domestic production is clearly seen in a statement the same year that a planter in three months trained thirty negroes to make one hundred and twenty yards of cotton and woolen cloth per week, employing a white woman to instruct in spinning and a white man in weaving, and it was said : " He expects to have it in his power not only to clothe his own negroes, but soon to supply his neighbors." 10 A few plants may have approached a commercial char- acter. In 1790 it was related that "a gentleman of great mechanical knowledge and instructed in most of the branches of cotton manufactures in Europe, has already fixed, completed and now at work on the high hills of the Santee, near Stateburg, and which go by water, ginning ( ?) carding and slubbing machines, with 84 spindles each, and several other useful implements for manufacturing every necessary article in cotton." This establishment was coin- cident with Slater's famous factory at Pawtucket, Rhode Island, founded in 1790, and may have antedated it, though comparative credit to the Stateburg enterprise is perhaps diminished by information that while some long staple cot- ton was imported from' the West Indies, and a variety of goods were made, it was conducted as an adjunct to a plan- tation, parts of its equipment were later removed to and set up on another plantation, and much of its yarn was spun for persons in the vicinity. It is notable, however, that the machinery was made in North Carolina. 11 9 Ibid., p. 7. 10 South Carolina and American General Gazette, Jan. 30, 1770, quoted in ibid., p. 7. Cf. ibid., pp. 6-7. 11 American Museum, viii, Appendix iv, part 2, July 1, 1790, cited in ibid., p. 8. The question mark is Mr. Kohn's. If Mr. Kohn is correct in believing that " a regular cotton mill " was established by 14 THE RISE OF COTTON MILLS IN THE SOUTH [l20 The textile industry in the South in the latter part of the eighteenth and earlier part of the nineteenth centuries was stamped with the hallmark of domestic production. 12 How- ever, it is to be remembered that a century and a half ago this and other manufactures in every part of America and in England too bore very much of the domestic character, 13 and that probably Southern States showed instances of power-driven machinery before Slater set up the first Ark- wright mlill in Rhode Island. The South had planter- manufacturers it is true, but this link between agriculture and industry as contrasted with New England is easily ex- plained in the more general fertility of the soil and the effect this of course had upon the occupation of the people. Furthermore, 20 THE RISE OF COTTON MILLS IN THE SOUTH [l26 spicuous participation in the business since the early thirties, said : The mills built after the war were not the result of pre-bellum mills. This is trying to ascribe one cause for a condition which probably had many causes. The industrial awakening in the South was a natural reaction from the War and Reconstruction. Before the War there was first the domestic industry proper. Then came such small' mills about Winston-Salem as Cedar Falls and Frank- linsville. These little mills were themselves, however, hardly more than domestic manufactures. When, after the War, competition came from the North and from the larger Southern mills, the little mills which had operated before and had survived the war lost their advantage, which consisted in their possession of the local field. . . . The ante-bellum domestic-factory system did not produce the post-bellum mills. 29 It must be obvious from foregoing considerations that a census enumeration of mills of the period cannot show in- ternal characteristics which are all-important. But even the census returns, counting one plant like another, display _ the Southern industry at this stage as being feeble. Some primary descriptive factors are lacking in the earliest re- ports of the census which are at all useful, but taking the four Southern States which were farthest advanced in the year9 1840 and 1850 — Virginia, North and South Carolina and Georgia — and comparing 'the whole of the South with New England, the showing may be summed uip thus : 30 29 John W. Fries, ibid. It is not to be forgotten that lack of trans- portation facilities necessarily cramped the old mills, and that this operated also to keep out competing product, but their essential character was independent of this consideration. The superior trend of capital into agriculture limited ante-bellum cotton mills by pre- venting profitable extension of plant and embarrassing advantageous marketing of product which might require some waiting. Cf. Ed- ward Ingle, Southern Sidelights, pp. 70-71. Another with a broad view of the history of the industry was willing to include the Gran- iteville enterprise, about which some controversy has clustered, in his judgment: "The cotton mills in the South before the War were third-rate affairs. I speak of Graniteville and Batesville and such plants as these. I remember my mother's telling me that the warp . . . used to be supplied by the mills for use in the homes of the housewives. They were not regular cotton mills as the plants of later establishments have come to be" (W. W. Ball, int., Columbia, S. C, Jan. 1, 1917). "The mills built in the eighties were a part of a new spirit from the ante-bellum mills. The old mills — Bivings- ville, Valley Falls, Crawfordsville, in Spartanburg County — were small and insignificant affairs. They lived from hand to mouth " (Cleveland, int., Spartanburg). »» * 30 U. S. Census of Manufactures, 1900, "Cotton Manufactures," p. 127] THE BACKGROUND 21 Virginia N. Carolina S. Carolina Georgia So. States New England . . Census Plants I84O 22 1850 27 1840 25 1850 28 1840 15 1850 18 I840 19 1850 35 I84O 248 1850 166* 1840 674 1850 564 Capital $1,299,020 1,908,900 995.300 1,058,800 617,450 857,200 573,835 1.736,156 4,331,078 7,256,056 34,931,399 53,832,430 Ops. I,8l6 2,963 1,219 1,619 570 1,119 779 2,272 6,642 10,043 46,834 61,893 Spin. 42,262 47,934 531,903 16,353 42,589 i8o,927 5 1,497,394 Bales Consump- tion 17,785 (a) V 13,617 9,929 20,230 78,140 430,603 Many single mills in the South today represent more than the extent of the whole industry in the most forward Southern State in 1850. 31 n Some writers have pointed to evidences of industrial ac- tivity in the period to 1840 as presaging the later develop- ment. A localizing tendency in the textile manufacture along the fall line of rivers in the decade following 1830, has 'been called a "slow and unconscious development" 32 George Tucker in 1843 first pointed out that slavery was showing signs of decay from economic causes and as a sys- tem would finally lapse of its own accord. 33 A study of 54 ff. (a) Thompson gives 700 looms and 7000 bales consumed (p. 49 ff. (b) An obviously incomplete summary. 31 Cf. Thompson, p. 49 ff. " The number of small carding and fulling mills and of little water-driven yarn factories, in this section [the South] before 1850, may have approached the number of textile factories in the same region today; . . . but few of these establish- ments became commercial producers" (Clark, in South in Building of Nation, vol. v, pp. 310-320). A map showing distribution of cot- ton spindles in 1839 indicates a good representation for all the Southern States except Mississippi, Louisiana, Arkansas and Florida, as to mills of small size, but the localization both as to plants and spindles in New England is marked (Clark, History of Manufac- tures, pp. 533-560). See the whole section for an excellent discus- sion of both historical and economic phases). "Few mills south of Virginia had power looms prior to 1840" (ibid., in South in Building of Nation, vol. v, p. 321). Notice omission of looms for Southern States in census returns referred to above. 32 Clark, in South in Building of Nation, vol. v, p. 322. 33 « Progress of the United States in Population and Wealth in Fifty Years," referred to by William E. Dodd, in South in Building of Nation, vol. v, pp. 566-567. 22 THE RISE OF COTTON MILLS IN THE SOUTH £l28 North Carolina industrial history of the period has led to the conclusion that " The people of the state became inter- ested and soon a class of small manufacturers . . . came into prominence and continued to thrive down to i860." 34 It is questionable, however, whether it may be truly said that " the people of the state became interested " ; certainly there wasi nothing like the sweep of public sentiment that appeared in 1880, and the suggestions relied upon in mak- ing the inference show as much against as for the likelihood of their taking effect. 35 »t The foregoing paragraphs lead up to a more important judgment of Mr. Clark that " In the South the most strik- ing feature of this period [1840-1860] was the gradual breaking down of a traditional antipathy to manufactures. This hostility was opposed to the obvious interests of a region where idle white labor, abundant raw materials, and >t ever-present water-power seemed to unite conditions so ' 3i Pleasants. Reference is had especially to items in State papers and in Niles' Register. The Tarboro Free Press declared that should a tariff measure of the time meet with success, the people of the Carolinas would have to "join in the scuffle for the benefit anticipated from this new American system, and they will have to bear a portion of its burdens and buffet the Northern manufacturer with his own weapons." It is noticed that a report to the North Carolina legislature in the late twenties, looking back upon the dis- integrating process of the preceding two decades, said : " There must be a change. But how is this important revolution to be accom- plished? We unhesitatingly answer — by introducing the manufac- turing system into our own state and fabricating at least to the extent of our wants. . . . Our habits and prejudices are against manufacturing, but we must yield to the force of things and profit by the indications of nature. The policy that resists the change is u unwise and suicidal. Nothing else can restore us," 35 With preemption of land into large estates and consequent in- jury to small farming, discovery of gold, agitation for railroads and improvements in cotton manufacturing machinery, the people of Mecklenburg County, North Carolina, " many years before the war were beginning to realize the importance of diversified industries. . . . An industrial crisis was imminent, and the problem would have solved itself by natural agencies within a few more years, had not sectional differences brought on the war" (Tompkins, History of Mecklenburg, vol. i, p. 124). Cf. ibid., pp. 126-127; Kohn, Cotton Mills of S. C, pp. 18-19. That the war did come to render such an industrial impulse impossible of effects shows the relative weakness of the spirit at this time. The preoccupation with intersectional dif- ferences was of greater potency than the intrasectional change of mind, if such there were. I29] THE BACKGROUND 2$ favorable to textile industries. Cotton planting engaged the labor and the thought and capital of a directing white class, but the natural operative of the South remained un- employed, and the capital of the North and of Europe was mobile enough to flow to the point of maximum profit with- out regard to sectional or national lines, were such a profit known to be assured by Southern factories. Slavery as a system probably had less direct influence upon manufactures than is commonly supposed, but the presence of the negro through slavery was important." It is frankly recognized that white immigration from Eu- rope, which at this time supplied the most considerable me- chanical skill, avoided districts heavily populated with ne- groes ; that plantation self-sufficiency meant isolation with small need for good communicating roads; that the market for middle-grade goods was restricted by the servile character of the colored inhabitants ; that the credit system, by which factors controlled the direotioning of productive capital, rested upon cotton culture by negro labor; that while the corn laws held in England, reciprocity between the South- ern States and the mother country tended to discourage manufactures in this section while the conditions of com- merce favored manufacture in the North. " These business interests, supported by social traditions and political sec- tionalism, were strengthened in their opposition to new industries by a widespread papular prejudice against or- ganized manufactures. . . . Nevertheless the South chafed continually under the discomfort of an ill-balanced system of production. . . ." Mention is made of the canal at Au- gusta and of cotton mills at Charleston, Mobile, Columbus, New Orleans and Memphis directly following the writings to and object lesson of William Gregg in his Graniteville fac- tory, and it is concluded that "modern cotton manufactur- r. ing in the South date9 from the founding of Graniteville -x rather than from the post-bellum period. . . . However, viewed in comparison with the cotton manufactures of the North, those of the South were still insignificant. . . . if 24 THE RISE OF COTTON MILLS IN THE SOUTH [13O Nevertheless, the present attainment of the industry as- sured its definite future growth, and ultimate national im- portance." 36 It is not hard to justify disagreement with this view. The basis of probable industrial development before the War appears in hindsight only if the pervasive numbing influ- ence of slavery, made more powerful in the last years through the frantic effort at its maintenance through exten- sion, is forgotten. Well enough to assert that the capital of the North and of Europe was mobile enough to flow across the Atlantic and across Mason and Dixon's line were a profit in manufacture in the South known to be assured, but the fact is that capital did not -come in for industrial pur- poses because bright prospects had not been proved, and this largely because home enterprise was a laggard while slavery claimed the section's capital resources for cotton cultivation. 37 It is difficult to see the distinction which Mr. Clark desires to draw between the effect of the presence of the negro and the presence of slavery. While it is true that for long years after emancipation, and continuing to this day, the influence of the negro's presence in restraining in- flow of immigrants is evident, the lessening of this deterrent and the removal of nearly equal drawbacks could not pro- ceed or commence while slavery existed. From the point of " 36 Clark, History of Manufactures, p. 553 ff. Cf. ibid, in South in Building of Nation, vol. v, pp. 213-214, and p. 316 ff. Cf. Kohn (Cotton Mills of S. C, p. 16) : "The real and the lasting develop- 1 , ment of cotton mills in South Carolina might be started with the y Graniteville Cotton Mill. . . ." Cf. Gregg, Domestic Industry, pp. »• 24-25. 37 " Cheapness of cotton, abundance of water-power, the resources of the coal-fields, when steam began to supplant the dam, the other mineral resources, and the wealth of forests . . . did not even at- tract from other parts sufficient capital to develop the section to anything like its full extent. No artificial expedients were neces- sary there. But capital did not come" (Ingle, p. 73). A propa- " gandist of the early eighties, desiring to organize small cotton mills in the South, quoted with approval a correspondent of the Morning News of Savannah, declaring that before the War the planters saw the advantage for such establishments but were deterred from manufacturing because " slavery and the factory were declared to be incompatible institutions. They could not exist together " (W. H. Gannon, The Landowners of the South, and the Industrial Classes // of the North, p. 9 ff.). I31] THE BACKGROUND 2$ view of the independent white workman the presence of the negro in slavery held as a far more forcible objection than the presence of the negro in freedom. His killing economic competition and radiated social poison were beyond dispute and beyond prospect of remedy until he was made at least a free producer. Any prospect of immigration for the South has taken its rise from the Civil War. M It was slavery that made plantation self-sufficiency in primitive needs universal, that made isolation and physical barriers to intercourse. The credit system in its heyday rested in large degree upon supply by the factor of all in- dustrial products, which needs must be sustained so long as every local energy was foredoomed for absorption into H cotton growing. It cannot rightly be said that the traditional antipathy to ', manufactures was " opposed to the obvious interests of a region where idle white labor, abundant raw materials, and ever-present water-power seemed to unite conditions so favorable to textile industries," if it is meant that these in- terests, clear enough to us now, were obvious to Southern consciousness and purpose then. This applies particularly to the labor factor. It will be seen later that in the period"! before the War the mills often employed slaves as the ex- clusive operatives; in some cases negroes were employed with whites, and finally and more importantly, through Re- construction years and at the very outset of the cotton mill era the inclination of establishers of factories was frequently to engage negro hands and to induce operatives to come from the North and even from England and the Continent — overlooking the native white population as a useful sup- ply of workers as though it had not been there. Before the War the presence of raw cotton was certainly thought of rather as a guarantee of economic independence than as a stimulus to produce within the section those products of manufacturing which the staple was potent to purchase „ from outside. It is not implied that conspicuous promulgators and ex- 26 THE RISE OF COTTON MILLS IN THE SOUTH I 32 emplars of the need for a change in economic activity, such as William Gregg and some others, were not products of a reaction that showed itself from the long continuance of slavery, but they stand out, impotent as they are striking, against a dull and motionless background of prevalent sys- * tern. 1 They cried in a wilderness. Materials and viewpoint are both too well understood to require further demonstration of the preventive influence which slavery and cotton had upon industry in the South. Yet a few observations of Southern men are interesting h just at this point. Henry Watterson has said : " The South ! The South! It is no problem at all. The story of the South may be summed up in a sentence : she was rich, she lost her riches ; she was poor and in bondage ; she was set free, and she had to go to work ; she went to work, and she is richer than ever before. You see it was a groundhog case. The soil was here, the climate was here, but along with them was a curse, the curse of slavery." 38 Probably not over-induced by bitter animus is Helper's direct charge : i, In our opinion, an opinion which has been formed . . . from as- siduous researches, . . . the causes which have impeded the progress and prosperity of the South, which have dwindled our commerce . . . into the most contemptible insignificance; sunk a large majority of our people in galling poverty and ignorance, rendered a small minority conceited and tyrannical . . . ; entailed upon us a humiliat- ing dependence upon the Free States ; disgraced us in the recesses of our own souls, and brought us under reproach in the eyes of all civilized and enlightened nations — may all be traced to one common source, and there find solution in the most hateful and horrible word that was ever incorporated into the vocabulary of human r, economy — Slavery! 39 Tompkins saw clearly and in effect said again and again, " the result of the introduction and growth of the system of slavery was revolutionary; it turned the energies of the people almost wholly to the cultivation of cotton ; it prac- tically destroyed all other industries. . . ." 40 '' 38 Quoted by A. B. Hart, The Southern South, pp. 231-232. 39 Helper, p. 25. 40 Tompkins, History of Mecklenburg, vol. i, p. 100. " There were no industries requiring skill or thought, and there was no necessity for scientific farming or anything else scientific. . . . Slavery not only demonstrated that people will not think unless it is necessary, 1333 THE BACKGROUND 2"J Not only did slavery hold the South down to supplying the raw material, but while its baneful influence lasted few improvements were made in the methods or appliances even for the growing and preparation of cotton for the market. As in India and China today, the cheapness of labor made ingenuity, enterprise and machinery unnecessary. Except in size and superficial appearance there was no change in the ante-bellum gin, gin-house and screw from 1820 to i860. But after the War came a feeder, a condenser, a hand- press in the lint room, and cotton elevators. 41 If Cotton was King, the monarch was an imperious and but also that they will not work unless it is necessary (ibid., pp. 98-99). This statement is strongly influenced by Tench Coxe. It has been said of the Irish people by Lord Dufferin that "the entire nation flung itself back upon the land, with as fatal an impulse as when a river, whose current is suddenly impeded, rolls back and drowns the valley which it once fertilized." Sir Horace Plunkett comments : " The energies, the hopes, nay, the very existence of the race, became thus intimately bound up with agriculture" (Sir Horace Plunkett, Ireland in the New Century, p. 20). "By the influence of the negro the South lost its manufactures and largely its commerce, and became practically a purely agricultural section of the nation" (Tompkins, ibid., vol. ii, pp. 200-201 ; cf. ibid., Cotton Growing, pp. 3-4). As to the usefulness of negroes in latter-day cotton mills, this manufacturer advised : " Dependence upon the negro as a laborer has done infinite injury to the South. In the past it brought about a condition which drove the white laborer from the South or into enforced idleness. It is important to rees- tablish as quickly as possible respectability for white labor " (ibid., Cotton Mill, Commercial Features, pp. 109-110). Cf. ibid., Building and Loan Associations, p. 43 ; The Cultivation, Picking, Baling and Manufacturing of Cotton, from the Southern Standpoint, pp. 5-6; F. T. Carlton, History and Problems of Organized Labor, pp. 19-20. 41 " The cotton was packed by hand, carried into the gin-house in baskets by laborers, carried to the gin by laborers, pushed into the lint rooms, carried to the screw, packed in the box of the screw and bound with ropes, all by hand," but since the abolition of slavery " all the machinery and appliances for preparing cotton for the mar- ket have been revolutionized" (Tompkins, Cultivation, Picking, etc., of Cotton, pp. 5-6). See others of his writings for a full discussion of this point. Cf. M. B. Hammond, The Cotton Industry, pp. 77-78, and, for a detailed account of bad preparation of cotton down to 1880, Edward Atkinson, in U. S. Census of Manufactures, 1880, " The Cotton Manufacture," p. 4 ff. " No slave-holding people ever were an inventive people. In a slave-holding community the upper classes may become luxurious and polished; but never inventive. Whatever degrades the laborer and robs him of the fruits of his toil stifles the spirit of invention and forbids the utilization of inven- tions and discoveries even when made" (Henry George, Progress and Poverty, twenty-fifth anniversary ed., p. 523). 28 THE RISE OF COTTON MILLS IN THE SOUTH ^34 narrow-minded tyrant, who cramped the development and put blinders to the vision of the country. Said William Gregg in 1845 : " Since the discovery that cotton would mature in South-Carolina, she has reaped a golden harvest; but it is feared it has proved a curse rather than a blessing, and I believe that she would at this day be in a far better condition, had the discovery never been made. . . . Let us begin at once, before it is too late, to bring about a change in our industrial pursuit's. ... let croakers against enterprise be silenced. . . . Even Mr. Calhoun, our great oracle ... is against us in this matter; he will tell you, that no mechanical enterprise can succeed in South-Carolina . . . that to thrive in cotton spinning, one t should go to Rhode Island. . . , 42 " The invention of the cotton gin," wrote Tompkins, "... before i860 . . . was nearer anything else than a blessing. It was primarily responsible for the system of slavery. . . . Cotton ... in its manufacture ... is the life of the South, but we could probably have done as well with- out it until we began to manufacture it." 43 42 Domestic Industry, pp. 18-19. 43 History of Mecklenburg, vol. i, p. 194. For a careful descrip- " tion of the circumstances surrounding the invention of the cotton gin, and the legal documents in the dispute over the rights to it, cf. ibid., Cotton and Cotton Oil, pp. 19-31, and appendix. " We aban- doned a once leading factory system ; we imported slaves ; we let all public highways become quagmires; we destroyed every possibility for the farmer except cotton and by cutthroat competition amongst ourselves we reduced the price to where there was not a living in it for the cotton producer. We made cotton in a quantity and at a t price to clothe all the world excepting ourselves" (ibid., Road ' Building, p. 24). "The economic history of the South from the ,/ Revolution to the Civil War is a record of the development of one natural advantage to the neglect of several others. Fitted by nature " to support a large population engaged in a variety of pursuits based upon agriculture, it had a small population occupied in the produc- tion of raw material that contributed to the maintenance of a dense population in regions where artifice contended against harsh climate /fand a stubborn soil" (Ingle, p. 47). Cf. Burkett and Poe, Cotton, pp. 312 and 313; E. C. Brooks, The Story of Cotton, p. 157; Thomp- son, pp. 44; Miller and Millwright, quoted in Manufacturers' Record, Baltimore, Feb. 22, 1883. Gregg showed that cotton, the great god, drove agricultural enterprise from South Carolina, for, with the returns to its cultivation under ordinary management amounting to only 3 or 4 and in some instances only 2 per cent, the inclination for planters to remove with their slave capital to the richer Southwest was strong, thus keeping the population of the State at a standstill — - — {TXom£sticJLridustry ! _p J i8) . "Perhaps the most striking economic change that thenew industry [cotton culture] effected in the South 1 3 5^] THE BACKGROUND 29 The old South had much in common with mercantilist feeling. Though coin for coffers was not precisely the aim, there was the settled ambition for exportation of a money- crop that involved self-exploitation and left no room for sectional introspection. The economic system was full of inhibitions, the all-pervading effect of which cannot be cal- culated. In accounting in 1856 for the stagnation of Vir- ginia as compared with the industrial activity of New Eng- land and Old England, Olmsted wrote : It is the old, fettered, barbarian labor-system, in connection with ** which they [Virginians] have been brought up, against which all their enterprise must struggle, and with the chains of which all their ambition must be bound. This conviction ... is forced upon one more strongly than it is possible to make you comprehend by a mere statement of isolated facts. You could as well convey an idea of the effect of mist on a landscape, by enumerating the number of particles of vapor that obscure it. 44 Duping of the people through charlatan guidance of polit- ical leaders is too evident in the South of today to require description of its operation in an earlier period. 45 A re- after the reintroduction of slavery was the speedy abandonment of manufactures. . . . What was the use of nerve-racking investment in elaborate and costly machinery when a land-owner could reap ten per cent net profit from a few negroes and mules and a bushel or two of the magical cotton seed? And yet the South had unusual manufacturing facilities. . . ." (Scherer, p. 168 ff . ; cf. ibid., pp. 243, - ^4J_-lngl£ < __pfL_4Q 1 _i39 ; New York Herald, quoted in News and fc ^iirier r Xhaxleston, March 9, 1881 ; F. L. Olmsted, Seaboard Slave rt ates, p. i^ ff). The social difference between North and South before the War, so often remarked as existing of itself apart, is accounted for by slavery, which arrested development on Southern soil of the industrial type of American civilization (A. D. Mayo, in ^ The Social Economist, Oct., 1893, pp. 203-204). ^^ . 44 Olmsted, pp. 140-141, cf. ibid., p. 185 ; pp. 213-214. "The "*S amount of it, then, is this: Improvement and progress in South Carolina is forbidden by its present system" (ibid., pp. 522-523). And for his general philosophy of the subject, see ibid., pp. 490-491). He took as an average expression of the views " of the majority of those whose monopoly of wealth and knowledge has a governing influence on a majority of the people," the statement of a newspaper in 1854: ''African slavery ... is a thing that we cannot do without, that is righteous, profitable, and permanent, and that belongs to Southern society as inherently, intricately, and durably as the white race itself" (ibid., pp. 298-299). 45 There are many instances similar to that of a famous election speech in Virginia in the fifties, in which the aspirant declared to his audience : " Commerce has long ago spread her sails, and sailed \ 30 THE RISE OF COTTON MILLS IN THE SOUTH [136 flection as sorrowful, however, as the confirmed bias of the people shown in applause to such guidance, is the blindness of the leaders who, no doubt with strong elements of trick- ery, gave even stronger signs of being themselves duped by a situation. Not that the crowd was believing, but that spokesmen were so largely sincere, was most melancholy. The drug had ceased to lead to remorse, and began to bring hallucinations. 46 Approaches to rational statesmanship and reasonable moves toward balanced economic activity, found especially in the border States, could be nothing more than away from you . . . you have set no tilt-hammer of Vulcan to strike blows worthy of the gods in your iron-foundries; you have not yet spun more than coarse cotton enough, in the way of manufacture to clothe your own slaves. . . . You have rallied alone on the single power of agriculture — and such agriculture! . . . Instead of having to feed cattle on a thousand hills, you have had to chase the stump- tailed steer through the sedge-patches to procure a tough beef-steak (laughter and applause). . . . The landlord has skinned the tenant, and the tenant has skinned the land, until all have grown poor together." " And how," asks Olmsted, "does the fiddling Negro propose, it will be wondered, to remedy this so very amusing stu- pidity, poverty, and debility? Very simply and pleasantly. By building railroads and canals, ships and mills ; by establishing manu- factories, opening mines. . . . And, ' Hurrah ! ' shout the tickled electors ; ' that's exactly what we want' " And then he showed that it was much like the quack telling the confirmed paralytic to live generously, take vigorous exercise and grow well; that with the disease of slavery in its vitals the South could not do else than languish ; that in promising wholesome measures which contemplated everything but the attacking of slavery the politicians were just 1/ laughing at the people (Olmsted, p. 288 ff. ; cf. ibid., pp. 179-180). 46 A passage of Sir Horace Plunkett in comment upon Irish poli- tics is much to the point : " Deeply as I have felt for the past suffer- ings of the Irish people and their heritage of disability and distress, I could not bring myself to believe that, where misgovernment had continued so long, and in such an immense variety of circumstances and conditions, the governors could have been alone to blame. I envied those leaders of popular thought whose confidence in them- selves and in their fellows was shaken by no such reflections. But the more I listened to them, the more the conviction was borne in upon me that they were seeking to build an impossible future upon an imaginary past" (Ireland in New Century, p. 147). Cf. Tomp- kins, Cotton Mill, Commercial Features, preface to appendix, for an incident related of William Gregg and an opponent in an election campaign, which, despite its incidental happening, shows aptly just the point of preoccupation with politics to which the Southern mind came, the degree of trifling with which the most sober proposals were met, the hopelessness of change from this state of affairs by anything short of a fundamental moral awakening. I37] THE BACKGROUND 3 1 ineffectual stirrings while slavery persisted, and were less likely of success because the last years before the War, in which they emerged, were given over to such passionate, defiant advocacy of the " Southern institution." 47 The deterrent effect of slavery upon immigration has been noticed above. In i860 only 6 iper cent of the white population of the South was foreign-born, but immigrants made up nearly 20 per cent of that of the North. In the decade 1850-1860 the South's quota of foreign-born in the whole country dropped from 14 to 13 per cent. 48 Independent white artizans, so important in the industrial history of the North in this period, avoided competition with slave labor; if this drawback to coming to the South was removed by their acquiring slaves themselves where a few had the means, they must then leave mechanical pur- suits; many disapproved of slavery anyway. 49 Completer evidence of the damage wrought by slavery is the actual emigration of natives from the section when slaves were crowding; a portion of the population which under other circumstances might have taken root in industrial enterprise within the South was thus driven off. •1 47 " With the line around slavery being drawn more closely . . . the cotton South lagged in the industrial race, and the border States were hampered by the institution that they felt to be a burden, but which they could see no safe way to abolish. Compassed as it was by political compromises, slavery must ultimately have toppled through its own overweight; but in i860 it was so valuable for the plantation that it was not only not readily converted into the fac- tory, but was an obstacle in the way of the employment of capital and of other labor in that direction" (Ingle, pp. 68-69). f t 48 Ingle, p. 11. i, 49 Clark, in South in Building of Nation, vol. v, pp. 213-214. Southern whites were indisposed to welcome those who could not or refused to grow into the slavery system. A newspaper in the fifties betrayed this : " A large proportion of the mechanical force that migrate to the South, are a curse instead of a blessing; they are generally a worthless, unprincipled class — enemies to our peculiar institutions . . . pests to society, dangerous among the slave popu- lation, and ever ready to form combinations against the interest of the slaveholder, against the laws of the country, and against the peace of the Commonwealth." But slave-acquiring merchants were cordially received (quoted in Olmsted, p. 511). For interesting facts as to immigration to North Carolina, cf. Tompkins, History of ^Mecklenburg, vol. ii, p, 204; vol. i, p. 153. 32 THE RISE OF COTTON MILLS IN THE SOUTH Q 1 38 Communities with strong foreign infusion and slight or no reliance upon slavery, showed a vigor before the War which has been to them a continuing advantage into the present. 50 It was observed that competition of the slave was almost matched in hurtfulness by the example of the prosperous white man with whom acquisition of the com- forts and dignities of life did not proceed from daily toil. 51 The dependence of the ante-bellum South upon the North and upon Europe for the most substantial and trivial ap- purtenances of civilization was spectacular. It might be argued in apology for the total one-sidedness of the old South, that the section was responding to the principle of comparative economic advantage. Certainly the most ab- solute adherence to the territorial division of labor could not require a more exclusive devotion to the making of cotton and fuller reliance upon less peculiarly favored dis- tricts for manufactured goods and certain foodstuffs and materials, than the South displayed. But however strict in its conformity to superficial dictates of this policy, the pro- gram was ruinous to the section and the country, and was hurtful to the economic welfare of the world. Easy yield- 50 In the fifties it was declared that the most prosperous commu- nity in South Carolina was a settlement of Germans in the western part of the State. Here had been founded an educational institution, varied manufactures, farming was successful and capital was in- vested in a railroad venture. Slavery bore small part (Olmsted, p. 511). In 1865 the northwestern counties of Georgia, strongly op- posed to secession and which furnished soldiers to the federal armies, were held to be better disposed toward the national govern- ment than any other part of the State; slaves had constituted less than a fourth of the population. Though cruder than those from the seaboard, delegates from this section to the constitutional con- vention of 1865 were said to have a well-informed outlook for the Commonwealth (Sidney Andrews, The South Since the War, pp. 342-343). Study of the conventions of other States immediately succeeding the War shows " up-country " representatives, as con- trasted with those of the " low country," more easily adjusting themselves to the new condition and readier to go ahead with a changed program. It was said that at a time when the average wage of female operatives in Georgia cotton mills was half that paid in Massachusetts, New England factory girls were induced by high wages to go to the Southern State, but returned North because their position was unpleasant in " the general degradation of the laboring class" (Olmsted, p. 543). 31 Ibid., p. 201. 139] THE BACKGROUND 33 ing to the principle did not suggest to statesmen that the South after all was in only partial compliance — that even ifor the most efficient production of cotton as such there needed to be a wholesome admixture of manufacturing and of other agricultural interests. Post-bellum industry brought not a less but a more economical and larger output of the staple. The very humor of many passages in the literature of the economic history of the South, describing the need of the section to go to the North for a thousand and one essentials of daily existence, shows the seriousness of the situation. Gregg, too lonely in his advocacy of home industry to treat the subject in other than its fundamental aspects, declared: *' A change in our habits and industrial pursuits i9 a far greater desideratum than any change in the laws of our government . . . ," and " if we continue in our present habits, it would not be unreasonable to predict, that when the Raleigh Rail-Road is extended to Columbia, our mem- bers of the Legislature would be fed on Yankee baker's ■bread. Pardon me for repeating the call on South Carolina to go to work." His own city of Charleston, than which there was no greater sinner, had regulations against the employment of steam engines that stand 1 in striking con- trast to the arguments for the comparative advantage of steam as against water power at a later date when the city centered attention upon building a cotton factory. 52 52 " God speed the day when her [South Carolina's] politicians will be exhorting the people to domestic industry, instead of State resistance; when our Clay Clubs and Democratic Associations will be turned into societies for the advancement of scientific agriculture and the promotion of mechanic art; when our capitalists will be found following the example of Boston and other Northern cities, in making such investments of their capital as will give employment to the poor, and make them producers, instead of burthensome con- sumers ; when our City Council may become so enlightened as to see the propriety of following the example of every other city in the civilized world, in removing the restrictions on the use of the Steam Engine, now indispensable to every department of Manufacturing. . . ." And again : " He who has possessed himself of the notion that we have the industry, and are wronged out of our hard earnings by a lazy set of scheming Yankees, to get rid of this delusion, needs only seat himself on the Charleston wharves for a few days, and 3 34 THE RISE OF COTTON MILLS IN THE SOUTH [I4O A decade later Helper reproached a South that had not given heed to Gregg : " It is a fact well known to every in- telligent Southerner that we are compelled to go to the North for almost every article of utilty and adornment, from matches, shoepegs and paintings up to cotton-mills, steamships and statuary. . . . All the world sees, or ought to see, that in a commercial, mechanical, manufactural, fiancial and literary point of view, we are as helpless as babes. . . ." 53 Gregg remarked the supply by the North not only of the articles of major manufacture, but of those adjuncts of agriculture which would naturally be made within the South — axe, hoe and broom handles, pitchforks, rakes, hand-spikes, shingles and pine boards. 54 A newspaper in Richmond chronicled the sale to North- ern interests of a large coal field in the State, and in un- conscious irony placed in juxtaposition to the notice this confident exhortation: behold ship after ship arrive laden down with the various articles produced by Yankee industry" (Domestic Industry, p. 9ff.). ."The labor of negroes and blind horses can never supply the place of steam, and this power is withheld lest the smoke of an engine should disturb the delicate nerves of an agriculturist; or the noise of a mechanic's hammer should break in upon the slumber of a real estate holder, or importing merchant, while he is indulging in fanci- ful dreams, or building on paper, the Queen City of the South. . . ." (ibid., p. 23). 53 Helper, pp. 21, 23. Cf. for other interesting illustrations of de- pendence upon the North, some of which influenced Henry W. Grady. An orator at the Southern Commercial Convention, New Orleans, 1855, adapted for the occasion the famous speech in the British Parliament on taxes, and beginning, in the Southern version : " It is time that we should look about us, and see in what relation we stand to the North. From the rattle with which the nurse tickles the ear of the child born in the South, to the shroud that covers the cold form of the dead, everything comes to us from the North. We rise from between sheets made in Northern looms, and pillows of northern feathers, to wash in basins made in the North . . . ," and continuing in the strain which was a favorite with plat- form and pen, and many examples of which may be found (Olmsted, p. 544). Cf. Grady, New South, (ed. 1890), p. 188 ff. 64 Domestic Industry, p. 8; cf. ibid., p. II. Olmsted instances a case, probably common enough, where a North Carolina planter was buying hay grown in New York or New England with very large charges for carriage (pp. 378^-379). Cf. ibid., p. 175. When South- ern industrial resources were exploited, the total benefit might not come to the locality. Thus shipwrights at Mobile were from the North (Olmsted, p. 567). 141 ] THE BACKGROUND 35 '* It is plain that a new and glorious destiny awaits the South, and beckons us onward to a career of independence. Shall we train and discipline our energies for the coming crisis, or shall we continue the tributary and dependent vassals of Northern brokers and money- changers? Now is the time for the South to begin in earnest the work of self-development! Now is the time to break asunder the fetters of commercial subjection, and to prepare for that more com- plete independence that awaits us. 55 Other appeals to domestic industry were as clearly in- spired by sectional animosity; they were incidental to polit- ical ambition, and are to be contrasted with the generous, wholesome rallying-cries of the cotton mill campaign twenty- five years later, when economic sanity had gotten the better of partisan futilities. Another Virginia paper, wiser than that just quoted, urging manufacturing in the State and particularly textile mills for Richmond, anticipated with different mind the event invited by its contemporary, and foretold what was later too patent : " It must be plain in the South that if our relations with the North should ever be severed — and how soon they may be, none can know (may God avert it long!) — we would, in all the South, not be able to clothe ourselves. We could not fell our forests, plow our fields, nor mow our meadows. . . . And yet, with all these things staring us in the face, we shut our eyes, and go on blindfold." 56 In addition to the barrier to manufactures formed by 55 Olmsted, p. 363. 56 Ibid., p. 166. An " Address to the Farmers of Virginia," read at a convention for the formation of a State Agricultural Society in 1852, adopted, reconsidered and readopted with amendments, and finally reconsidered again and rejected on the ground that it con- tained admissions, however true, which would be useful to aboli- tionists, contained the words : " . . . thus we, who once swayed the councils of the Union, find our power gone, and our influence on the wane, at a time when both are of vital importance to our prosperity, if not to our safety. As other States accumulate the means of mate- rial greatness, and glide past us on the road to wealth and empire, we slight the warnings of dull statistics, and drive lazily along the field of ancient customs, or stop the plow to speed the politician — should we not, in too many cases, say . . . the demagogue? . . . With a wide-spread domain, with a kindly soil, with a climate whose sun radiates fertility, and whose very dews distill abundance, we find our inheritance so wasted that the eye aches to behold the pros- pect " (ibid., p. 169). 36 THE RISE OF COTTON MILLS IN THE SOUTH [^42 cotton cultivation under slave labor, and the silent opposi- tion which the prevalent system engendered, were not in- frequent outspoken declarations against industry. William Gregg was one of the few in the South to rise superior to Calhoun's sway, and asserting that there were some who were better able to speak of the propriety of factories than even that statesman, faced him squarely but tactfully: "The known zeal with which this gentleman has always engaged in every thing relating to the interest of South- Carolina, forbids the idea that he is not a friend to domestic manufactures, fairly brought about, and, knowing, as he must know, the influence which he exerts, he should be more guarded in expressing opinions adverse to so good a cause." 57 And again, speaking of manufactures, he was regretful of the fact that " our great men are not to be found in the ranks of those who are willing to lend their aid, in promoting this good cause. Are we to commence another ten years' crusade, to prepare the minds of the people of this State for revolution ; thus unhinging every department of industry, and paralyzing the best efforts to promote the welfare of our country ? " 58 57 Domestic Industry, p. 20. '* 58 Ibid., p. 14. "Lamentable, indeed, is it to see so wise and so pure a man as Langdon Cheves, putting forth the doctrine, to South- Carolina, that manufactures should be the last resort of a country. With the greatest possible respect for the opinions of this truly great man, and the humblest pretensions on my part, I will venture j, the assertion, that a greater error was never committed by a states- y' man" (ibid). The Southern Quarterly Review in 1845, the same year as Gregg's publication, quoted Cheves : " Manufactures should be the last resort of industry in every country, for one forced as with us, they serve no interests but those of the capitalists who set them in motion, and their immediate localities." And Mr. Kohn remarks, " This expression was not peculiar to any one class of leaders in South Carolina at that time," and instances other exam- ples (Cotton Mills of S. C, p. 13). Tompkins comments: ". . . as slavery grew, . . . there was a period from 1840 to i860, when the interest of the South sorely needed manufacturing as well as agricultural development. Only those men who appreciated this condition undertook to go counter to the growing sentiment in favor of agriculture and slave labor. Those who did continue to manufacture, were necessarily men of broad views and great abili- ties," and he speaks of some of the notable few — Gregg, Fries, Holt, Leak, Morehead, Hammett (Cotton Mill, Commercial Features, p. 180). Cf. also references to Burkett and Poe and to Brooks, n. 42. »4^ 143] THE BACKGROUND l"J In public-mindedness, in breadth of view, in qualities of imagination, in sanity of judgment that did not sacrifice un- derstanding of his misguided contemporaries, in power of analysis of the confronting situation, William Gregg stood . head and shoulders above other Southerners of his time. And only now, seventy-five years later, can his wisdom be thoroughly appreciated. The Lancashire opposition, which, despite the cotton famine, hated slavery and led to British disaffection when the warring South two decades after- wards most needed an ally, brilliantly vindicated his warnr ing to his antagonists that even their selfish ambitions could only be served by attention to such reasoning as he ad- vanced. Gregg said: Those who are disposed to agitate the State and prepare the minds of the people for resisting the laws of Congress, and particularly those who look for so direful a calamity as the dissolution of our Union, should, above all others, be most anxious so to diversify the industrial pursuits of South-Carolina, as to render her independent of all other countries ; for as sure as this greatest of calamities be- falls us, we shall find the same causes that produced it, making enemies of the nations which are at present the best customers of our agricultural productions. 59 Because of the striking reversal of front of the city at a later date, which will be of central importance in subse- quent chapters of this study, Gregg's advice to Charleston's capitalists in 1856 is interesting. Condemning, as a mem- ber of the legislature, a proposed subsidy to a railroad to link Charleston and Cincinnati, put forward in furtherance of commercial policies selfishly followed by " wealthy gen- tlemen, some of whom have ships floating in every sea," he declared that Charleston's destiny was " fixed and indis- soluble with the State of South-Carolina, and . . . mainly her great investment in Internal Improvements should be made with a view to developing the resources of the imme- diate country around her . . . cheap modes o.f transporta- 11 Cf. Gregg, Domestic Industry, pp. 19-20. For a very fine passage refuting _ Cheves' position and defining what the writer meant by " domestic manufactures " — not household industry, but cotton fac- tories throughout the State and craftsmen at every cross-roads — see "ibid., pp. 14-16. 59 Domestic Industry, p. 14 ; cf. ibid., p. 52. 38 THE RISE OF COTTON MILLS IN THE SOUTH [l44 tion from all quarters of the State could not fail to re-act on the general prosperity of the city . . . the dormant wealth of Charleston might be so directed as to be felt in the remotest parts of the State, in stimulating agriculture, draining our . . . swamps and putting into renewed cul- ture our worn-out and waste lands; diversified industry, stimulating the mechanic arts and increasing the population and wealth of the State." Instead of this he found that " there is no city in the Union which has accumulated more wealth, to its size, than Charleston — none that has shown so little inclination to develop the resources of the State. Her millionaires die in New York. There is scarcely a day that passes that does not send forth Charleston capital to add to the growth and wealth of that great city." 60 The characteristic inclination toward the individual rather than the corporate form of enterprise which was noticed as showing itself in the South of the Revolutionary Period, was still strong up to the Civil War. In 1845 Gregg inveighed against it, particularly as crystallized in legislative refusal to grant charters of incorporation; he was quick to hold up New England as a business model to the South. Those who have sought to magnify the industrial activities of the old South have frequently failed to take into account the differences in organization which distinguished enter- prises then from those of post-bellum years. The textile in- dustry could not be a movement in economic society, sink- ing its roots deep and extending them broadly, so long as investment participation sprang from and ended with indi- vidual initiative. Until the widespread emergence of the joint-stock form, the mills could not claim and embrace the generality of the community's resources. And in a period when this device was not largely turned to, it is plain that industrial stirrings were comparatively feeble. 61 60 Speech on Blue Ridge Railroad, p. 67. Cf. ibid., p. 29. 61 Gregg hoped that dangers to be apprehended from indiscrimi- nate granting of charters to banking institutions would "not be confounded with, and brought injuriously to bear against the char- ters which are necessary to develope [sic] the resources of our country, and give an impetus to all industrial pursuits. . . . The 14S2 THE BACKGROUND 39 " The individualism of the old South, the inability to co- operate was due no less to physical than social isolation be- tween portions of the population. Not only was there self- satisfaction coupled with dependence upon the North for manufactured commodities in the low-country, but the up-country, the frugal population of which was better dis- posed for manufacturing development, was so segregated as to be kept in mean state, or actually dependent itself upon the coastal districts. Between the Piedmont and the sea was the barrier of plantations ; between the Piedmont); and the industrial North were no transportation facilities. Concentration of capital, especially in the corporate form of industrial enterprise, is a mark of economic integration; in the ante-bellum South many other facts besides the absence of capital concentration show the lack of team work, of conditions making for unity of thought or action. 62 practice of operating by associated capital gives a wonderful stimulus to enterprise. . . . Why is it that the Bostonians are able in a day, or a week, to raise millions at one stroke, to purchase the land on both sides of a river, for miles, to secure a great water power and the erection of a manufacturing city? . . . The divine, lawyer, doc- tor, schoolmaster, guardian, widow, farmer, merchant, mechanic, common labourer, in fact, the whole community is made tributary to these great enterprises. The utility and safety of such institu- tions is no longer problematical. ... If we shut the door against associated capital and place reliance upon individual exertion, we may talk over the matter and grow poorer for fifty years to come, without effecting the change in our industrial pursuits, necessary to renovate the fortunes of our State. . . . About three-fourths of the manufacturing of the United States, is carried on by joint-stock companies ; ... we shall certainly have to look to such companies to introduce the business with us." He showed, by South Carolina examples, the perpetuity of the corporate form as contrasted with the frequently limited life of the personal enterprise (An Enquiry into the Propriety of Granting Charters of Incorporation for Manu- facturing and Other Purposes, in South Carolina, pp. 4-11). 62 " Isolation gave birth to individualism, as marked upon the mountain-clearing as upon the plantation ; and beginnings of the co- operative spirit were dwarfed by nature and by human inclination . . ." (Ingle, p. 32 ff.). Cf. Clark, in South in Building of Nation, vol. v, pp. 314-315. Olmsted found mountain wagons coming some- times two hundred miles to the head of navigation in North Caro- lina (p. 361 and pp. 358-359). The division of capital among small mills rather than its investment in larger factories is paralleled by the relatively larger number of church buildings in the South than in the North; with, however, relatively small seating capacity (Ingle, p. 32 ff.). The same tendency may be seen in respect to poorhouses, 40 THE RISE OF COTTON MILLS IN THE SOUTH {^146 The non-industrial character of the old South may be seen not only in internal fact, but in external reflection equally conclusive. Of external evidences, the political per- haps most readily occurs to one. Pervasive economic con- ditions come certainly to the surface in political pretensions ; economic transitions are registered in alterations of political front. The protective tariff of 1816 was introduced and defended, respectively, by two South Carolinians — Lowndes and Calhoun. The signature of a Virginia president — Madison — made it a law. This tariff was opposed by New England in the person of Webster. In 1828, in the debate over the "Tariff of Abominations," the situation was just the reverse — Calhoun opposed protection, Webster cham- pioned it. In swapping sides, both men were answering to the changed economic interests of their respective sections. 11 No clearer picture is needed of the trend of the South in ante-bellum years than the spectacle of Calhoun trans- formed from nationalist to sectionalist. 63 Cotton, nearly exclusively in the South, and to a notable degree in New England, was responsible underneath for the alterations which were displayed in the superficial play of politics. It was the disintegration of manufactures brought about by more and more extensive embracing of cotton cultivation that turned the South from protection to free trade;, it was the growing absorption in industry, espe- cially cotton manufacture, and relative relinquishing of commerce, that made New England protectionist instead of, as before, the champion of free trade. 64 asylums, hospitals and jails (Dodd, Expansion and Conflict, p. 231; cf. industrial map for i860, p. 188, showing few plants of an output of $250,000 south of Maryland). 63 Upon this whole matter, see Scherer, p. 179 ff. "In 1816, when Webster opposed protection, there was a capital of only about $52,000,000 invested in textile manufacture, of which much still lay in the South. In 1828, when he reversed his position, this capital had probably doubled, and had become localized in and about New England" (ibid., p. 181). Cf. ibid., p. 234. 64 Ibid., p. 152. Slavery added to cotton brought the extra confu- sion of purely political animosities. " At the beginning of the nine- teenth century the tariff was not a matter which was exclusively political. . . . The subject ceased to be an economic one and became 147] THE BACKGROUND 4 1 This is not the place to remark at length how economic interests are changing the South back, in partial measure, to the first position. Cotton is again central. Cotton fac- tories are largely responsible for the little leaven that is working in a large loaf, producing in the heart of the Solid South Republican adherents and voices for protection. " Slavery has been abolished. The South has reestablished manufactures. Its interests in free trade and protection are changed from what they were in i860. We need not only domestic trade, but foreign markets. We need, ap- parently, protection and free trade at the same time. . . . The South is as much interested in protection to home markets as New England is. New England is as much in- terested in export markets as the South is. In this situation we ought to get together . . . for ' Protection and Reci- procity.' " 65 It is interesting to examine a summary of the industrial history of the South in the fifty years preceding the Civil War, given by an important writer : it Between 1810 and i860 three periods of progress marked the fac- tory development of the cotton states. During our last war with England. . . . mill builders from the North migrated to the Southern highlands, and with local cooperation established small yarn factories at several places in the Carolinas, Georgia, Tennessee, and Kentucky. . . . During the decade ending with 1833, when hostility to the tariff made the Southern people bitterly resent economic dependence on the North, there was a second movement towards manufactures, especially in South Carolina and Georgia, directed mainly towards the erection of larger and more complete factories. This agitation bore fruit in some corporate enterprises, most of which had but qualified success. Finally, in the late forties real factory develop- ment began simultaneously at several points, and had not two finan- cial crises and a war checked its progress, we should probably date from this time the beginning of the modern epoch of cotton manu- facturing in the South. 66 a political one in proportion as slavery grew in the South and dimin- ished in the North, and in inverse proportion as manufactures dried up in the South and became of greater importance in the North" (Tompkins, The Tariff and Reciprocity). 65 Tompkins, Tariff and Protection. 66 Clark, in South in Building of Nation, vol. v, p. 316 ff. ; Cf . ibid., pp. 330r33i. Contrast Tompkins, History of Mecklenburg, vol. i, pp. 133-137- 42 THE RISE OF COTTON MILLS IN THE SOUTH [148 Two objections against this view have pertinence. In the first place, these three periods of comparative interest in manufactures can hardly be called "movements" in any social or economic sense. That of the twenties and running into the thirties may claim more color of this than the other two. 67 The plants set up by New Englanders earlier were in response to individual enterprise, and that enterprise born out of the boundaries of the South. Cooperation with the newcomers was not of the sort that marks the considerable interest of a community. To the extent that mills were built in the forties as a result of public agitation, William 67 But some of the agitation for industries in these, as in other years, had a flavor not symptomatic of healthy desire for improve- ment. Conventions looking to railroad development were held in North Carolina and Tennessee in the middle thirties. Of the advan- tages which it was agreed would flow from the building of the Charleston and Cincinnati Railroad, it was declared that " it will form a bond of union between the States [i.e., Southern States] which will give safety to our property and security to our institu- tions " (Tompkins, History of Mecklenburg, vol. i, p. 125). Of more positive character was the utterance of a Southerner who viewed with concern the danger that the North would crush slavery and place the South under complete submission to tariff aggressions, congressional representation for the latter section finding a stop in the limit of slave territory. " Under these circumstances, the true policy of the South is distinct and clearly marked. She must resort to the same means by which power is accumulated at the north, to secure it for herself." If the South should manufacture a large portion of its cotton crop " we reduce the quantity for export, and the competition for that remainder will add greatly to our wealth, while it will place us in a position to dictate our own terms. The manufactories will increase our population; increased population and wealth will enable us to chain the southern states proudly and indissolubly together by railroads and other internal improvements; and these works by affording a speedy communication from point to point, will prove our surest defense against either foreign aggres- sion or domestic revolt. ... If the evil day shall ever come when the south shall be satisfied that she cannot remain in the Union with safety to her institutions, it [i.e., industrial self-sufficiency] will place her in a condition to maintain her separate nationality" (E. Steadman, of Tennessee, quoted in J. D. B. DeBow, Industrial Re- sources of the South and Southwest, vol. ii, p. 127). Objection to massing poor whites in mills was combatted by a Charlestonian with the reflection that small farming with slave labor brought discontent that might mean social upheaval, whereas the factory opened a' door of opportunity making for stability; when poor whites should have the chance of owning a slave " they would increase the demand for that kind of property, and would become firm and uncompromising supporters of Southern institutions" (Ingle, pp. 25-26). 149] THE BACKGROUND 43 Gregg was almost wholly responsible. It has been pointed out above that Gregg was a missionary who preached an unaccepted faith. He was not a social exponent. In the second place, it is gratuitous to count upon what would have been the case had not the war broken in upon declared industrial beginnings. The Civil War was not a fortuitous event. It had to come. It was the disastrous evidence of the -dominance in the South of a system which gave no room to widespread industrial enterprise. Could the war be re- garded simply as an occurrence, an unfortunate happening, there would perhaps be ground for assuming that indus- trial enterprise might have been built into, and finally changed wholesomely, the economic regime of' the ante-bel- lum South, but facts show that it was a case where mastery between mutually exclusive plans had to be tried on the basis of comparative strength. The spirit for manufactures had not sufficient force to avert the war, but only enough life to show, in expiring, that it had begun to be born. i-» The decade 1850-1860 has been reserved for specific treatment at this point because two Southern writers have sought, rather dogmatically, to invest it with a character of industrialism! superior to that of ante-bellum years gen- erally and to show that it fathered later growth. Mr. Ed- monds has said: "A study of the facts . . . should con- vince anyone that the South in its early days gave close attention to manufacturing development, and that while later on the great profits in cultivation caused a contraction of the capital and energy of that section in farming opera- tions, yet, after 1850, there came renewed interest in in- dustrial matters, resulting in an astonishing advance an v railroad construction and in manufactures." 68 C 68 Edmonds, P-1&. It is shown how the course of cotton prices affected-irliluWyfTrom 1800 to 1839 cotton averaged a fraction over 17 cents; in 1840 the price dropped to 9 cents, continuing to decline to the 1846 average of 5.63 cents, when, after a short crop, there was a sharp rise in 1847, only to be followed by a fall to 8 cents and less. " These excessively low prices brought about a revival of public interest in other pursuits than cotton cultivation. . . ." It is said that from 1850 to i860 the South quadrupled its railroad mi- 44 THE RI SE 0F COTTON MILLS IN THE SOUTH £l$0 It is stated that " Cotton manufacturing had commenced to attract increased attention, and nearly $12,000,000 were invested in Southern cotton mills. In Georgia especially this industry was thriving, and between 1850 and i860 the capital so invested in that State nearly doubled." 69 The assertion that in i860 the South had in all 24,590 in- dustrial establishments with an investment of $175,000,000 loses force when, by a simple division, it is seen that on an average this made the investment in each only $7,144.37, which is surely not indicative of considerable importance. Many of the establishments must have been much smaller than would be represented by this average, and the few which were a great deal larger were rare exceptions. The very disparity in size of enterprises points away from any concerted movement toward manufacturing. As to the rail- roads, many of them were narrow-gauge, and all the facts tend to show that railroads were looked upon as facilitating commerce rather than manufactures. 70 In vaunting property figures of the South of i860 as compared with those of the North, Mr. Edmonds has given himself to the most obvious and serious error of including slaves. 71 Slaves, though in the legal sense agreed to belong leage, in the latter year being 387 miles in advance of New England (ibid., p. 10 ff.). For an account of late colonial and Revolutionary development, see ibid., p. 3 ff. Cf. DeBow, vol, iii, p. 76 ff. 69 Edmonds, Facts about South, p. 10 ff. Judging by the United States census of manufactures, these figures are grossly inaccurate. In i860 the Southern States had $9,840,221 invested in cotton manu- facturing, and in Georgia the investment increased from $1,736,156 in 1850 to $2,126,103 in i860, or less than 30 per cent (United States Census of Manufactures, 1900, Cotton Manufactures, p. 56). 70 Even after the war the pet scheme to build a railroad over the mountains gathered sentiment in the long-cherished desire to link Charleston with " the producing interior " typified in Cincinnati ; as rails were laid, piece-meal, through the Piedmont, advantages thus afforded for the erection of factories were seldom mentioned. The easier transport of cotton and the development of the South Atlantic ports were the thoughts uppermost. See above, p. 37. In the case of North Carolina, it is said that the railroads by bringing in manu- factures cheaper than local plants could supply them, actually hurt the advance of individual enterprise (Thompson, p. 31). 71 " Blot out of existence in one night every manufacturing enter- prise in the whole country, with all the capital employed [he was writing in 1894], and the loss would not equal that sustained by the I5l] THE BACKGROUND 45 to certain persons, were, socially and economically consid- ered, no more property and wealth than were their masters. In their emancipation the South did not lose, but gained, if their labor in freedom may be thought to be more produc- tive than when they were chattels. 72 Mr. Edmonds makes such over-zealous statements as that vy "The energy and enterprise displayed by the South in the extension of its agricultural interests was fully as great as the energy displayed in the development of New England's manufactures or that of the pioneers who opened; up the West to civilization,"' and greatly overreaches in his disap- proval of the phrase " The New South," " a term which is so popular everywhere except in the South, . . . supposed to represent a country of different idea9 and different busi- ness methods from those which prevailed in the old ante- bellum days. ... Its use ... as intended to convey the meaning that the South of late years is something entirely new and foreign to this section ... is wholly unjust to the South of the past and present. It needs but little inves- tigation to show that prior to the war the South was fully ft abreast of the times in all business interests." 73 His real purpose, which does not require ill-considered harking South as a result of the war. . . . New England and the Middle States, having grown rich by the war, almost trebled their property [from i860 to 1870] while the South drops from the first place to the third. In i860 it outranked the Northern section by $750,000,- 000." Mr. Edmonds does not note the inclusion of the slaves in his "property" figures (p. 18 ff.). In reference to the false idea ot pros- perity in the ante-bellum South, it has been said: "A delusion of great wealth was created in the listing as taxable property of slaves to the amount of at least two thousand millions" (Hart, p. 218). 72 " As commonly used the word ' wealth ' is applied to anything having an exchange value. But when used as a term of political economy it must be limited to a much more definite meaning, because many things are commonly spoken of as wealth which in taking account of collective or general wealth cannot be considered as wealth at all. . . . Such are slaves, whose value represents merely the power of one class to appropriate the earnings of another. . . . All this relative wealth, which, in common thought and speech, in legislation and law, is undistinguished from actual wealth, could, without the destruction or consumption of anything more than a few drops of ink and a piece of paper, be utterly annihilated" (George, PP. 38-39). -$c n 73 Edmonds, pp. 1-2. 46 THE RISE OF COTTON MILLS IN THE SOUTH [_l$ 2 back to ante-bellum years, is to show that "the wonderful industrial growth which has come since 1880 has been due mainly to Southern men and Southern money," and it is well to rest bis exposition with the proper statement that " Since 1880 " the people of the South " have turned to manufacturing with a facility that not only shows that they are in no way lacking in capability to compete in manufac- turing pursuits, but, considering the limited capital, this section has exhibited remarkable gains in developing its resources under adverse conditions. In a little more than a decade from the time the work of development may be said to have begun . . . nobody . . . doubts that the South can compete with New England in the manufacture of cotton goods, but many do doubt whether New England can com- pete with the South. . . ." 74 Edgar Garner Murphy embraced the viewpoint and made more categorical the statements of Mr. Edmonds respect- ing Southern industrial history. "The present industrial development of the South," he wrote, " is not a new crea- tion. It is chiefly a revival. . . . Instead of industrial in- action we find from the beginnings of Southern history an industrial movement, characteristic and sometimes even provincial in its methods, but presenting a consistent and creditable development up to the very hour of the Civil War. The issue of this war meant no mere economic re- versal. It meant economic catastrophe, drastic, desolate. . . . Thus the later story of the industrial South is but a story of reemergence." 75 The steps of Mr. Edmond's argu- 74 Ibid., p. 21. Cf. ibid., pp. 19-20. 75 E. G. Murphy, The Present South, p. 97. With modifications prompted by deeper study, Clark has presented about the same inter- pretation of the decade of the fifties as that of Edmonds and Murphy : " The South resented economic dependence, yet lacked the population, the experience, the capital and the habits that foster manufactures and diversify industries. It was topheavy with cotton, and slave agriculture unbalanced its economic life. . . . Yet had the war not intervened, manufactures would have revived and increased as settlement became denser, railways more numerous, and capital more abundant in proportion to resources, until these states by their own potency would have remoulded their industrial economy " (in South in Building of Nation, vol. v, pp. 330-331). For statements 153] THE BACKGROUND 47 merit are then repeated, except that Mr. Murphy failed to see the almost total lapse of industrial activity by 1840. The incentive to discover an industrial past for the sec- tion, which Mr. Edmonds found in the desire to establish the South as the magician of her post-bellum awakening, was matched in Mr. Murphy's motive by a more penetrat- ing purpose. In commenting upon the growing importance of manufactures as contrasted with agriculture, which was the most distinctive economic movement after 1880, he de- clared that " it is but one reassertion of the genius of the old South." Though his words boldly invite such a con- struction, it was outside of his object to mean by this that a genius for industrialism had run through the earlier history of the section. His true desire was to assert that " The old South was the real nucleus of the new nationalism," the old South in the sense of " the South of responsibility, the men of family, the planter class, the official soldiery, or (if you please) the aristocracy, — the South that had had power, and to whom power had taught those truths of life, those dignities and fidelities of temper, which power always teaches men. . . ." He regretted that this old South was not able to come into force until after Reconstruction be- cause " a doubt wa9 put upon its word given at Appomattox. . . . Power was struck from its hands. Its sense of respon- sibility was wounded and confused." 78 This is a fine statement of a primary truth in the devel- opment of the South that began about the year 1880. The old South did draw breath with the new. The permanent character of the South, the forces resident in the South of earlier as of later years, were those which largely made pos- sible a complete change in viewpoint, which carried through the measure of, if not indeed giving birth to, a reversed program. But, as Mr. Murphy did not see, there is a radi- cal distinction between the continuity of this quality in the probably influenced by Edmonds or Murphy, or both, see St. George L. Sioussat, in The History Teacher's Magazine, Sept., 1916, p. 224, and J. J. Spalding, in Proceedings, Fourth Annual Convention, Georgia Industrial Association, pp. 44-45. 76 Murphy, pp. 10-11. 48 THE RISE OF COTTON MILLS IN THE SOUTH £l 54 South and any continuity of its evidences in industrial pur- suits. The new South did not receive from the old South a heritage of industrial tradition; what it received was an ingrained and living social morality, not marred in its es- sential characteristics, and very likely, strange as it may appear, even assisted, by the institution of slavery. 77 Against some suggestions of an industrial character for the fifties, 78 may be placed much evidence of an opposite nature. Thus Hammond, of South Carolina, in the United. States Senate on March 4, 1858, goaded, perhaps, by the assaults of Helper and Seward, is found setting up figures of supposed per capita surplus production of the South as superior to those of the rest of the world, and forgetting that not wealth but economic power is the measure of the strength of a people. 79 The obsession with cotton, and the crazy confidence which the staple engendered, come out in the defiant valedictory which this spokesman flung to the North : " . . . would any sane nation make war on cotton? Without firing a gun, without drawing a sword, should they make war on us we could bring the whole world to our feet. . . . What would 77 " This sense of responsibility, deepened rather than destroyed by the burden of slavery, was the noble and fruitful gift of the old South to the new, a gift born of the conditions of an aristocracy, but responsive and operative under every challenge in the changing con- ditions of the later order" (ibid., p. 21). 78 A list of cotton factories in Alabama in 1852, the largest of which had only 3,080 spindles, is contained in DeBow, vol. i, p. 233. For a similar list for South Carolina in 1847, see Kohn, Cotton Mills of S. C, pp. 17-18; cf. Gregg, Domestic Industry, pp. 24-25. As to railroads, see DeBow, vol. iii, p. 76 ff. Where cotton mills were urged, the tone of the press might be casual as compared with that characterizing the later period of the eighties when advocacy was passionate ; e.g. : " We are glad to learn that our men of enterprise and capital are at length waking up on the subject. This is the best business that they could turn their attention to with the view of realizing profits . . . while at the same time it gives new life and energy to the surrounding community" (North Carolina Standard, Feb. 27, 1850, quoted in Pleasants MS.). 79 Scherer, p. 235 ff. Cf. Friedrich List, National System of Po- litical Economy. Hammond indulged largely in estimates ; as to untrustworthiness of census figures of wealth in these years, see Olmsted, pp. 512-513, and M. T. Copeland, The Cotton Manufactur- ing Industry of the United States, p. 18, note. 155]] THE BACKGROUND 49 -# happen if no cotton were furnished for three years? . . . England would topple headlong and carry the whole civil- ized world with her, save the South. No, you do not dare to make war on cotton. No power on earth dares to make war upon it. Cotton is- King." 80 Propaganda toward sweeping in Mexico and the Spanish West Indies to the Southern slavery system, when it became apparent by 1856 that further expansion in the West was impossible, paralleled the academic instruction given throughout his whole career by Professor Dew in the Col- lege of William and Mary. 81 il Ship-building, often urged because of superior advan- tages for the industry, did not take hold in the South. 82 In capital investment, presumption was against everything but cotton cultivation. Those who in the later period invested in manufactures were before the war slave holders. Fear that the presence of manufactures might undermine free trade tenets of the South had some influence against indus- try. 83 Only inhibitions against manufacturing as pervasive and unconsicous as they were effective can explain the sur- prise with which Southerners contemplated the failure of cotton mills set in the midst of cotton fields. 84 The pro- > 80 Quoted in Scherer, p. 235 ff. How little thought had been given to the South's economic self-sufficiency appears in this warning to the North: " The South have (sic) sustained you in a great measure. You are our factors. . . . Suppose we were to discharge you; sup- pose we were to take our business out of your hands; we would consign you to anarchy and poverty" (quoted in Scherer, p. 241). Cf. the spirited dissent from such thinking by Cassius M. Clay, as quoted in Helper, pp. 206-207. Hammond's views are readily am- plified by reference to proslavery writings, especially those of Christy, Bledsoe, Stringfellow, Harper, Dew. 81 Dodd, in South in Building of Nation, vol. v, p. 573. 82 Cf. Olmsted, p. 539, note, and table on p. 541 ; Ingle, pp. 70-71. 83 Cf. Ingle, pp. 70-71. " Of the twenty millions of dollars an- nually realized from the sales of the cotton crop of Alabama, nearly alt not expended in supporting the producers is reinvested in lands and negroes," and from this proceeded "senility and decay" (Hon. C. C. Clay, Jr., speaking to a horticultural society in 1855, quoted in Olmsted, p. 577). Cf. B. F. Perry, in address before S. C. Insti- tute, 1855, quoted in Helper, pp. 229-230. 84 Cf. Sparta, Ga., dispatch to Charleston News, July, 1855, in Olmsted, pp. 543-544. The decade 1850-1860 was the most pros- perous for the cotton industry in the country up to that time (Cope- 4 50 THE RISE OF COTTON MILLS IN THE SOUTH [156 portion of slaves in the ten cotton States was greater in i860 than in 1850, 85 the border States showed a positive in- crease in number of slaves, cotton planters of the older sec- tions gave themselves to breeding slaves for the Texas market, 86 and the amount of cultivated land increased 16.4 per cent. 87 The cotton crop of 1859-1860 was the largest to that time, being in excess of two billion bales. 88 No distincter picture of the growing trend in the South away from balanced economic development can be wished than that presented by the series of commercial conventions held in the fifteen year9 preceding the Civil War. The 1845 meeting, in Memphis, did not allow the recording of a proposition that the seat of government be removed to a place west of the Alleghanies, and passed a resolution af- firming that the convention " far from desiring to engender sectional prejudice . . . regard the North and the South, the East and the West, as one people, in sympathy and in interest, as in government and country " ; in accordance with the purpose to build up the South, the questions brought before the convention were at first of a practical nature, concerning commerce, manufactures and education. Gradually, however, border States ceased to send dele- gates, and the conventions were dominated by the political aims of the cotton belt, with politicians, rather than men of affairs, as spokesmen. Such practical measures as were discussed were on lines too broad to be capable of realiza- tion. They were such proposals as made resolutions rather than results. The South, so far as she sought industrial advancement, was in a maze, a novice not knowing to what projects to lend strength, never thinking of looking inward and never willing to start with homely enterprises that are suggested by genuine recognition of economic needs. It land, pp. 73-74). Following opening of railway communication after 1850, which brought in outside manufactures, " the home industry was, as a whole, distinctly less successful" (Thompson, p. 31). 85 Hammond, pp. 60-61. 86 Ibid., p. 59. 87 Ibid., p. 102, note 1. See table in ibid., p. 129. 88 Ibid., pp. 73-74- 157] THE BACKGROUND 51 was sought to secure the free navigation of the Amazon, to make passage of the Isthmus at Tehuantepec, to build a railroad from the Mississippi to the Pacific, to secure the introduction of slavery into Central America, to remove obstacles to filibustering plans in Nicaragua. Through the discussions in successive years at Charleston, New Orleans, Richmond, Savannah, Knoxville, Montgomery and Vicks- burg the tendency toward politics grew ; rather forced pro- nouncements of belief in the Union carried implication against their sincerity, and were mocked by speedy develop- ment of wrangles over the tariff into open use of the word " secession." " Hail Columbia " might be played at a ban- quet, but response was given to a toast, "The District of Columbia, the 'battleground for Southern institutions.'* Washington, and not the Southern States, drew the eye of all. 89 89 See Ingle, p. 221 ff. "... in all that they said there was an undertone of disappointment and apprehension. They wished to take part, but could not, in what was going forward in the rest of the country. They spoke hopefully of national enterprise, but it was evident that the nation of which they were thinking . . . was not the same nation that the Northern man had in mind when he thought of the future of industry" (Woodrow Wilson, Division and Reunion, p. 164). Cf. Scherer, p. 204. Cassius M. Clay in a speech in 1856 relentlessly pointed out the futility of all the plans proposed : " If there are not manufactures, there is no commerce. In vain dj> the slaveholders go to Knoxville, to Nashville, to Memphis and to Charleston, and resolve that they will have nothing to do with these abolition eighteen millions of Northern people ; that they will build their own vessels, manufacture their own goods, ship their own products to foreign countries, and break down New- York, Philadel- phia and Boston ! Again they resolve and resolve, and yet there is not a single more ton shipped and not a single article added to the wealth of the South. But . . . they never invite such men as I am to attend their Conventions. They know that I would tell them that slavery is the cause of their poverty, and that I will tell them that what they are aiming at is the dissolution of the Union. . . . They well know that by slave labor the very propositions which they make can never be realized ; yet when we show these things, they cry out, 'Oh, Cotton is King!"' (quoted in Helper, pp. 206-207). An ob- servation of Sir Horace Plunkett with respect to Irish leaders is peculiarly applicable here, if Irish nationalism be understood as paralleling true Southern economic needs : "... I always felt that an Irish night in the House of Commons was one of the strangest and most pathetic of spectacles. There were the veterans of the Irish party hardened by a hundred fights, ranging from Venezuela to the Soudan in search of battlefields, making allies of every kind 52 THE RISE OF COTTON MILLS IN THE SOUTH (^ 1 58 The 'bias of these last ante-bellum years, lashed to pas- sion by a guilty sectional conscience, or made more wild by the lack of any connected thinking, precluded even the pos- sibility of industrialism. When a gambler on the verge of ruin is desperately playing his last cards he has no time to reflect on past errors of judgment, and no inclination to think of better methods than the fortunes of chance by which to repair a pocket that likely momentarily will be emptied. 90 What did not occur to the leaders did not rise in the thoughts of the people. 91 Industrialism and the growth of cities are closely con- nected, yet in the decade of the fifties the advance in popu- lation of representative Southern cities was tardy as con- trasted with the North and West. 92 It has been noticed earlier that before the war even agri- culture was carried on with the roughest, least efficient tools, such as the "scooter," the "bull-tongue," the scraper, the sweep and hoe. 93 It is found that as late as 1880 patents issued to Southerners were for devices to be employed on of foreign potentate, from President Cleveland to the Mahdi, from Mr. Kruder, to the Akhoom of Swat, but looking with suspicion upon every symptom of an independent national movement in Ire- land ; masters of the language of hate and scorn, yet mocked by inevitable and eternal failure; winners of victories that turn to dust and ashes, devoted to their country, yet, from ignorance of the real source of the malady, ever widening the gaping wound through which its life-blood flows. . . . Irishmen have been long in realizing that . . . there are battles for Ireland to be fought and won in Ire- land " (p. 91 ff.). 90 " . . . the Irish mind has been in regard to economics, politics, and even some phases of religious influence, a mind warped and diseased, deprived of good nutrition and fed on fancies or fictions, out of which no genuine growth, industrial or other was possible " (Plunkett, pp. 122-123). 91 At the height of this period Helper wrote : " . . . the stupid and sequacious masses, the white victims of slavery . . . believe, what- ever the slaveholders tell them ; and thus it is that they are cajoled into the notion that they are the freest, happiest and most intelligent people in the world, and are taught to look with prejudice and dis- approbation upon every new principle or progressive movement. Thue it is that the South, woefully inert and inventionless, has lagged behind the North, and is now weltering in the cesspool of ignorance and degradation" (pp. 44-45. Cf. Page, pp. 22-23). 92 Ingle, pp. 14-15. 93 Cf. Hammond, pp. 77-78. Cf. George, pp. 522-523. 1 59J THE BACKGROUND 53 •the farm or in the home rather than in mechanical pursuits, thus arguing against any considerable industrial tradition or stirrings before that date. 94 Gregg warned the South that as surely as she separated from the Union, she would find herself economically un- equipped to maintain her position. His words were real- ized with bitter force. The trial of the war showed how far industry had been neglected. It tore away in an instant a veil of fiction, and showed deplorable fact beneath. Not even the immediate needs of an army, in munitions and ordnance, could be met within the South. Clothing for soldiery and people was lacking, shipyards were small; transportation was insufficient. When cotton could no longer bring in the manufactures of others, the South was left without essentials. 95 It has been seen how lacking was the ante-bellum South in any industrial character, and how some tendencies in this direction, showing themselves in the years just before the outbreak of conflict, were choked off or perverted by polit- ical motive in the rapidly growing hostility to the North. The Civil War, which brought into* glaring view the ab- sence of Southern economic self-sufficiency, cleared the 94 Under date of Nov. 14, 1882, the patent for a loom shuttle was issued to D. A. Willbanks, High Shoals, Ga., but this is the only invention connected with cotton manufacturing revealed by a search of patent lists for many weeks (Baltimore Journal of Com- merce and Manufacturers' Record, Nov. 18, 1882). Typical lists of patents issued in the same year to Southerners show : cultivator, saw gin filing machine, vehicle wheel, quilting attachment for sewing machines, rotary engine, couch, combined cotton-planter and fer- tilizer-distributor, grate feeder, paint, devices for holding the fingers in writing, hoe, animal trap, bottle washer, automatic fly can, spoke socket, cotton chopper, coffee roaster, revolving plow, bread cutter, etc. (ibid., Sept. 26, 1882, and Nov. 4, 1882). 95 " The story of manufactures in the South from i860 to 1865 is a record of the efforts of a people, deprived in large measure of the materials that satisfy their needs, to supply themselves without pre- vious preparation with the equipment of war and the resources of peace" (Clark, in South in Building of Nation, vol. v, pp. 330-331). Cf. Scherer, p. 260, note; R. D. Stewart, "firearms of the Con- federacy," in Magazine of Antique Firearms, Dec, 191 1; Tompkins, Tariff, p. 5 ; Thompson, p. 55 ; ibid., p. 44. It is significant that the exigency was met only by leaning heavily upon domestic household production. See above, p. 35. 54 THE RISE OF COTTON MILLS IN THE SOUTH [l60 fevered, suffocating atmosphere like an electric storm. Mis- conceived sectional political ambition and fierce protest had ridden to a fall; talent and energy theretofore absorbed to such ends were freed for wholesome introspection and ma- terial upbuilding. The Civil War set at rest the political inconclusiveness of the Union, which had operated so harm- fully for the South. The political bee, which had been en- couraged to buzz in the Southern bonnet by the planter par- ticularism, was silenced. 96 This was the first condition of economic advance. Besides the negative effect of the war, through the issue of the struggle the South was drawn into the national life, and thu9 was given positive stimulus through the industrial example of the North and East. 97 With the removal of political obsession vanished its co- hort, slavery; slavery gone, it not only became apparent that the South had to change tactics, but that it could change tactics. Thus practical pointings were not more powerful than mental consequents — not just the slaves, but the South as a whole was emancipated. 98 96 Southerners "now renewed once and for all time their allegiance to the Union which had up to that time been an experiment, a gov- ernment of uncertain powers " (Dodd, Expansion and Conflict, p. 328). 97 " The planter culture, the semi-feudalism of the ' old South ' was annihilated, while the industrial and financial system of the East was triumphant. . . . the east was the mistress of the United States, and the social and economic ideals of that section were to be stamped permanently upon the country" (ibid., p. 328. On the non- industrial quality of the ante-bellum South, see also ibid., pp. 214- 215). After emancipation, "the Southern people felt themselves in the throes of an economic revolution leading to a future of diver- sified industries. The old sentiment in favor of agriculture sur- vived ; but faith in it as the sole support of a nation was disappear- ing. The wealth and power which the North had derived from manufactures was better appreciated" (Clark, in South in Building of Nation, vol. vi, p. 254). 98 As will be seen later, new opportunities and duties did not break on the South with full force at first. What the war made possible, however, is seen in the following striking statement of a Southern periodical some years afterward : "... it has been a very common thing ... we all know, for one generation after another in southern cities ... to beguile the monotony of their humdrum life with rosy day-dreams of a far-off greatness that has been always coming but has never come. At last, however, since the annihilation of the institution of slavery, the new awakening of the world under l6l~] THE BACKGROUND 55 It will be seen in a later chapter, in examining the wide- spread building of cotton mills, how completely the South was altered in economic outlook after the Civil War. Not the least satisfactory evidence of this changed character is in the frank avowal of it by Southerners on every hand. The war was in Southern economic history a watershed. In 1882 a publisher in the heart of the South could say: "The old sectional spirit is dying out. You can find few men now who hold the narrow views of former years." 99 The newness of cotton manufacture, as of industry gen- erally, to the post-bellum South is evidenced in the type of enterprisers who entered the field when its opportunities were understood. There were few experienced men upon whom to rely ; it is safe to say that after the war more of the men projecting cotton mills came from any one of the accustomed callings of agriculture, commerce and the pro- fessions than from industry. 100 Before the war such propa- the intelligent energies of an age of unprecedented progress, the delusive mirage now disappears ; and the desert of hope in the South begins truly to grow green with ... a harvest that is really ripening before the impoverished people who have so long been looking for it and have been so drearily disappointed. ... At last we know that the South need no longer be nodding, and dreaming, and drooping, over the faded hopes that have for ages attended her traditions ; but, under the auspices of a new order of things, that her people have to go on only a little further with the same heroic endurance and the same brave energies now characterizing them, to realize in all its fullness and all its force the great established and imperishable fact that the old Slave States of the Union — themselves emancipated from the industrial incubus of an institution which contracted their spirit of enterprise, enfeebled their energies, and smothered all their industries except that of agriculture, — are now at last standing straight and strong, with a cheering consciousness of their native power in the bounties God has given them. . . ." (Industrial South, Richmond, quoted in Baltimore Journal of Commerce and Mfgrs. Record, June 17, 1882). Cf. Clark, in South in Building of Nation, vol. vi, p. 254, and Grady, p. 270. Tompkins said of one community now noted for its manufactures, " The effect of emancipation upon all classes of industrial life was immediate and revolutionary," and attributed the interest in factories chiefly to abolition of slavery (History of Mecklenburg, vol. i, p. 150. Cf. ibid., pp. 151, 194-106). 99 Patrick Walsh, of Augusta Chronicle, quoted in Journal of Commerce and Manufacturers' Record, Baltimore, Sept. 30, 1882. 100 See Goldsmith, pp. 7-8; Clark, in South in Building of Nation, vol. vi, pp. 266-267; Tompkins, Cotton Mill, Commercial Features, p. 180; and the present writer, in Manufacturers' Record, Baltimore, May 10, 1917. 56 THE RISE OF COTTON MILLS IN THE SOUTH \_\62 gandists as DeBow, hammering away in his Review for railways, cotton manufactures and direct trade with Europe, were pitifully in the minority. After the war, such adher- ents of the old order as Bledsoe ridiculed industrialism in vain ; warnings against making the " New South " only an- other North made small appeal to thinking men who cher- ished precisely this ambition. 101 How great is the temptation to conceive and attempt to carry through political and social reforms which are really contingent upon economic reorganization, is nowhere more clearly seen than in the period of Reconstruction in the South. These years, filled with the clamor of jealousy and vindictiveness and hurt and passion and greed needed so much of wisdom and patience and, above all, work. For- tunately, economic processes by some magic can usually, however uncertainly, go forward in spite of every political hindrance; the South, if hearing with one ear insults from without, listened with the other to voices from within. The degree of distraction and torment of Reconstruction tes- tifies to the strength of purpose with which the South at- tended to her own best promptings. It may even be held, perhaps, that Reconstruction, in a certain point of view, wa9 of positive assistance in nurturing the mind for indus- trial beginnings. There was no question but that the South was exhausted and was being drained of all but self-respect ; she was humbled beyond compassion. Former slaves were apparently becoming masters. As a participant in national- ity, in appreciation of broad social policies, the South knew that she had made a terrible failure. The fierce pride of the first war years had waned into the hopeless, dogged re- sistance of the days before Appomattox and flickered out in the degradation that followed. During Reconstruction the South, like a man thrown into prison, had time to reflect on past sins. Though perhaps it was not admitted in word, 101 See Dodd, in South in Building of Nation, vol. vi, p. 546. For an excellent account of post-bellum activity as contrasted with ante- bellum quiescence, see Tompkins, History of Mecklenburg, vol. i, pp. 150-151, 194-196; Clark, in South in Building of Nation, vol. vi, pp. 262-263. 163] THE BACKGROUND 57 it was soon to foe shown in deed that the South understood the part that slavery had played. A new course must surely thenceforth foe adopted. In Reconstruction the South found itself. Not without the material assistance and more gen- erous view that came through agency of Northern men who in this period learned to know the industrial opportunities of the section and were willing to contribute toward its de- velopment, it was still primarily a change of heart which the South experienced. In the face of a freed negro popu- lation, the idea of work first seriously presented itself to the Southern white mind. Lack of tangible evidences of this psychological change should not hinder understanding of its presence. During Reconstruction little that was practical could be done, but how earnestly the South had been introspecting and plan- ning is splendidly apparent in the suddenness and vigor with which industrial development commenced once im- pediments were removed. 102 It will be seen later that no agency bore a larger part in the rise of cotton mills in the South than the News and Courier, of Charleston. It is therefore important to know that, according to a statement made by the paper in 1880, on the very eve of the great development, its philosophy of 102 Mr. Clark has well called Reconstruction " a germinal period for manufactures." For a sympathetic interpretation of the mean- ing of Reconstruction years, see Clark, in South in Building of Na- tion, vol. vi, pp. 254-255, 262-263, 265-266. Grady wrote in 1889, speaking principally of the period of Reconstruction : " For twenty- five years the industrial forces of the South have been at work under the surface. Making little show, experimenting, working out new ways, peering about with the lamp of experience barely lit, dig- ging, delving, struggling, until at last the day has come, and inde- pendence is proclaimed. Now watch the change take place with al- most comical swiftness " (p. 270) . One cannot' but second the ap- peal of Professor Sioussat : " The political history of reconstruction has been narrated from many points of view, . . . but the vast social and economic changes, which beginning in the reconstruction time are still in progress, usually receive in our text-books less attention. Our girls and boys study carefully the work of the Gracchi, the organization of the medieval manor . . . and the condition of the peasants in France before the revolution. Is it not possible to awaken an intelligent interest in the tasks with which emancipation and the industrial revolution have confronted the people of the South?" (p. 223). 58 THE RISE OF COTTON MILLS IN THE SOUTH [164 manufactures had been conceived in the thick of Reconstruc- tion. " Ten years ago," it was said, " The News and Cour- ier formulated what is now an accepted truth, in declaring that the remedy for commercial distress in the North and the secret of sure fortune in the South was to bring the mills to the cotton." The thought was not balked by the small success of ante-bellum factories, one of which, established in Charleston long before the war, was at the date of this writing "in the irony of fate, the City Alms-House"; nor was it unassisted by the presence of men in the State "who understood that large profits could be made by well-man- aged cotton factories." There were at the close of the con- flict such mills as Graniteville and Batesville which were gaining reputation, and another important venture was being projected. Around these a body of thought, favorable to manufactures, and new to the South, grew up, and "the expectation of profit, which in those days had something of a theoretical basis," was by 1880 able to stand upon " a solid foundation, supported by . . . indisputable and con- vincing facts. . . ." 103 103 p e b I0> 1880. A South Carolinian, reminded of the cotton mill boom of the early eighties, led by the press, said " the South had begun to develop and revive before 1880. The papers probably stressed a program which they had already seen started " (M. L. Bonham, interview, Anderson, S. C, Sept. 10, 1916). "No appre- ciable break occurred in the continuity of cotton manufactures in the South, in spite of the mills destroyed or closed by the war. Be- fore 1870 several of the ruined factories had been rebuilt, and long prior to that others had resumed operations. ... In 1868 . . . there were sixty-nine mills ... in operation south of the Ohio and Po- tomac. . . . By 1870 Southern mill owners were confident they could make yarn five cents a pound cheaper than the Northern factories " (Clark, in South in Building of Nation, vol. vi, pp. 254-255). Cf. ibid., pp. 262-263. The News and Courier declared that " nothing did more to show the practical advantages of a cotton producing State in this matter than the calculation made and published a num- ber of years ago by the President of the Saluda Factory, which showed by actual figures that South Carolina mills could sell ordi- nary yarns in New York at the price which it cost the mills in New England to make these yarns, and still realize a considerable profit " (ibid.). See a list of mills in operation in South Carolina two years after the war, published in an almanac of Joseph Walker, Charleston, quoted in Kohn, Cotton Mills of South Carolina, p. 19. With reference to the fifteen years following the war, see Thomp- son, p. 59 ff. For a sketch of the career of H. P. Hammett, typical 165] THE BACKGROUND 59 We may leave now the period of Reconstruction, with its formative influences, and come to the evidence bespeaking material proof of industrial beginnings after political hin- drances were removed, economic strength was being re- gained and the South could concentrate on its task of manufactures. 104 The Southern States, though regaining self -government generally about 1876, did not get economic freedom of action with political rights. Later, in another connection, it will be shown how the issue of the Hayes- Tilden presidential election helped to delay for four years industrial beginnings. But aside from this, waving the wand of civic independence could not produce cotton mills immediately from a magic hat. Additional years of recov- ery were necessary, years far from idle, but not marked by widespread activity. The war saw a fevered South com- pletely stricken ; during radical rule the victim lay on a bed of torture; while convalescent after 1876, the patient did not comence to sit up and take solid food until about 1880. There is every reason for selecting the year 1880 as the beginning of cotton manufacturing development in the South. Negatively, foregoing pages have shown that it did not exist, in a proper sense, earlier. Remaining parts of this study will exhibit very positive evidences of alertness and progress after that date. Though there are material bases for grounding the genesis in the year 1880, it is not meant to insist dogmatically upon this precise point of time. of the South Carolinians who after the war understood that a profit could be made from well-managed cotton mills, and who in the sixties and seventies was mayor of Greenville, a member of the House of Representatives, a railroad president and mill builder, see Tompkins, Cotton Mill, Commercial Features, pp. 180-190. Renewal of cotton manufacturing in the South closely following disappear- ance of slavery was generally on old lines and with old machinery, but Hammett's Piedmont Factory was " designed, built and equipped after strictly modern plans" (ibid.). 104 " "While some retrospect is necessary [in studying the history of the New South] the period . . . covered is principally that which began with the close of the reconstruction era, at the time when the South was permitted once more to exercise self-government, and when some progress had been made toward repairing the economic losses of the war" (Sioussat, pp. 223, 228). Cf. Tompkins, Tariff, P- 3- 60 THE RISE OF COTTON MILLS IN THE SOUTH \_l66 Certainly, however, much in the way of convenience would be sacrificed by choosing 1879 or 1881. Writers touching the subject, whether careful students or casual comimen^ tators, have very generally selected this date as the initiation of the cotton mill era. 105 105 « The scope of the history of Southern progress along indus- trial lines is embraced mostly within the last twenty-five years " (T. C. Guthrie, in Proceedings, 7th Annual Convention, Southern Cotton Spinners' Assn., 1903, p. 44). See this and following pages for an extraordinarily good interpretation of stages antecedent to the rise of the mills. The suddenness with which development began is indi- cated : " If some soothsayer . . . twenty-five years ago . . . had essayed to predict what the South would accomplish in industrial development . . . and particularly in cotton manufacturing; if he had foretold the hundreds of millions of capital that would be in- vested ; the number of mills ; the number of spindles ; . . . the quan- tity of cotton consumed each year ; the number of operatives ; the value of the annual output — if he had prophesied concerning the meeting here today, the capital, labor, values and territory repre- sented here, he would have been set down as a dreamer of dreams." Another speaker at the same convention referred to slavery as turn- ing back the clock of progress, which, however, started ticking off industrial advance after 1880 (Averill, ibid., pp. 123-124). Noticing the decrease in price of cotton from 1870 to 1879 from 23 to 10 cents, the growing impatience with unreliable freed negroes, the movement of people of means to the cities and willingness to invest in other things than mortgages, Mr. Thompson assumed the same date of commencement (p. 59 ff.). Cf. E. C. Brooks, Story of Cotton, p. 215. Professor Brooks prefers 1880 as the date of the Southern economic renaissance (interview, Durham, N. C, Sept. 18, 1916), and his Story of Cotton shows this clearly, as, e.g., " It was in 1880 . . . that the Southern states turned seriously to manufacturing cot- ton" (p. 261) ; he gives a table from which he says "It is apparent . . . that the real factory life in the South dates from 1880 ... ;" "... a new era started in the South about 1880 . . ." (p. 257) ; " The whole civilization of the South had been overturned, ... a new era in regard to the value of skilled labor and personal worth was taking the place of the old notions . . . and we have the begin- ning of the factory system in the South" (pp. 255-256). Mr. Gold- smith calls the year 1880 "epoch-marking" and declares it "marks the turning point in the development of modern cotton factories in the South. ... A new era dawned" (pp. 4-5). Tompkins related the third period in Southern population history to " the industrial expansion which grew from the business revival . . . following the war," and quoted figures from 1880 (History of Mecklenburg, vol. r ) P- !97)- Murphy put stress upon a psychological reversal which argued industrialism : " About the year 1880 the long-waited change begins. By 1890 the industrial revival is in evident progress. By 1900 the South had entered upon one of the most remarkable periods of economic development to be found in the history of the modern industrial world" (pp. 101-102). "From the ashes and ruins left by the war a ' new South ' has emerged. Between the cessation of i6 7 ] THE BACKGROUND 6 I Innumerable evidences of the newness of the South to cotton manufacture in 1880 crop out, making it clear that united building of mills cannot be placed before that date. hostilities and the beginning of this development, a period of fifteen years, the South had slowly recovered from the losses which it had suffered. . . . The cotton manufacturing industry has grown up in the South . . . since 1880" (Copeland, pp. 32, 34). "The revolu- tion, . . . the evolution on the ' double quick,' began about 1880 in South Carolina. . . ." (Kohn, Cotton Mills of S. C, p. 20). Cf. ibid., pp. 18-19. " One of the most remarkable features in the indus- trial history of the Southern States has been the phenomenal growth of cotton manufactures there . . .; from 1880-1890 the number of spindles increased twofold . . , whilst in the following decade the growth was still greater. . . ." (T. W. Uttley, Cotton Spinning and Manufacturing in the United States of America, p. 43). This selec- tion of 1880 is by an English student. Some references far from studied are especially confirmatory; often a painter will half close his eyes to discern tone values : " United States Census figures show that since 1880 the consumption of cotton in mills in the cotton grow- ing states has increased 1,502 per cent. . . ." (Advertisement of Southern Railway in Textile Manufacturer, Charlotte, N. C, Aug. 19, 1915). "In other words, since 1880 the investment in Southern cotton mills has increased from less than fifteen million dollars to more than three hundred and fifty million dollars " (John A. Law, in Proceedings, Robert Morris Club, National Association of Credit Men, 1916, pp. 18-19). Cf. Henry D. Phillips, in The South Mobil- izing for Social Service, p. 566; Hart, pp. 224, 232, 242. " It will be seen that the South has been taking stock since 1880, and that eco- nomic forces and influences are now better understood than ever be- fore . . ." (Dodd, in South in Building of Nation, vol. vi, p. 550). " Mills were established in Spartanburg County first in 1879 and 1880 in numbers. About these years was the first great activity. The County was crushed before 1879. Before 1876 there was no capital, and the domination of the carpet bag government" (Cleveland, int., Spartanburg). For a looser statement, hardly to be taken in contra- diction, see Tompkins, Cotton Mill, Commercial Features, preface. The year 1880 marks not only the beginnings of cotton manufactur- ing, but was signalized by recovery or new enterprise in other direc- tions. Ante-bellum cotton production of over 5,000,000 bales had been reached again (Sioussat, p. 227, and News and Observer, Raleigh, N. C, Sept. 15, 1880) ; Tennessee and Alabama boom towns, resting on hopes of iron and steel manufacture, came a little later (Sioussat, ibid.) ; railroad development took its rise (Hart, p. 227) ; "... it was not until amost 1880 that the public-school idea was accepted as the best solution of the educational problem" (U. S. Bureau of Education, Negro Education, 1917) ; furniture and vehicle factories appeared in the upland, hardwood sections (Brooks, p. 217) ; agricultural method and rural life began undergoing reorgani- zation and betterment (ibid., pp. 221-222) ; public interest in cotton seed oil manufacture started with 1882 (Tompkins, Cotton and Cot- ton Oil, pp. 210, 214) ; right of suffrage was withdrawn from illiterate whites and negroes (ibid., p. 64) ; as to good roads, see Tompkins, History of Mecklenburg, vol. ii, p. 213; the speculation 62 THE RISE OF COTTON MILLS IN THE SOUTH £l68 In this year only one establishment in South Carolina was located within the corporate limits of a city. 106 Descrip- tions of cotton manufacturing processes had to be of the most primary sort, without technical language. 107 Lack of specialization and even the link with domestic industry showed in at least one conspicuous instance as late ag^ 1880. 108 How largely thought of industrial matters was delayed until 1880 by the issue of the Hayes-Tilden contest will be seen in detail later. 109 Contributing to the lateness of the economic awakening was the fact that South Carolina, which proved so strong in leadership when the movement commenced, was one of the last States to be freed from carpet-bag rule. The panic of 1873 and the following depression may be considered alone sufficient cause for the failure of these years to show more industrial progress in the South. From the combined causes of war, paper money, and scarcity of cotton, the price of the staple and of manufac- turing machinery soared to monstrous figures, and did not return again to the level of i860 until about 1880. 110 In a list of the thirty cities having the largest gross manu- of 1879 was held to have set in motion European and American spindles (Commercial and Financial Chronicle, quoted in News and Courier, Charleston, Sept. 12, 1881) ; "The cotton-manufacturing industry in almost every part of the world has continued to* prosper during the past twelve months" (Financial and Commercial Chron- icle, quoted in Baltimore Journal of Commerce and Manufacturers' Record, Sept. 9, 1882) ; " . . . the sudden and wonderful revival of business which took place in the republic during the last half of 1879 . . . had the effect of withdrawing us from the foreign markets to supply our home demands " (American Rail and Export Journal, quoted in ibid., Aug. 26, 1882). 106 J. K. Blackman, The Cotton Mills of South Carolina, p. 13. 107 See as to Clement Attachment, Daily Constitution, Atlanta, Jan. 23, 1880. 108 In connection with the Glendale Factory, D. E. Converse & Co. operated a flouring mill, several gins, a saw and planing mill, and a wool carding mill in which upwards of 10,000 pounds of wool was prepared for the country people (Blackman, p. 10). 109 See especially, however, correspondence signed " Local," in News and Observer, Raleigh, N. C, Nov. 21, 1880, and quotations from New York Herald and Washington Post in News and Courier, Charleston, March 8, 1881. 110 U. S. Census of Manufactures, 1880, " Cotton Manufacture " p. 8. 1 69] THE BACKGROUND 63 facturing product, the census of 1880 enumerated none in the South, unless Baltimore and St. Louis be counted, and in neither of these did cotton manufacture rank with their six principal industries. 111 Census figures, inconclusive when examined for particular aspects of the history of the cotton manufacture, show strikingly, when taken for a considerable period, that the Southern industry had its rise in 1880. The following table, covering the years 1850 to 1900 inclusive, gives the course of the mills of the South as exhibited in the most salient features: 112 Year Estab. Capital Opera- tives Spin. Looms Lbs. Cotton I9OO I89O I880 I870 i860 I850 40I 239 I6l 151 165 166 $124,596,874 53,821,303 17,375,897 11,088,315 9,840,221 7,256,056 97,559 36,415 16,741 io,i73 10,152 10,043 4,299,988 1,554,000 542,048 327,871 298,551 IIO,OI5 36,266 11,898 6,256 8,789 707,842,111 250,837,646 84,528,757 34,351,195 45,786,510 That 1880 was the date of commencement, clearly seen in this tabulation, is also interestingly apparent in interpre- tations of the figures brought out in successive census re- ports. No better picture of the way in which the Southern development broke on the national consciousness can be had than by a glance at some of these comments seriatim. As has been said, up to 1880 the Southern industry had evidenced no extraordinary or convincing advance. It is natural, therefore, to find the census of this year remarking on the degree of Southern growth merely as an extension of the manufacture, and classing the Southern mills with some new ones in the West. 113 111 " Remarks on the Statistics of Manufactures," p. xxvii. 112 U. S. Census of Manufactures, 1900, " Cotton Manufactures," p. 57. These figures, strictly taken, indicate the decade, rather than the year, of commencement of striking growth. Comments in the census and other evidence, however, fill in the outline here pre- sented. 113 " Th e cotton manufacture is almost monopolized by New Eng- land, Massachusetts alone producing to the value of $74,780,835. The other New England states produce in the aggregate about as 64 THE RISE OF COTTON MILLS IN THE SOUTH [17O As will be seen later, Edward Atkinson, of Boston, had much to do with rousing the South to economic activity. However, he admitted Southern industrial prospects only when he could not urge a superior advantage in New Eng- land or when he knew that to do otherwise would be futile. His comments in the census of 1880 are interestingly in- dicative oi his frame of mind. Dwelling on the new through rail connections in this country, he computed in pound- cents the saving of New England over Lancashire in raw cotton ; recognizing that this argument of relative proximity to cotton fields proved too much, applying with greater force to the Southern States, he was obliged to say that "If Georgia has twice the advantage over Lancashire that New England now possesses, it will only be the fault of the peo- ple of Georgia if they do not reap the benefit of it." 114 He went on to assert, somewhat contradictorily, that " The charge for moving cotton is becoming less year by year, and it will soon matter little where the cotton factory is placed, so far as distance between the field and the factory is con- cerned," and suggested that this allowed location of mills so as to utilize assets in climate, labor, and repair facilities much more. . . ." And in the list of States producing in excess of $2,000,000 each are mentioned Georgia, Maryland, New Jersey, New , /York, North Carolina, Pennsylvania, South Carolina (U. S. Census * of Manufactures, 1880, " Remarks on the Statistics of Manufac- tures," by Francis A. Walker, pp. xix-xx). Two obvious advantages of Southern mills seemed to be sufficient cause for greater per- centage of increase in that section than in other sections. "... tables indicate the rapid extension of the cotton manufacture to the southern states, where the cotton is at hand and labor is much cheaper than at the north." Southern spindles increased from 1870 to 1880 by 65 + per cent, in New England 57 per cent, in the Middle States 11 + per cent, in the Western States 46 -f- per cent, and in the whole country 49 -f- per cent. " It will be seen that the states of Michigan, Wisconsin, and Minnesota have been added to the list of cotton manufacturing states since 1870" (ibid., " The Factory System of the United States," by Carroll D. Wright, p. 16). "After the success of the power loom the cotton manufacture took rapid strides. . . . Factories sprung up on all the streams of Yorkshire and Lanca- shire, . . . while in this country the activity of the promoters . . , won cities from barren pastures. They erected Lowell, Lawrence, Holyoke, Fall River . . . and now in this generation the industry is taking root upon the banks of Southern streams" (ibid., p. 8). 114 U. S. Census of Manufactures, 1880, " The Cotton Manufac- ture," p. 12. Cf. p. 13. I7l] THE BACKGROUND 65 which were possessed by New England ; that the lowest cost of production existed where wages were highest. 115 He was fond of trying to center the attention of the South on the "preparation" of cotton rather than on its manu- facture. Thus he declared that ginning, which must be car- ried on among the plantations, " is the most important de- partment in the whole series of operations to which the cotton fiber must be subjected; and as yet there has been less of science and art . . . applied to this department than to any other." He exhibited in much detail, on the basis of a private investigation made before the census year, the careless and wasteful way in which cotton was handled in the Southern gins and " screws," but was obliged to admit that by 1883, when his report was transmitted, "the old methods, by which the cotton has been depreciated after it had been picked, are rapidly going out of use." This was partly by agency of the Atlanta cotton exposition of 1881, in which he had been a prime mover, and which it is clear he hoped might direct efforts increasingly to the growing of the staple in the uplands, and the utilization of seed for its oil and food substances. 118 The position taken in this study, that the Southern cotton manufacturing development really began in 1880, receives striking justification in the comments on the statistics of the industry by Edward Stanwood in the census of 1890. In the figures collected in this year the Southern develop- ment since 1880, as contrasted with the previous record of the section, and as compared with the proportionate ad- vance of other seats of the manufacture, was too apparent to be accorded other than frank avowal, leading to specula- tion as to chances of the rest of the country in maintaining accustomed superiority. " The geographical distribution of the cotton manufacturing industry is an interesting study," Stanwood said, "and it is more especially so at the present time by the fact that during the last ten years a change has been taking place, which, if it should continue, will become 115 Ibid., p. 14. " 116 Ibid., p. 4 ff. 5 66 THE RISE OF COTTON MILLS IN THE SOUTH [l72 highly important." He recited that from the beginning New England) had been chief in the industry, in 1840 having 70 per cent of the spinning machinery, in i860 (spindles were not taken in 1850) 74 per cent, J J per cent in 1870, and 81 per cent in 1880. The 1890 census, however, showed for New England a drop to 76 per cent. In the face of this decrease, he enlarged on the steadiness of concentration in certain New England districts and the success with which Massachusetts alone had maintained its percentage of spin- dleage increase. But, in spite of having added 2,000,000 spindles, New England was a relative loser by nearly 5 per cent, and for the first time in the census occurs the heading " Growth in the South." And it is declared : In considering the geographical distribution of the cotton manu- facturing industry the most important fact is the extraordinary rate of its growth in the South during the past decade. For a great many years, probably ever since the cultivation of the cotton plant in the South Atlantic states had a beginning, domestic spinning and weaving of coarse cotton fabrics has been a common fact in the household economy of that part of the country. Here and there small factories were established for the production of heavy fabrics. It is only in the period since the close of the civil war that mills have been erected in the South for the purpose of entering the general market of the country with their merchandise, and almost all the progress made in this direction has been effected since 1880. It was remarked that the 1880 census showed for all the States south of the District of Columbia, only 542,048 spin- dles, and that had all these been concentrated in one State it would have raised that State only to seventh place in point of production capacity. " A remarkable development of manufacturing enterprise in the South, based on the near- ness of supplies of raw material, which began ten years ago, had no more reasonable field in which to exercise itself than that of cotton spinning. New mills sprang up all over the region, but particularly in the states of North Carolina, South Carolina, and Georgia." In 1890 these three States reported 75 more establishments than in 1880, but even this did not indicate the increase, because some antiquated mills had ceased operation forever, and the average number of spindles to the mill had advanced nearly 73 per cent. 173] THE BACKGROUND 6? Quite as large proportionate increase had taken place in other Southern States ; markets previously in exclusive pos- session of Northern mills had been occupied by Southern products, finer goods were being manufactured and the new mills were " for the most part equipped with the latest and most improved machinery." Outstanding Southern ad- vantages were partially offset by disadvantages, " some of which time and experience will cause to disappear," and, in place of Atkinson's determined preference for New Eng- land, it was declared that " It can not be doubted that the development of this industry in the cotton-raising states is based upon sound commercial reasons, and that it is destined to continue." Increase of manufacturing in the Middle States had been at a slower rate than in any other part of the country, and the development in the. West, while ex- hibiting a good rate of advance, was too small to call for extended notice. While it was recognized that the future growth of the industry, considered geographically, de- pended upon a variety of factors — cheapness of transpor- tation of raw cotton, nearness to markets for finished goods, economy of power, supply of adaptable labor, spirit of State laws, and, perhaps, degree of humidity — the South was not held to be militated against in any of these respects. 117 By the time Stanwood came to analyze the figures of cotton manufacture for the 1900 census, events had further 117 U. S. Census of Manufactures, 1890, "Cotton Manufacture," by Edward Stanwood, pp. 171-172. As will later appear, a good deal had been made of the alleged disadvantage of the South in not having a sufficiently humid climate, but Stanwood showed that the superiority possessed in this particular by the British Isles hadbeen overcome in American mills through use of artificial humidifiers. The whole of his estimate in this census report is interesting as indi- cating how the southern development was breaking on the national consciousness ; special New England localities were given praise, but the rise of the South as a cotton manufacturing section held promi- nent place in the writer's thought. In addition to the percentage increases in spindles, it is important to notice that in looms, repre- senting completer commencement of capture of the industry, the percentage advance in the United States was 43, in the Middle States 28, Western States 85, New England 35, and in the Southern States was 204 (ibid., p. 171). 68 THE RISE OF COTTON MILLS IN THE SOUTH [_ 1 74 clarified his thought. 118 Covered apology for New England in stress laid upon records of special localities, such as that of Providence County, Rhode Island, which had more spindles than any Southern States except South Carolina, had to give way to the frank assertion that " the percentage of New England as a whole has suffered a considerable de- cline," from 81 in 1880 to 76 in 1890 to 67.6 in 1900. 119 " The growth of the industry in the South is the one great fact in its history during the past ten years." From 1880 to 1890 the number of establishments advanced 48.4 per cent, from the latter date to 1900 the increase was 67.4 per cent, and the size of mills had easily kept pace. The interpretation of the growth of the Southern industry rep- resents one of the earliest conscious attempts at scrutiny with desire to analyze — Southern cotton manufacture had become not only a fact, but a fact to be studied, appre- ciated, understood. 120 \J Comments on returns in the 1910 census showed the per- 118 U. S. Census of Manufactures, 1900, Cotton Manufactures, pp. 28-29. 119 Decrease in number of establishments in New England and the Middle States was said to be more apparent than real, by reason of consolidation of plants and changes in census classification. The Western States were shown to work under disadvantages which dis- missed them from further solicitude. Cf. ibid., p. 48. 120 " Speaking broadly, the cotton manufacturing industry did not exist in the South before the Civil War, and it existed on only the most restricted scale before 1880. ... It is probably not an exagger- ation to say that prior to 1880 there was not a mill south of the latitude of Washington that would be classed as an efficient modern cotton factory, even according to the standard of that time. Before the Civil War the people of the South were almost exclusively en- gaged in agricultural pursuits. After the war closed it was some years before the people had recovered sufficiently from the disaster to undertake manufacturing." Extended reference to the effects of the Atlanta cotton exposition, the character of the cotton mill cam- paign, and the lessons learned in matters of machinery will be noticed in another place. It was remarked that the South was making experiments of value to the whole industry, the first and, for some time, the only electrically operated factory being in that section. Instead of the former speculation as to the permanence of Southern mills, it was declared that " The fact that after a phe- nomenal growth during more than twenty years the expansion of old mills and the erection of new ones are still going on in the South is ample proof of the success of the enterprise," and the steady in- crease in spindles is given by years. 175]] THE BACKGROUND 69 centages of increase in the leading Southern States to be decidedly greater than those in Northern States, but South was merged with North as going to make up the nearly exclusive seat of the industry, the East. Records of indi- vidual Southern States are intermixed with those of States of New England, the former having come into proper com- parison with the latter in point of absolute importance. 121 Census reports uncovered fully, after a period of time, facts which were in part contemporaneously recognized. The following chapter will exhibit this proclaiming of a new day in the South of 1880 in detail; but the whole study really is a justification of the assertion that this date ushered in industrialism. The consciousness of a new economic era, arising in the mind of a theretofore sluggish and perverse South, is the best evidence of the beginning of manufactur- ing for the very good reason, as will presently appear, that expression of this consciousness went far to create the de- velopment. Preliminary notice of a Charleston newspaper's trade review covering months in 1880 and 1881 said: " In the An- nual Review will be exhibited the course and strength O'f the manufacturing revival in South Carolina, with especial reference, of course, to the progress of manufactures in Charleston." 122 And the summary itself declared: "The industrial feature of the year is the rapid extension of cot- ton manufacturing in South Carolina in common with other Southern States. . . . diversified industries are taking the place of the exclusive cultivation of cotton. . . ." 123 Another paper commented on the desire of a Northern contemporary that New England should take steps to pro- gress into the manufacture of finer grades of cotton goods, since it recognized " the great advance we are about to make at the South." 124 How certainly this was a change in South- ern experience is shown in the assurance with which altera- 121 U. S. Census of Manufactures, " Cotton Manufactures," pp. 38-39. 122 News and Courier, Aug. 16, 1881. 123 Ibid., Sept. 1, 1881. 124 The Observer, Raleigh, March 26, 1880. yO THE RISE OF COTTON MILLS IN THE SOUTH ^1/6 tion for the better was sensed. Thus, "The cities of the South are rapidly learning to appreciate the great value of manufacturing industries, and the great' development of the last year or two is only a beginning of what may be ex- pected when that whole section throbs with industrial life and activity in the near future." 125 By 1884 the new turn in events was so evident that, in brief retrospect, the date of genesis could be discerned. Of South Carolina it was said : " The State has now recovered the ground that was lost by emancipation, by negro suf- frage, by political misrule and official corruption. And the most significant circumstance is that the industrial triumph now proclaimed is mainly the result of the work of four or five years." And a significant point was touched in the ob- servation that " agricultural operations could be carried on with reasonable success, in even the darkest days of strife and misrule, but the undertakings which were dependent on the concentration of capital for their development re- mained torpid, if not dead, until the return of confidence breathed into them new life and vigor." 126 By 1880 one of the oldest Southern cotton manufac- , turing towns had recovered. In 1865 the Federal army burned 60,000 bales of cotton and all the mills of Colum- bia " The very heart of the city was burned out. ." . . Within fifteen years the waste places have been rebuilt and industry revived from its very ashes." 127 125 Baltimore Journal of Commerce and Mfgrs. Record, Aug. 26, 1882. "... too little heed is given by manufacturers and mechanics to the immediate prospects opened up by what is termed the new departure of the South; . . . there is no possibility that the South can immediately become a section of great manufacturing centres; but it is unquestionable that a combination of present efforts will in time yield important results" (American Machinist, quoted in ibid., July 15, 1882). Cf. ibid., July 15, and, in connection with buying by Southern merchants, Aug. 26, 1882. 126 News and Courier, Charleston, Feb. 4, 1884. Giving figures of cotton manufacture, it was said : " In a little more than three years . . . the increase in production was a third more than in the ten years ending in 1880, and the whole production in 1883 was ten times as great as the product in i860" (ibid.). As to the process of agri- cultural recuperation, cf. Hammond, p. 166. 127 Observer, Raleigh, Sept. 10, 1880. I77U THE BACKGROUND 71 Newspaper notice of organization of the Charleston Manufacturing Company in 1881 was headed, "The dawn of a new era," and the same paper, which did so much to bring about cotton manufacturing, often showed how sharply denned was the movement's beginning. 128 The 1880 census enabled .the South to take stock of its in- dustrial condition as a section and as part of the nation, and furnished a definite basis on which to calculate improve- ment. Speaking of the increase in manufacture in Augusta, a cotton manufacturer of that city summed up what had been done since the census of 1880, as follows: Well, to particularize, the Sibley Mill has been completed; the King and Goodrich Mills built up entirely since that time. The Summerville, McCoy, Globe and Sterling Mills have all been in- creased largely, and the Enterprise Factory more than doubled. These increments since the meagre census reports were sent in mean 63,000 new spindles, 2,200 additional looms and about 2,200 fresh hands . . . the increase in cotton manufacturing property alone since the census amounts in Augusta to $30o,ooo. 129 It will presently be shown that the Atlanta Exposition of 1 88 1 had much to do with stimulating interest in cotton manufacturing in the South, and in accelerating and broad- 128 News and Courier, Charleston, Aug. 1, 1881. Commenting on an address of H. P. Hammett, " Cotton Mills in the South," which was in itself a full exposition which indicated widespread popular inquiry into the subject, it was said that the speaker's own factory "was projected and built before the opening of the Cotton Mill Campaign in the South, and Major Hammett ranks, therefore, as one of the pioneers . . ." (ibid.). 129 Manufacturers' Record, Baltimore, Feb. 15, 1883. Nine months earlier a Georgia paper could read in the progress since the census the promise of a time when the South might " spin every pound of cotton made upon her fields" (Columbus Chronicle, quoted in Balti- more Journal of Commerce and Manufacturers' Record, Oct. 14, 1882). Cf. Atlanta correspondence of Augusta Chronicle and Con- stitutionalist, quoted in Manufacturers' Record, Baltimore, Feb. 8, 1883, and Augusta Trade Review, Oct., 1884. A special issue of the Baltimore Journal of Commerce and Manufacturers' Record, Sept. 2, 1882, denominated " an exponent of the new South," gave sta-r tistics of the important features of cotton manufacturing in the South, by States, at that date, indicating that from $15,000,000 to $18,000,000 had been invested in the business since 1880. Cf. Manu- facturers' Record, Baltimore, March 8, 1883; Baltimore Journal of Commerce and Manufacturers' Record, July 29, 1882. 72 THE RISE OF COTTON MILLS IN THE SOUTH [178 ening and lending confidence to the " cotton mill campaign." But it wa9 result as well as cause. The rapidity with which the exposition was planned and opened in a small town in the heart of a section unaccustomed to such ventures, and the readiness of response to its appeal cannot be explained except in recognizing that the Southern thought for indus- try had gone far toward crystallizing. A few years earlier it would have been impossible because the suggestion of such a scheme would have been unmeaning. 130 After Atlanta had had the faith to act host to the first exposition predicated upon belief in the South's industrial future, other places, by entering eagerly into plans for sim- ilar undertakings, testified to the awakening. It was even proposed to duplicate the Atlanta Exposition in Boston; this was perhaps a sophisticated suggestion intended to lessen the enthusiasm for the manufacturing of cotton in the South that had been the rather unexpected outcome of the original exhibit. 131 Baltimore in 1882 tried to launch an exposition that would allow the city to spring into lead- ership of a movement of proved success, and it was even said that the future of Baltimore would depend upon the way in which the proposal was met. 132 The next year Louis- ville and Nashville actively entered into rivalry for another 130 « T^ Atlanta Exposition, in 1881, was the hopeful and con- scious expression of the opening of a new era for Southern indus- try; . . . consequently, wonderful as has been the growth of this quarter century, it is but the realization of what was even then prac- tically assured by existing attainments and conditions" (Clark, in South in Building of Nation, vol. vi, p. 280). See editorial giving a summary of Atlanta's prosperity in The Daily Constitution, Jan. 2, 1880. "... it was all the work of merely ten months from the time the project was conceived until the exposition was thrown open to the people. It was impossible in that short time, at that remote dis- tance, and in that small city, to do the whole South complete justice. But a knowledge of the South's resources was demanded . . ." (J. W. Ryckman, secretary of the exposition, in Baltimore Journal of Commerce and Manufacturers' Record, June 24, 1882). "The visi- tors to this [exposition] were convinced that ' an industrial revolu- tion had actually been effected in the South . . .'" (Hammond, pp. 328-329). 131 See Philadelphia Industrial Review, quoted in Baltimore Jour- nal of Commerce and Manufacturers' Record, June 10, 1882. 132 See ibid., June 10, Sept. 23, Oct. 7, 21, 1882. I79] THE BACKGROUND 73 exhibition. 133 In 1883 the board of agriculture of North Carolina, aroused to the possibilities of the State, paid a visit to Boston, and the next year occurred the Raleigh ex- position. The New Orleans undertaking followed in 1885. The detailed description of the condition of the cotton manufacture in South Carolina, published by the News and Courier in 1880, was evidence of the same consciousness of industrial stirrings as was the Atlanta Exposition. 134 There was abundant recognition outside of the South of the industrial awakening that occurred about 1880 and was made manifest in the Atlanta Exposition. Agreement among Philadelphia cotton manufacturers to shorten pro- duction of coarser fabrics was held to be as wise as it was significant, for " the time can not be far distant when all our coarse cottons will be supplied from the cotton belt; and the child is born who will see the great mass of cotton manufacturing in all its diversified branches, carried on where the fleecy staple is cultivated." 135 It naturally took a little time for the reality of the South- ern awakening to break upon observers who had hardly ex- pected industrialism from that section. 136 133 Manufacturers' Record, Baltimore, Feb. 22, 1883. 134 " Attempts have been made at different times to show the ex- tent of the cotton manufactures in South Carolina, but until to-day- no thorough and complete statement upon that subject has been given to the public" (Blackman, p. 3). Cf. Kohn, Cotton Mills of S. C, p. 20. 135 Chicago Herald, quoted in Baltimore Journal of Commerce and Manufacturers' Record, July 29, 1882. A Boston journal struck a generous note that differed from some emanating from New Eng- land in an article, " The Drift of Manufacturing " : " Another Pitts- burgh is growing at Birmingham, Alabama; another Lowell at Au- gusta; another Lawrence at Columbus. . . . The East has no sole right to the term ' manufacturing ' ; the drift is Westward and South- ward, and is already a larger one than is generally supposed. . . . The time is not far distant when the breeders of domestic strife will be relegated to another clime, or at least to where they will cease attempting to array one set of industries in this great country against another set" (Commercial Bulletin, quoted in ibid., Sept. 23, 1882). 186 " Progress has been made with considerable acceleration as the wisdom of the new order of things became apparent, until now, when it appears that a new state of things has become established " (Miller and Millwright, quoted in Manufacturers' Record, Feb. 22, 1883). After speaking of the local character of ante-bellum mills, the Dry 74 THE RISE OF COTTON MILLS IN THE SOUTH [l80 Space only remains for bare mention of some objective evidences recommending 1880 as the date to be chosen as that marking the South's industrial awakening. The return to specie payments, bringing confidence to enterprise, showed itself in the veritable boom of the fall of 1879, pre- cipitating events in the South as all over the nation. 137 In 1880 Southern railway building took on new life, roads in financial difficulties being reorganized and narrow gauge being changed to broad gauge. 138 Southerners were accu- mulating a little surplus cash, as was indicated by their abil- ity to go again to Saratoga and other watering places. 139 Charleston shipbuilders were busy. 140 Plans for a cotton mill in Charlotte, though going the full length of organiza- tion of a company in the middle seventies, did not mature until 1881. 141 Something of the changed impulse back of cotton manufacturing about 1880 may be indicated in the fact that little was heard of extensions of woolen mills, though there had been many small ones in the South. The Clement Attachment, coordinating the work of ginning and spinning cotton, apparently did not cause pilgrimages and attract discussion until 1880. 142 Goods Economist, in 1896, said : " Whatever the expansion of the cotton industries of the South in the years following close upon the war, . . . such progress pales into insignificance when compared with what has taken place almost within the last decade " (Jubilee number, p. 78). Cf. Baltimore Journal of Commerce and Manufac- turers' Record, July 29, 1882; Sept. 23, 1882; News and Observer, Raleigh, Oct. 10, 1880; early suggestion of English interest is seen in a quotation from Iron, Philadelphia, in Manufacturers' Record, Feb. 8, 1883 ; cf. ibid., Dec. 21, 1882. 137 See Commercial and Financial Chronicle, Jan. 10, 1880; Cope- land, p. 266; Clark, in South in Building of Nation, vol. vi, pp. 264-265. 138 See Observer, Raleigh, Jan. 15, 1880, quotation from Railway Age; ibid., Jan. 8; Baltimore Sun, Jan. 22, 26, Feb. 2, 20, 1880. 139 News and Courier, Charleston, May 30, 1881. 140 News and Courier, April 13, 1881. 141 Tompkins, History of Mecklenburg, vol. i, pp. 181-182. Agita- tion for a special school tax, bringing several unsuccessful elections, during which time the school was suspended, resulted in an over- whelmingly favorable vote only in 1880 (ibid., p. 168). The streets of Charlotte began to be paved (Tompkins, Road Building and Broad Tires, p. 6). 142 See Blackman, pp. 18-19, and many other references in this pamphlet. l8l] THE BACKGROUND 75 The economic South was coming rapidly to a national point of view, strikingly signalized in the invitation of busi- ness men to Edward Atkinson to address them in the Senate chamber of Georgia in October of 1880. 143 Cotton goods in 1880 were in brisk demand, their price advancing more rapidly than that of the raw material; in this benefit Southern mills shared. 144 Production of cotton in the South had gradually increased by 1880-1881 to three times the number of bales of 1865- 1866, 145 and exports of the staple from the section to for- eign countries regained i860 figures by 1880. 146 The abun- dance of cotton in the section where factories would be likely to start, 147 coupled with the price (on the average about 11 cents), which had resulted through a general fall in the fifteen previous years, 148 was of consequence. Shortly after 1880 the manufacturing development of the South required special spokesmen and interpreters, and brought publications with such an aim, a9 the Manufac- turers' Record of Baltimore, the Industrial South, of Rich- mond, and Southern Industries, of Nashville, into exist- ence. 149 143 On this occasion, called by him (proceed. Southern Cotton Spinners' Assn., 1903) "The first opportunity ever given to a North- ern anti-slavery man to speak words of truth and soberness to Southern men," Mr. Atkinson said : " Malignant conditions [of dis- union] have passed away. The active and vigorous men born of the new South refuse to be controlled any longer by the Bourbons of that section, and the ' stalwarts ' of the North, who dare not trust the principle of liberty to work its first results, are being themselves classed as Bourbons incapable of guiding or directing the true union that now exists in this Nation" (Address at Atlanta, p. 8). See also ibid., p. 12, and John W. Ryckman in author's preface of ibid. 144 See Commercial and Financial Chronicle, Jan. 3, 1880; Balti- more Sun, Jan. 8, 20, 28, 1880; Blackman, p. 15. 145 Quotation from Bradstreet's, in Baltimore Journal of Com- merce and Manufacturers' Record, Nov. 4, 1882. 146 Brooks, p. 209. 147 Blackman, p. 7. 148 Quotation from Bradstreet's, in Baltimore Journal of Com- merce and Manufacturers' Record, Nov. 4, 1882. As to improve- ments in agricultural implements in the South by 1880, see Tomp- kins, History of Mecklenburg, vol. i, p. 181 ; a Georgia community wanted an agricultural implement factory; steam engines were sold for farm use (Manufacturers' Record, Baltimore, Nov. 30, i£ 76 THE RISE OF COTTON MILLS IN THE SOUTH [Jl82 Managers of the New England Manufacturers' and Me- chanics' Institute announced in March of 1883 that space in the exhibition to take place in the fall had been applied for by Southern exhibitors. 150 Suggesting something as to the date of commencement of cotton manufacture is the fact that in 1886 South Carolina repealed an act of 1872 exempting from state, county and municipal taxes for ten years capital invested in cotton, woolen and paper mills. 151 In the next chapter it will be seen what positive bearing the defeat of the Democratic candidate in the presidential election of 1880 had upon the Southern cotton manufactur- ing industry. In this place it is only necessary to note that after 1880 Southern political animus never gave itself again to such bitterness against the North, and thus one undoubted obstacle to economic advance was removed. 152 149 See Baltimore Journal of Commerce and Manufacturers' Rec- ord, Aug. 5, Nov. 18, 1882; Manufacturers' Record, Baltimore, Nov. 23, 1882, Jan. 25, 1883. 150 Manufacturers' Record, Baltimore, March 29, 1883. 151 Clark, in South in Building of Nation, p. 282. 152 See statement of executive committee of Columbia and Lexing- ton Water-Power Company, in News and Courier, Charleston, March 25, 1881. CHAPTER II The Rise of the Mills It has been seen how cotton for long years had been hurt- y, ful to the South; how it had joined with slavery and seces- sion to bring the disaster of the Civil War; how after Aft humiliating but sobering Reconstruction years the curtain was ready to lift on a new act in which the characters should be chastened in spirit, clarified in thought, and quick to discharge changed roles. The South by 1880 was ready to be no longer negative, but affirmative; not just the passive resultant of its past, but the conscious builder of its future. From a consequence, the South was to become a cause. 1 The determination with which the South entered the War was to hold over to receive new application. " The forti- tude o.f the march, the courage of the charge, the heroism of the retreat, the touching sacrifices of the ill-paid and ill- equipped soldier-life — these were to be emphasized and pro- longed, when the tattered flag no longer flew, the quick roll of the drum had ceased, and the comradeship of the camp and march was dissolved. From defeat and utter poverty were to be wrought victory and plenty." 2 The South suffered a change of heart. An altered pur- pose animated its leaders, and gradually but certainly seized upon its rank and file. President Baldwin, of the Louis- x " There are scores of turning-points" in the history of cotton in America " where, if wisdom had taken the skeins from the hands of prejudice and passion, a righteous and peaceful pattern might have been the result" (Scherer, p. 296). This was a juncture where judgment was to prevail. 2 Grady, The New South, p. 166. On the Confederate monument in the busy little city of Anderson, South Carolina, are the words : "And above all let him [the truthful historian] tell with what sub- lime endurance they met defeat, and how in poverty and want, broken in health, but not in spirit, they have recreated the greatness, and made it again the sweetest land on earth. In grateful acknowl- edgment of their prowess in war, and of their achievements in peace, this monument is erected." 77 ?8 THE RISE OF COTTON MILLS IN THE SOUTH ("184 ville and Nashville Railroad, born in Maryland and for many years resident in New York, and so competent to speak for both sections, declared with force : The commercial men of the cotton States fully appreciate the sit- uation. . . . They now see clearly how very little politics have done for them, and seriously turn toward the real " reconstruction " which active trade will inaugurate. . . . All the war issues are dead and buried — except to a few politicians who misrepresent their constit- uents and merely use the language of the past to give them, person- ally, . . . prominence. . . . True, we hear a great deal more about the few men who stand forth prominently as the advocates of these dead issues than we do of the thousands of young and energetic Southern men who are building cotton and woolen mills ; who are opening mines and starting iron, copper and zinc furnaces, or who are relaying the roads between the Atlantic and the Ohio and the Gulf. These men don't talk, they don't write books, they don't go to the Legislature or to Congress. They speak, trumpet toned, in results. . . . Years have brought time for thought, and compulsory thinking has produced marvellous results. . . . The people of the South have suffered — it is not pertinent whether we regard their sufferings as just or unjust — but they have put aside mourning and are ready for work. 3 A Georgian in welcoming South Carolinians to the At- lanta Exposition said of the display that " It comes at a most propitious moment, for the South, in sympathy with the quickening energies which excite the continent, is even now trembling in the initial throes of the mighty industrial revolution that surely awaits her. A great change is evi- dently about to come upon us. 'In the fabric of thought and of habit' which we have woven for a century we are no longer to dwell, and a new era of progressive enterprise opens before us." 4 This whole study goes to show a f unda- 3 Quoted from New York Herald, in News and Courier, Charles- ton, July 11, 1 881. "Mills for the weaving of the coarser cotton fabrics are now in successful operation in Tennessee, Georgia, Ken- tucky and several of the Atlantic Coast States, all of which have been built by native labor, mostly with local capital and are managed by Southern men. . . . The class formerly known as ' poor whites ' are . . . assimilating with their more fortunate neighbors. They are making good workers in mine and field, good operatives in fac- tories. . . ." 4 News and Courier, Charleston, Dec. 27, 1881. "... there are 2I 3i I S7 spindles to Georgia's credit. . . . These are the weapons peace gave us, and right trusty ones they are. . . . The story the spindles tell is one of joy to all, and show (sic) how rapidly we are climbing the hill of prosperity" (Columbus Enquirer, quoted in Daily Constitution, Atlanta, March 9, 1880). Professor Hart has I85] THE RISE OF THE MILLS 79 mental distinction between the English Industrial Revolu- tion and that in the South, namely, that the former was, certainly in its immediate causes, unanticipated, accidental, while the latter was deliberately planned. 5 This is plain in the quotation just given, and at a dinner of the Burns Char- itable Association in Charleston, along with toasts to the poet and the queen, this was offered : " The State of South Carolina — A new era of prosperity is about to dawn upon her: increasing commerce, manufactures, agriculture and population, are the echoes of its coming.'" 6 Reconstruction governments, under radicals, outsiders and blacks, had attempted a political display through waste- ful, ruinous expenditure; it will be seen how different was the program of economic advancement embraced in the " Real Reconstruction " of Southerners come into their own. 7 Observing that "These old commonwealths 1 were quoted an editorial in a Southern newspaper, presumably of the early eighties, declaring that " the great South ... is self-contained, and what is more, she is self-possessed, and she has set her face reso- lutely against the things which will hurt her " (p. 219). 5 Cf. B. L. Hutchins and A. Harrison, A History of Factory Legis- lation, pp. 19-20. 6 News and Courier, Charleston, Jan. 26, 1881. 7 Cf. Dunning, Reconstruction, pp. 205-206. How much earlier reorganization might have come in the South had not the carpet-bag regime been instituted, may be guessed from the frankness with which South Carolina, which so largely led the revival in the eighties, reentered the Union in 1865. The sincerity and dignity of sur- render is sufficiently apparent in the speech of Huger, the aged post- master of Charleston, in seconding the motion nullifying secession, in the constitutional convention following the war. Of South Caro- lina he said : " She is my mother ; I have all my life loved what she loved, and hated what she hated; everything she had I made my own, and every act of hers was my act; as I have had but one hope", to live with her, so now I have but one desire, to die on her soil and be laid in her bosom. If I am wrong in everything else, I know I am right in loving South Carolina, — know I am right in believing that, whatever glory the future may bring our reunited country, it can neither brighten nor tarnish the glory of South Carolina. She has passed through the agony and the bloody sweat; as we now return her to the Federal Union, let every man do his duty bravely before the world, trustfully before God, remembering each man for himself that he is a South-Carolinian. She has been devastated by the invader, reviled by the hireling, mocked by the weak-hearted, but she has accepted the invitation to return, — accepted it in good faith, with the assurance of a word better than a bond ; and now, no matter what she gives up, no matter what there is to endure and to 80 THE RISE OF COTTON MILLS IN THE SOUTH [l86 arrested in their development by slavery and by war and by the double burden of a sparse population and of an ignorant alien race," Walter Page recognized that " The process that has been going on in the upland South in particular is a process of conscious and natural State-building, construc- tive at every important step," and working itself out through the two instruments of industry and popular education. 8 The quickness with which creativeness displaced destruc- tion showed a purposeful people. " Eighteen years ago," it was written in 1882, "the upper bank of the Augusta canal was walled up with a chain of turretted tenements of brick . . . over which stood, in lofty suggestiveness, the smoke spire. . . . These buildings were frequented by silent men who worked in quiet and in gloom, and who sifted through their machinery the acids and minerals which go to form the explosives of war. From a hundred battle- forget, let us all do our duty as becomes her children, counting it our chiefest honor to stand by her in evil report as well as in good report, honor alike to live with her and to die with her" (Andrews, The South since the War, pp. 52-53). Orr, deploring quibbles and extenuations, declared : " We must put it in the constitution that slavery is dead, and that we will never attempt to revive it. . . . We seem to forget where we stand; we forget that we made the war and have been beaten ; we forget that our conquerors have the right to dictate terms to us. . . . Let us be wise men. Let us strengthen" Jackson's " hands by graceful and ready acquiescence in the results of the war. So shall we strengthen ourselves, and soon bring again to our loved State the blessings of peace and civil rule" (ibid., pp. 61-63). Cf. ibid., p. 94. 8 Rebuilding of Old Commonwealths, p. 139. Grady's plan — " the settlement of the race problem and the development of the material resources of the South " — was nothing different (see Oliver Dyer, Sketch of Grady, in The New South, pp. 76-77). "Mr. Grady's patriotism partook of the quality of his love; although romantic and general, it was also practical and local. ... It took hold of the . . . condition and interests of the country — of its diversified indus- tries, its agriculture, its manufactures, its commerce, its internal de- velopment, its external relations, its education and its religion " (ibid., p. 20). He said in 1889: " The industrial growth of the South in the past ten years has been without precedent or parallel. It has been a great revolution, effected in peace" (New South, p. 191). On Professor Hart's discussion of the comparative wealth of the South and other sections, it may be commented that given the fact of huge potentialities in the South and of an awakened eagerness to develop these, status counts for little ; given the loaf, and the leaven working in the loaf, and the most exacting of economists must be satisfied" (see Southern South). 187] THE RISE OF THE MILLS 8 1 fields of the civil strife, the blackened granulations of the Augusta Powder Mills flashed and thundered, and when the war was over the mills went down before the ravages of time. . . . To-day, the same spire, with extinguished craters, overlooks the same spot. The same river rolls at its feet ; the same hills confront it on the other side. But in place of the scattered walls of war, a massive structure, granite and compact, is reared. In the place of musty ex- plosives of darker days, the purest productions of peace are fed into the present mill, and from its looms will go forth the texture to clothe the people of the land, to weave the white wings of commerce and to float the bunting of the Newer South. The old picture has rolled away — the new one has received a solid setting/' 9 One cannot view the passion with which revival was un- dertaken without realizing how pointed were the lessons taught the South in the war and its aftermath. 10 Convinced of old errors, the remaking of the South was emphatically in response to a moral stimulus, mot less real because not always outwardly apparent. "A man who has been in the whirl of New York or in any of the brand new cities of the great West coming into Charleston might easily enough come to the conclusion that the old city was in a sad state of decadence — but our own people who have been accus- tomed to its quiet way of doing business, if they have their 9 Chronicle and Constitutionalist, Augusta, Feb. 23. Though four years earlier North Carolina " would not be caught " in the " Yankee money trap " of the Philadelphia Centennial Exhibition, in 1880 it was being asked : " Shall our Commissioner of Agriculture or our State Geologist be . . . subjected to the mortifying . . . task of standing in those grand halls [of the proposed world's fair of 1883] . . . and present the ridiculous farce of representing this . . . State by showing a dump-cart load of rocks?" (News and Observer, Raleigh, Nov. 12, 1880) . Cf . ibid., Dec. 2, 1880. The ten " supreme advantages" claimed by Augusta in 1884 were every one economic, the first being its superiority as "a manufacturing center" (Trade Review of Chronicle and Constitutionalist, Oct., 1884). 10 Citing statistics of property losses to South Carolina between i860 and 1870 and the relative gain to a state such as Rhode Island, Murphy wrote : " Beneath these cold and unresponsive figures there lie what tragedies of suffering, what deep-hidden recurrent pulses of despair, of self-repression, of patience, of silent and solemn will, of self-contest, of ultimate emancipation!" (Present South, p. 101). 6 82 THE RISE OF COTTON MILLS IN THE SOUTH [l88 eyes open (or hearts open would perhaps be the better ex- pression) could not fail to see manifest improvement — progress even, if you like the word better." 11 As the movement proceeded from introspection, the very geniu9 of " Real Reconstruction " was self-help. It took courage to begin, but confidence rallied about every sign of genuine performance. Thus it was said that " Every true South Carolinian must rejoice at the . . . energy exhibited by the citizens of Columbia in their management of the Cotton-Mill Campaign. For years they have appeared to depend on somebody else to help them. The Legislature made liberal concessions. No effort was spared to interest Northern capitalists in the splendid water power. . . . But nothing was done. Tired oi waiting a number of business men in Columbia took up the matter themselves. They soon found that the citizens generally would sustain them. . . , the city is full of life again. A handsome sum of money has been subscribed .already to the capital stock of the Cot- ton Mill Company. ... It will be a happy day for the whole State when the hum of a myriad spindles is heard on the banks of the historic Canal." 12 11 News and Courier, Charleston, March 24, 1881. Timrod wrote of Charleston : " How know they, these busy gossips, what to thee The ocean and its wanderers may have brought? How know they, in their busy vacancy, With what far aim thy spirit may be fraught? Or that thou dost not bend thee silently Before some great unutterable thought?" (Henry Timrod, Poems, Memorial Ed., 1899, p. 172). Professor Sioussat has stressed the significance of the economic readjustment between 1865 and 1880, "a readjustment more fundamentally im- portant than the political events which in large degree overshadowed the less dramatic factors" (History Teacher's Magazine, Sept., 1916, p. 224). Cf. Ingle, p. 5. Declaring right after the war that negro slavery had been hardly more debasing than white slavery, Andrews foresaw that the remaking of the South must reach down to basic tasks : " That is the best plan which proposes to do most for the common people" (pp. 387-388). Cf. Clark, in South in Building of Nation, vol. vi, p. 254. 12 " The News and Courier busies itself with every enterprise, big and little, that will turn a dollar's worth of raw material into more than a dollar's worth of manufactures. ... we confess to a weak- 189J THE RISE OF THE MILLS 83 This self-reliance never meant exclusion of assistance from the North or elsewhere ; it meant a broadening, not a contracting of view. " Some of that credit which was ac- corded to the man who caused an additional blade of grass to grow should be given to everyone," whether home or out- side enterpriser, " who affords facilities to manufacture an additional! boll of cotton. . . ." 13 The South, ready to plunge into its task, took stock of itself. " All questions of domestic economy, and especially those involving the capi- tal of our people, whether in the shape of labor or dollars, will necessarily be canvassed and scrutinized very closely in their bearings on our material progress. . . ," 14 Even those communities most earnest in social regenera- tion, and most anxious to forget the past in looking to a saner future, very occasionally slipped back into old ruts, and found in material advancement the means of satisfying spitefulness. Thus an attempt to settle foreigners upon a large tract in eastern Tennessee was commended partly be- cause it would increase congressional representation of the ness for Columbia, which suffered so sorely at the end of the war. . . . But cotton mills will soon make amends for the vicissitudes and hopelessness of the past . . ." (News and Courier, Charleston, March 19,1881). Another paper discouraged reliance upon the government for prosperity, and pointed to relief that had come to the West only through self-help: "That government is the best which is not re- quired ... to pass new laws, leaving to the people the utmost free- dom, with full liberty to devote their energies to the improvement of their own condition. . . . We know of no people more favorably situated than North Carolinians are in this respect" (Observer, Raleigh, Jan. 9, 1880). As Ireland in its cooperative agricultural efforts later, the South was experiencing a " combination of eco- nomic and human reform" (see Plunkett, pp. 205-206). "There came a different viewpoint," said one informant. " The old South was done away with. The problem was to utilize the thing nearest at hand to support a large portion of our people." And so the North Carolina Board of Agriculture made an investigating trip to New England, and an industrial exhibit was held " (Henry E. Fries, int., Winston-Salem, N. C, Aug. 31, 1916). 13 News and Courier, Charleston, June 28, 1881. 14 News and Observer, Raleigh, Dec. 1, 1880. " South Carolina in 1884," a 60-page pamphlet published by the News and Courier after a comprehensive survey of economic conditions prevailing in each county, shows the strength of this spirit. In descriptive detail it is a valuable photograph of agriculture and industry in the State at that date. 84 THE RISE OF COTTON MILLS IN THE SOUTH [19O South and enable it the better to protect itself against " ad u verse legislation." 15 Once awake, how immediately the South went to work is evidenced in notices proclaiming the new order of things. " The time was when the South was exclusively agricultural in its pursuits, but the past few years have seen factories springing up all over this section. . . . The South is destined at no distant day to not only raise cotton . . . but to manu- facture it . . . thus keeping at home all the profits." 16 It was recognized that Southern economic life was becoming more diversified, in agriculture and in industry, and so com- munities were growing independent. 17 The franker and more generous Northern papers joined writers at the South in encouraging the new development. It was generally held at the time that internal impulse wa9 chiefly responsible for the change in program. It could not be said of the South as of the establishment of the factory regime in England 15 Observer, Raleigh, Aug. 25, 1880. Virginia, never so ardently back of economic recuperation as States to the south, was perhaps hindered by internal dissension over repudiation of part of her debt; the papers at this juncture were filled with political wrangles (cf. Daily Dispatch, Richmond, Feb. 9, March 24, 1880). The proposal to exempt manufacturing plants from taxation, already bringing results further South, could raise protest from the farming interest (cf. ibid., Jan. 14, 27, 1881). Public solicitude over industrial devel- opment was far less marked than in the Carolinas and Georgia, partly because of border position of Virginia, partly, perhaps, be- cause there was not the one chief manufacture, cotton, on which to center attention. There was less reliance on home effort, more look- ing to outside assistance (cf. ibid., March 29, 1880). Mississippi had time for childish vituperation over dead issues. A Wisconsin editor had asked a Mississippi contemporary, " Did you ever read of Appomattox?" He received the reply: " O, Yes! We've read of Appomattox, where a few hungry and ragged thousands surrendered to a man with a million of men under his command. .. . . the whole .wide world remembers that it required five of your federals to whip one of our confederates. . . . Will you fight for Grant if he should slap a golden crown on his cranium? . . . The last man of you that shoulders a shot-gun in behalf of your gory god will be hunted down like dogs . . ." (quoted in Daily Constitution, Atlanta, Feb. I, 1880). Cf. a headline in Daily Constitution, Atlanta, April 11, 1880, and colloquies in ibid., March 14, 1880; News and Observer, Raleigh, Dec. 18, 1880; News and Courier, Charleston, June, 1881. 16 Americus, Ga., Recorder, quoted in Baltimore Journal of Com- merce and Manufacturers' Record, Oct. 14, 1882. 17 Cf. Miller and Millwright, quoted in Manufacturers' Record, Baltimore, Feb. 22, 1883. I91] THE RISE OF THE MILLS 85 that "As a great fact the system originated in no precon- ceived plan ; on the contrary, it was formed and shaped by the inevitable force of circumstance. . . . The first force which tended to create this system was that of invention. . . ," 18 After deploring " the errors of previous generations in their persistent blindness to home possibilities, while spending their money North and abroad," it was declared: " The war cost us heavily — oh ! so heavily — but we bent our stout hearts patiently to our tasks, and have profited, and will profit, by its lessons." 19 Contemporary spokesmen were naturally in some instances cautious to explain that " The New South " did not imply repudiation of the best spirit of the old South. 20 An understanding interpreter has ob- served that Southerners, when slavery and the war were past, "began . . . to beat their swords into plow shares and their spears into pruning hooks and to enter upon the childhood of material growth . . . , to give up the old time Southern ways and ideas of life, and to blend the character- istics of that day with the new spirit of business enterprise and thrift, changing from ' hornets in war to bees in in- dustry.' . . ." 21 Before 1880 the South had worn a veil before her eyes, had been running a temperature that distorted economic perspective, corrupted public judgment. When the veil was torn off and the fever subsided, normal thinking brought frank avowal of the past distemper. The section had woven " rosy day-dreams of a far-off greatness," and 1 been tortured 18 Carroll D. Wright, " The Factory System of the U. S.," in U. S. Census of Manufactures, 1880, p. 1. 19 Augusta correspondence of Savannah Morning News, July 4, 1882. Cf. News and Observer, Raleigh, Nov. 16, 1880, praising the industrial progress of Augusta. 20 Cf. News and Courier, Charleston, Dec. 27, 1881. Mr. Edmonds' solicitude on this point has been noticed ; cf . Edmonds, p. 1. 21 W. C. Heath, in Southern Cotton Spinners' Assn., proceed., 1903, p. 49. Post-bellum activity in mill building recalled the fact that years before planters had conceived the advantage in manufacturing, but were deterred by slavery; originality, to be effective, needed to work under a new dispensation. Cf. Gannon, Landowners of South and Industrial Classes of North, p. 6 ff . ; Andrews, pp. 224-226. 86 THE RISE OF COTTON MILLS IN THE SOUTH [192 by a " delusive mirage," 22 but now " The South must . . . look out for herself, and bring her great advantages to bear in her favor, asking only a free field and a fair fight against all competitors. ... It means work and not words." 23 To appreciate the strength of the demand for social re- generation, it must be recognized that while cotton 'manu- facturing formed its central purpose, the movement was comprehensive, embracing, in thought if not in deed, many departments of life. Progress along all lines was not simul- taneous or equal. It is not hard to see why public education, for example, did not so soon translate desire into realiza- tion as did industry. Bread and meat must first be looked to, and the South then could turn to plans which, if more truly fundamental, were still less instantly pressing. 24 If the will was surely present, and it was felt that " The South- ern States ought, in justice to posterity, to take this matter of public schools in hand," 25 it needed twenty years until performance could follow. When the South, after 1900, did embark on an educational campaign, the fervor pre- viously given to industry received new expression. 26 It was " Real Reconstruction "reaching another task. In the English Industrial Revolution other trades bor- rowed stimulus from textiles; 27 in the South, where the causal force was subjective rather than objective, this would more certainly be the case. Improvement in farming was 22 Industrial South, quoted in Baltimore Journal of Commerce and Manufacturers' Record, June 17, 1882. 23 Gannon, Landowners of South and Industrial Classes of North, pp. 6-7. 2i " I do not . . . suggest that any other agency of . . . economic progress can be more than a very partial substitute for education; but only that, in the peculiar circumstances of Ireland, we must have recourse to supplementary influences which will produce a more im- mediate effect upon the general life of the present generation while its young people are being educationally developed" (Plunkett, pp. 51-52). 25 News and Observer, Raleigh, Nov. 20, 1880. 26 Sioussat, p. 270. " Enthusiasm like that of a great religious movement developed and the result was that in the decade 1900- 1909 the total school revenues in these States had been more than doubled." 27 Cf. Scherer, pp. 51-52. 193] THE RISE 0F THE MILLS %7 especially significant. In the zeal for manufacturing, the temptation would be to neglect agriculture, the old bete noire, and so not keep ever in mind the higher wisdom of an economic balance. 28 But exodus of many negroes from a South Carolina county was thought by some a blessing in disguise, in that it would stimulate diversification and rota- tion of crops, rest land which needed rest; crops requiring less attention than cotton, grain for example, would be raised. 29 North Carolina farmers were encouraged to at- tend an agricultural meeting in far-away Connecticut. 30 " We at the South," it was said, "... if we intend to turn over a new leaf and seek a new development for our section," must inaugurate shipping relations with Brazil, 31 form something like a Southern chamber of commerce, 32 form a mercantile connection with Cincinnati, 33 send cotton abroad through Southern ports, 34 promote harmony within the section. 35 " The railroad fever is epidemic in Georgia," it was as'serted. " Every village wants a railroad to its neighbor." 36 The next year it could be said " There are now over 20,000 men and 100,000 horses and mules em- ployed in railroad building in Texas," 37 and a North Caro- lina editor even foresaw danger of railroad domination in state politics. 38 28 A friendly adviser pointed out the danger of excessive manu- facturing in England, and urged that the South seek development of agriculture beside industry (United States Economist, quoted in Baltimore Journal of Commerce and Manufacturers' Record, Sept. 30, 1882. 29 News and Courier, Charleston, Jan. 2, 1882. 30 News and Observer, Raleigh, Nov. 30, 1880. " Our system of agriculture is too much on the order of present enjoyment and does not have sufficient regard for future use. . . . We would gladly see all of the profits of this year's crop spent on the land itself. . . ." (ibid., Sept. 19, 1880). 31 News and Observer, Raleigh, Nov. 17, 28, 1880. 32 Ibid., Dec. 5, 1880. 33 Observer, Raleigh, April 1, 1880. 84 News and Observer, Raleigh, Nov. 13, 1880. 35 Observer, Raleigh, July 11, 1880. 36 Observer, Raleigh, Feb. 6, 1880. Cf. ibid., Jan. 15, Feb. 20, 1880. 37 News and Courier, Charleston, May 30, 1881 ; cf. ibid., April 29, 1881. 38 Observer, May 1, 1880. 88 THE RISE OF COTTON MILLS IN THE SOUTH [l94 Interest was taken in extension of telegraph and tele- phone lines. 39 Temperance societies showed augmented support. 40 Duelling was coming to be called murder in South Carolina. 41 The section exulted in the erection of cotton seed mills and exploitation of iron ores and phos- phates, cultivation of oranges and rice, and extension of cattle and sheep raising. 42 Cries for colonization of the negro, earlier condemned, 43 had hushed. More than a contributing cause in the growing desire for economic renovation of the South, and amounting certainly to a decisive accelerant, was the defeat of Hancock by Gar- field in the presidential election of 1880. The South, emerg- ing from the humiliation of Reconstruction, had centered hopes on a victory for Tilden over Hayes four years earlier, and when the Democratic candidate was counted out, by a likely fraud as the section was willing enough to believe it, despair gave way to resentment and the Solid South, nurs- ing its pride and revengefulness during Hayes' administra- tion, dedicated itself to Hancock's triumph. In the four years between elections, the South, bearing many real griev- ances, sought to lighten them by lashing itself to a false ambition. Hancock's success would give answer to the North and cure Southern sorrows. It was looked forward to as " the first full, and fair, and free presidential election in which the South has participated since the war. There will be no intimidation of voters by means of the army. . . . There will be no southern returning boards upon whose venality the republican leaders can rely in case of a close contest." 44 The shock of Hancock's defeat threw the South, so to speak, back upon its haunches. The days immediately fol- lowing are surcharged with interest for the student of Southern economic history. 39 See News and Courier, Charleston, Jan. 1, May 4, 1881. 40 Ibid., April 22, May 5, 28, June 13, 1881. «■ Ibid., March io, 1881. 42 Observer, Raleigh, Sept. 4, 1880. Cf. Daily Constitution, At- lanta, March 30, 1880. 43 Andrews, p. 158. 44 Daily Constitution, Atlanta, Feb. 15, 1880. 195] THE RISE OF THE MILLS 89 The News and Observer, of Raleigh, which had been vio- lently sectional and which for a few days after the election consoled its readers with hope of victory four years hence, within a week changed front and gave expression to a new spirit that, suddenly and with compelling force, was sweep- ing the people. 45 It was declared that " we have been de- feated in the national contest. In the administration of the national government for the next four years we need not concern ourselves, for as far as possible our councils will be ignored. What, then, is our duty ? It is to go to work earnestly to build up North Carolina. Nothing is to be gained by regrets and repinings. No people or State is better able to meet emergencies. . . . And what nobler employment could enlist the energies of a people than the developing of the great resources of our . . . State. . . . But with all its . . . splendid capabilities it is idle to talk of home independence so long as we go to the North for everything from a toothpick to a President. We may plead in vain for a higher type of manhood and womanhood among the masses, so long as we allow the children' to grow up in ignorance. We may look in vain for the dawn of an era of enterprise, progress and devolpment, so long as thou- sands and millions of money are deposited in our banks on four per cent interest, when its judicious investment in manufacture would more than quadruple that rate, and give profitable employment to thousands of our now idle women and children. " Out of our political defeat we must work ... a glor- ious material and industrial triumph. We must have less politics and more work, fewer stump speakers and more stump pullers, less tinsel and show and boast, and more hard, earnest work. . . . Work for the material and edu- cational advancement of North Carolina, and in this and 45 In quotations from influential newspapers it will be observed that the changed view, breaking on the South so quickly, at first carried something of sectional prejudice; industrial upbuilding would be partly spitefulness against the North. But this was the whimpering of a child while drying its tears. 90 THE RISE OF COTTON MILLS IN THE SOUTH [196 not in politics, will be found her refuge and her strength.''* 8 Following the installation of Garfield, another editor finely said: But if we lost the victory, in one sense, we have won it in another. We have been taught what the South can do for itself if it wills to do it. If we have lost the victory on the field of fight we can win it back in the workshop, in the factory, in an improved agriculture and horticulture, in our mines and in our schoolhouses. There is where our fight lies now, and the only enemies before us are the prejudices of the past, the instincts of isolation, the brutal indifference and harmful social infidelity which stands up in our day with the old slave arguments at its heart and on its lips, "I object" and "You can't do it." 4 ' No people less homogeneous, less one family, knit to- gether and resolute through sufferings, could have taken instant fire, as did the South, at such appeals. Facilities for satisfying the need were not narrowly investigated. The South was shut up to such and such means — they must fit into imperative requirement. 48 46 Nov. 9, 1880. " We must make money — it is a power in this practical business age. Teach the boys and girls to work and teach them to be proud of it. . . . Demand all legislative encouragement for manufacturing that may be consistent with free political economy." 47 Columbia Register, quoted in News and Courier, Charleston, March 18, 1881. Columbia at this time was entering upon the fervor to develop its canal and build a cotton mill. The editor of the Reg- ister had been a slaveholder. This pronouncement is purged of an earlier and unworthy jealousy which had sometimes appeared in such expressions as the following from another paper : " The South should depend upon its own virtue, its own brain, its own energy, attend, to its own business, make money, build up its waste places, and thus force from the North that recognition of our worth and dignity of character to which that people will always be blind unless they can see it through the medium of material, industrial and intel- lectual strength. We may proclaim political theories, but it is the more potent . . . argument of the mighty dollar that secures an audience there, and the sooner we realize it the better for us " (News and Observer, Raleigh, Nov. 27, 1880). 48 Also, as has been seen, a philosophy which had right quietly, sometimes half-consciously, been taking shape in the Southern mind, was just now becoming fully articulate. For example, some months before the election, it had been said : " While the politicians are making a great deal of noise over the states rights question, the people of the South are quietly making substantial industrial prog- ress. . . . The cotton mills in operation have proved very profitable. New mills are projected. . . . The signs of the great industrial change now going on in the South are plainly visible everywhere. 197] THE RISE OF THE MILLS 9 1 But the South did more than receive a new economic aim. Garfield elected, it began to show further the faith that had been welcomed, and moved to renounce political separatism : " The Southern people must be National themselves, in their aspirations and conduct, if they would have the Govern^- ment truly national in spirit," and Garfield president not of a section or party. " To have a government of ' the whole country,' to be entitled to it, we must think of the whole country as our own, and demand no more than we are ready to give. It must come to this." 49 Garfield'9 assassination showed how ready the South was to join hands with the North. " It could not have been fore- seen . . . that the outburst of sympathy and condemnation would have been universal in its manifestation, affectionate in tone and National in spirit. South Carolina does more . . . The people of the South are beginning to learn that the true road to power is not through the white house, supported by a swarm of federal officials. They are learning that solid wealth is power, and that wealth is attainable only by working up their cotton and wool into fabrics and their ores into metals" (Memphis Avalanche, quoted in Daily Constitution, Atlanta, March 30, 1880) . " Distinc- tion must be made between the political talk in the papers and what the people really wanted. There was a strong but silent undercur- rent for economic welfare, while the politician was still singing the old song" (E. C. Brooks, int., Durham, N. C, Sept. 18, 1916). Ap- proval of the thought that Hancock's defeat threw the South into a reversed frame of mind was received in interviews with some men who lived through the events, but from none of the newer generation. 49 News and Courier, Charleston, March 9, 1881. "In the near future the successful leaders, South and North, will be those whose first thought is for the Republic; men who are National in feeling and purpose; men who understand that the political and social strength and safety of each State depend not on isolation and sepa- ration, but on combination and union." Cf. ibid., May 7, 1881. A New Orleans editor said: "The bitterness, prejudice and hostility to the changes wrought by the war which were so marked a few years ago are disappearing. There is now a very noticeable . . . disposition to accept the situation as it is, and on this basis to build a new South which shall surpass in wealth, glory and greatness the old South. . . . before another National campaign opens this ele- ment will control the political and material affairs of the South " (Times-Democrat, quoted in ibid., Feb. 4, 1881). Cf. A. K. Mc- Clure, The South : Industrial, Financial, Political, p. 53 ; Dunning, p. 198. Not all leaders were so sensible; Senator Vance, of North Carolina, looking forward to Garfield's term, was belying his pro- fessions by asserting " The thing that has been is the thing that shall be" (News and Courier, Charleston, Feb. 23, 1881). 92 THE RISE OF COTTON MILLS IN THE SOUTH [198 than reprobate assassination. The . . . whole people, re- sent the deed because the victim is the President of the United States. . . . The forces of reunion had gone on with a rapidity which few appreciated. All the elements of cordial friendship and of national good-will were there. It needed only the threat of a common misfortune to give shape and voice to the recreate [sic] but sturdy love of the Republic." 50 It is clear that the pressing task of the South, from the day of Appomattox, was truly an economic and social and not a political one. 51 By 1880 this was publicly apparent, and no later expression of this view 52 has been plainer than contemporary exhortations that the people shut ears to politicians and open sympathies to constructive action. " So long as we have sectional enmity in politics in the 60 News and Courier, Charleston, July 13, 1881. "... the Presi- dent's desperate illness . . . has done more than years of ordinary events in bringing the North and South together. . . . Vainly will the politicians flourish the ' bloody flag.' The people will not rally on the ensanguined colors again" (ibid., July 18, 1881). Cf. ibid., July 14, giving interview with Jefferson Davis, and Sept. 20, 1881 ; William A. Harden, A History of Savannah and South Georgia, p. 485. The cordiality with which the First Connecticut Regiment was received in Charleston the month following Garfield's death was believed an outgrowth of the city's sorrow at the national tragedy. The first column of the News and Courier bore the flags of Con- necticut and South Carolina crossed, with the legends, " Yankee Doodle Come to Town," and " A Welcome Invasion." An editorial spoke of the war as a "grand lesson to the South," and declared: " We have learned that we cannot stand alone, that our fight must be made within the Union . . ." (Oct. 24, 1881). 51 Cf. Sioussat, p. 223. 52 " The greatest statesman of the South in recent times was Sea- man A. Knapp, who believed that the demonstration farm was of more value to society than the noisiest political convention . . . ; that a boy's corn club would do more to enrich materially the life of the people than the fattest office won on the hustings. . . . The unselfish servants of the people, working in humble ways to improve the farm, the road, the factory, the home, the school, and the church are the true statesmen of the South" (Samuel C. Mitchell, "The Challenge of the South for a Better Nation," in The South Mobilizing for Social Service, p. 46). Cf. ibid., p. 45. " Back of the patriotismof arms, back of the patriotism of our political and civic life, there lies, like a new and commanding social motive, the patriotism of efficiency. ... It is not merely the patriotism of industrial power. It is the patriotism of social fitness and of economic value. It is the passion of usefulness" (Murphy, Present South, p. 148). Cf. ibid., p. 316. 199] THE RI SE OF THE MILLS 93 South its material prosperity will be checked and an abso- lute injury will be sustained ... by exciting distrust of capital and prejudices of immigration. The Southern peo- ple, outside of the professional politicians, care very little about Federal politics. They are endeavoring to develop the resources of the South and regain the broken-down for- tunes left by the desolation of civil war." 53 Asserting that the South should welcome outside enterprisers, bid for gov- ernment appropriations and hold to the party that could insure peace in which to follow economic pursuits, a South Carolinian wrote that " The object of our politics should be the development of our resources. ... In this State we need capital and less party and politics." 54 It was not until 53 Sumpter, S. C, Southron, quoted in News and Courier, Charles- ton, May 14, 1881. " So taking the past and the present as indices for the future, it is plain to see that a dissolution of the Solid South will cut at the very roots of all these wrangles between the North and the South in which sectionalism is involved." Cf. Observer, Raleigh, Jan. 29, 1880, in comment on an editorial of Financial Chronicle. 54 "Brutus" in News and Courier, Charleston, May 25, 1881. For a list of Federal appropriations for North Carolina in the rivers and harbors bill the year previous, see Observer, Raleigh, May 6, 1880. It was said, apropos of the approaching meeting of the Southern Press Association, that the Associated Press in its selection of news did not always contribute sufficiently to the business progress of the South. " The commercial prosperity of the South is of far greater consequence to the Southern press than any mere political object. . . . Any association, therefore, that will aid in . . . dissemination of truthful information about the social, business and industrial life of the Southern States, should be encouraged by those who control the Southern press " (News and Courier, Charleston, March 29, 1881). "It is time to stop impeaching the South's development, for . . . business is driving sentimental politics to the woods" (Spring- field Republican, quoted in News and Observer, Raleigh, Dec. 31, 1880). Years earlier Gregg took McDuffie severely to task for his half-hearted entrance into cotton manufacturing : " Had you . . . mixed a little more patriotism with your efforts, you would have taken the pains to ascertain why your Vaucluse establishment did not realize . . . sanguine expectations. . . . You would have put your own shoulders to the wheel. . . ." Instead of political oppo- sition to protection (Gregg did not favor a tariff), McDuffie should have advocated turning it to economic advantage (Domestic Indus- try, p. 8). "It would indeed be well for us, if we were not so re- fined in politics — if the talent, which has been, for years past, and is now engaged in embittering our indolent people against their indus- trious neighbors of the North, had been . . . engaged in . . . the encouragement of the mechanical arts" (ibid., pp. 7-8). Cf. Ingle, 94 THE RISE OF COTTON MILLS IN THE SOUTH [200 1880 that men like Gregg, pleading that "politicians, in- stead of teaching us to hate our Northern brethren, en- deavor to get up a good feeling for domestic industry," 55 and who were overborne by such followers of Dew as Cal- houn, Simms, Hammond, Rhett, Davis, Yancey, and Cheves, 58 could be justified in the public judgment as ex- pressed by Grady when he said : " Every man within the sound of my voice, under the deeper consecration he offers to the Union, will consecrate himself to the South. Have no ambition but to be first at her feet and last at her serv- ice. . . ," 57 Nearly every stage of this study testifies to the large ex- tent to which such economic publicists as these, by conscious teaching and by example, were responsible for industrial growth. It is a nice matter to strike a balance between the force of their inner promptings and the external influences operating upon them. It has been seen that their identical philosophy, held by earlier Southerners, could not bear fruit before 1880, and certainly from this date forward moral stimulus gathered strength from the constantly more apparent physical advantages for manufacturing. As to whether desire for industry uncovered facilities, or facili- ties for industry suggested their employment, a careful thinker said : " My answer is for the ideas, the internal stim- ulus, but subject to the qualification that in a longer time we would have had the mills by force of external influence. So far as the period from 1880 to 1900 was concerned, it Southern Sidelights, p. 40; Andrews, South since the War, p. 96). The " old dislike of the peddling, money-making Yankee is being replaced by admiration for his thrift, and desire to adopt the means by which he has left his impress upon the nation's life" (Springfield Republican, quoted in News and Courier, Charleston, April 7, 1881). One who now feels this view has been overemphasized said, speak- ing of the lack of strong men in politics in South Carolina, " they have gone into this great economic movement, and let most other things go to the dogs" (Mrs. M. P. Gridley, int., Greenville, Sept. 9, 1916). 55 See Domestic Industry, pp. 14-16; 11, 24; Olmsted, Seaboard Slave States, p. 363. 56 Dodd, in South in Building of Nation, vol. v, p. 568 ff. ; Kohn, Cotton Mills of S. C, p. 13. 57 Dyer, in New South, p. 90. 20l] THE RISE OF THE MILLS 95 was as nearly the immediate result of internal agitation as any industrial growth could be." It is probably correct to conclude that " the social and economic influences cooper- ated with the human purpose." 58 Enough has been said to make it apparent that at the outset the employment of children in the mills, if not abso- lutely necessary, was practically so, and never excited the least question. Search has failed to reveal one instance of protest against their working, but, on the other hand, cotton manufacturing was hailed as a boon especially because it gave means of livelihood to women and children. Poverty- stricken, the South was mustering every resource to stagger to its feet. All labor power was empirically seized upon; response was eager. At that critical juncture, later results of the employment of children could not be looked to. The great morality then was to go to work. The use of children was not avarice then, but philanthropy ; not exploitation, but generosity and cooperation and social-mindedness. 59 58 "vy. W. Ball, int., Columbia, Jan. 3, 1917. " The building of the Pelzer Mill was the germination of the idea implanted by The News and Courier." A competent student wrote : " The growth of cotton manufacturing ... is significant of a change in Southern ideals . . . a change from a social system in which work was held to be degrad- ing, to one in which great interest is taken in industrial enterprise" (Copeland, pp. 32-33). Sir Horace Plunkett said unhesitatingly of a development not far different : " The story of the new movement . . . begins in the year 1889, when a few Irishmen ... set them- selves the task of bringing home to the rural population . . . the fact that their prosperity was in their own hands ... to arouse and apply the latent capacities of the . . . people . . ." (cf. pp. 178-179). An objective judgment is: "Other industrial conditions beside the nearness to the cotton crop produced this growth, chief of which has been the general industrial awakening experienced by the South " (New International Encyclopaedia, article on "Cotton," p. 159). Mr. Brooks leans toward environment when he says : " In . . . nat- ural resources the South has found the basis of . . . new economic policy, a new social order . . ." (p. 214). 59 Between 1880 and 1890 the number of children was doubled, and between 1890 and 1900 trebled (cf. U. S. Census of Manufac- tures, 1890, " Cotton Manufacture," by Stanwood, p. 173 ; ibid., 1900, p. 33). "Manufacturers took whom they could get for operatives in the new mills. The employment of children was not a matter of choice but of necessity. . . ." Cf. Edmonds, p. 20. It must be re- membered, too, that whole families were transferred from farms to mill villages, which alone, in the then condition of the South, would have required that the children work. Of course, the use of chil- g6 the rise of cotton mills in the south [202 Understanding that the South, from inner impulse, en- vironmental suggestion and the union of these two, was determined for manufacturing, the immediate reasons for the building of mills may now be considered. It must be remembered that there is a distinction between industrial advantages believed to be present, and facilities as they were afterwards proved out. In the next pages the effort is to discover the thought back of the erection of factories, rather than the evaluation of supposed advantages as re- vealed in actual operation. It is clear, first, that there could be no single proximate cause. A mill president said: "You cannot find any uni- formity in the reasons for establishment of mills. There were a thousand reasons. Sometimes it was salaries that were wanted ; sometimes commission houses that were after the charges ; sometimes it was to build up the community ; sometimes the profits of one mill that brought another into being; sometimes the machinery men; sometimes it was just because they were . . . fools." 60 When Mr. Edmonds declared that "What the South has done . . . has been without any special stimulus," he meant there were few demonstrated aids to manufacturing in the dren has long since become unnecessary, and has been as cruelly unjust as at first it was natural. Cf. the writer's " Some Factors in the Future of Cotton Manufacture in the South," in Manufacturers' Record, Baltimore, May 10, 1917, and " The End of Child Labor," in Survey, Aug. 23, 1919. Some of Murphy's eloquent pleas for aboli- tion of child labor, while courageous and fitting when he wrote, did not, perhaps, recognize sufficiently the facts of the inception of the system. Cf. Present South, pp. 114, 142-143, 147; George T. Win- ston, " Child Labor in North Carolina," in Pamphlet 262 of National Child Labor Committee, p. 1 ff. For a statement true for the eighties but not for 1916, see Hearings before Committee on Labor, House of Representatives, January, 1916, p. 12. 60 Landon A. Thomas, int., Augusta, Ga., Dec. 29, 1916. Others gave similar medleys : " I think the chief advantages observed were the possession of ample raw material and cheap motive power. . . . Also, cheaper common labor, and . . . the fact that the climate . . . is ... a good one . . ." (S. S. Broadus, Decatur, Ala., letter, Jan. 2 7, 1915)- "Mills were located about Spartanburg because they had cotton to grow to their doors, water power, tax exemption, encour- agement in railroads giving two-thirds rate on machinery and mate- rial hauled, and willingness of supply men to take stock " (J. B. Cleveland, int., Spartanburg). 203] THE RISE 0F THE MILL s 97 beginning; he neglected to take account of the subjective factor of popular resolve which flourished just because of the surrounding poverty. 61 " To help the city of Charles- ton and the people was the simple reason for starting the Charleston Manufacturing Company. The projectors thought the time had come for Charleston to do its part ; they had been sending a good deal of money to the Pied- mont mills and they thought they would build one at home." 62 It is not hard to discern several specific influences mak- ing for the industrial development, and these may be ex- amined separately, bearing in mind that all of them, in vary- ing degree, doubtless bore a part. Some, especially in North Carolina, have found a cause in manufacturing made necessary during the Civil War. The State, urged by Governor Ellis and Governor Clark, became a workhouse for the production of war supplies and goods no longer obtainable from outside. It is said that a vision of what lay in manufactures was firmly imbedded in the North Carolina mind, and that after Reconstruction the people went back to industry. 63 61 Facts about South, pp. 20, 22. Contrast R. M. R. Dehn, The German Cotton Industry, pp. 13, 16. 62 George W. Williams, int., Charleston, S. C, Dec. 27, 1916. Cf. a statement respecting development of English economic thought, in Edwin Cannan, Theories of Production and Distribution, ed. of 1894, pp. 147-148. 63 D. H. Hill, int., Raleigh, N. C, Sept. 16, 1016. Governor Clark r in his message to the Legislature, August 16, 1861, stated the situa- tion : " First, that in our commercial relations we have been depend- ent on the North for almost every article that we use connected with machinery, farming, merchandise, food and clothing . . . including almost every article we need for our defence. The second and more important fact is now established, that we have the means and mate- rial for supplying all these wants within our own borders. Neces- sity is developing these resources and driving us to the use of them. The continuance of this war and blockade for two or three years may inflict much personal suffering, but it will surely accomplish our national and commercial independence." Many cotton mills were chartered in North Carolina during the war. " War changes the habits of a people. After the Revolutionary War and the second war with England America relied less on England and became self- supporting. The Civil War changed the habits of the Southern people and made them rely on their own skill and energy for every necessity of life. Where there was no skill, attempts were made to 7 98 THE RISE OF COTTON MILLS IN THE SOUTH [204 Conveniently mentioned, too, in connection with North Carolina is the thought that certain groups of immigrants had planted their manufacturing tradition. This has been referred to in the previous chapter, and it will be remem- bered that slavery and agriculture forbade these foreigners making a lasting public impression. By maintaining an oc- cupation in particular families, however, late members of which came to bear in the industrial awakening, they did a service. 64 Entertaining a synthetic rather than analytic viewpoint, it has been sometimes said, with empirical reasoning, that in- dustry in the South grew out of a natural recovery follow- ing the war. While not accounting very well for a change of mind that was certainly present, this argument has point. A survey of South Carolina in 1884 asserted : " The State has now recovered the ground . . . lost by emancipation, by negro suffrage, by political misrule and official corrup- tion. . . . Since the redemption and regeneration of the State, in 1877, the growth of manufactures has been aston- ishing in its rapidity and volume. Agricultural operations could be carried on with reasonable success, in even the darkest days of strife and 'misrule, but the undertakings which were dependent on the concentration of capital for their development remained torpid, if not dead, until the develop it" (Brooks, pp. 199-200). That "a new form of expression of patriotism took the place of military service " after the Revolu- tion, — encouragement of home industry — is clear; it may be held that such economic patriotism was delayed in the South by Recon- struction following the Civil War, and that industrial progress was thus "the result of both moral and economical forces" (cf. U. S. Census of Manufactures, 1880, " Factory System of U. S.," by Car- roll D. Wright, p. 6). 64 Cf. Clark, in South in Building of Nation, vol. v, pp. 313-314- In a few instances, it is true, local communities were given an indus- trial character that resisted an enervating economic environment. Germans at Wachovia, in North Carolina, within a year after settle- ment had in operation a flour mill, carpenter, shoe and blacksmith shops, pottery, tannery and cooperage establishments (M. R. Pleas- ants, unpublished MS., " Manufacturing in N. C," p. 5). Cf. Tomp- kins, History of Mecklenburg, vol. i, pp. 24-25; Olmsted, p. 511. Winston-Salem, N. C, owes much to its Moravian settlers. 205] THE RISE OF THE MILLS 99 return of confidence breathed into them new life and vigor." 65 A similar account was given twenty years afterward : The war destroyed the capital and property of the South . . . and left in its wake a grinding poverty. . . . The problem of procuring wherewithal to feed and clothe themselves, the fight for a mere sub- sistence, employed all the energies ... of the people. This poverty and the struggle to get rid of the carpet-bag government, left no time for anything else. But there came a time when the people could pull themselves together and take an inventory of what they had accomplished . . . they had a little time to look about them, and to take some thought of the morrow. It required no particular wisdom to see that here where the raw material was produced, where natural resources abounded, and where there was . . . the steadiest and most intelligent class of labor, that in this favored land was the essential home of cotton manufacturing. So it became merely a question of providing capital with which to buy some machinery, the transfer of labor from the farm to the mill, and the South's career as a manufacturing people was fairly begun. 66 The war and Reconstruction took one generation of ac- tivity; by 1880 the South had convalesced. "We took our minds off the war, and began thinking about home affairs." Before 1880 there was a great social pressure that pre- vented attention to constructive measures. 67 65 News and Courier, South Carolina in 1884. " We shall see how the people of this section, reduced to poverty by . . . war, . . . be- stirred themselves cheerfully, amid the ashes and waste of their homes ; how they met new and adverse conditions with unquailing courage ; how they gave themselves cordially to unaccustomed work ; with what patience they bore misfortune, and endured wrongs put upon them through the surviving passions of the war. . . . How . . . at last controlling with their own hands their local affairs, they began, in ragged and torn battalions, that march of restoration and development that has challenged universal admiration. We shall see how . . . things despised in the old days of prosperity, in adver- sity won unexpected value. How frugality came with misfortune, fortitude with sorrow, and with necessity invention " (Grady, pp. 142-143). 66 Southern Cotton Spinners' Assn., proceed. 7th Annual Conven- tion, address of T. C. Guthrie, p. 44 ff. 67 E. C. Brooks, int., Durham, N. C, Sept. 18, 1916. The night of the Hayes-Tilden election the informant's father and uncle sat up all night with their shotguns, expecting trouble with the negroes. A representative of the old South said : " From the close of the war, all through Reconstruction time, we had it pretty hot. Politics took up the time of all of us. The effect of Reconstruction, even after we got rid of it, lasted us six or seven years. When that blew away, everything took on new life. We began to build up all sorts of IOO THE RISE OF COTTON MILLS IN THE SOUTH |^206 Mr. Clark has pointed out that with restoration of confi- dence in political conditions in the reconstructed States, outside capitalists no longer feared disorders that threat- ened safety of investments ; and when work became a neces- sity, opportunities for diversifying work were seized upon. 68 The South, of course, shared in the country's re- vival from the depression that followed the panic of 1873. The recovery symbolized in the return to specie payments in 1879, in its influence on Southern industry, will be spoken of later. 69 One is quite ready to agree to the suggestion, also, that "there was in the South a quiet element of business and professional men who did not approve the course of the leaders of the section, and who, smothered under, so far as public attention was concerned, kept up activity and stood forth when a liberal industrial and commercial program became the order of the day," and that the revulsion of feeling was not really so quick as study of public expres- sions might indicate. 70 The high price of cotton right after the war and a. belief that this condition would continue because cotton could be enterprises " (James Morehead, int., Greensboro, N. C, Aug. 30, 1916). "After they got straightened out, with their State govern- ments in their own hands, people began to feel there was a future for them" (Summerfield Baldwin, Sr., int., Baltimore, Md., June, 1917). Cf. Tompkins, Cotton and Cotton Oil, p. 64; Copeland, pp. 32-33. 68 In South in Building of Nation, vol. vi, pp. 265-266. This state- ment is one of the best-considered by this writer. _ 69 Cf. Clark, in ibid., pp. 262-263, 258 ff . " The growth of popula- tion, the building of railways, the accumulation of capital, the slow perfection of commercial finance, the spread of popular education, each assisted the imperative trend toward industrial diversification and expansion. In spite of the panic and depression . . . between 1870 and 1880 every important Southern manufacture was completely rehabilitated . . ." (ibid.). 70 Walter S. McNeill, int., Richmond, Va., Aug. 29, 1916; M. L. Bonham, Anderson, Sept. 10, 1916. Another said there is nothing esoteric in the cotton mill campaign, that the South was looking about for something to lay its hand to and naturally fell upon the omnipresent staple; the cables that moored the South to its past had worn thin, and it needed only some lucky accidents about 1880 to part the last strands and set the ship free on her course (J. L. Hart- sell, int., Concord, N. C, Sept. 2, 1916). Cf. Clark, in South in Building of Nation, vol. vi, pp. 254-255. 207] THE RISE OF THE MILLS IOI only scantily raised with free labor, focused attention again upon the staple ; the local merchant was given credit at the North, and he in turn gave credit to the farmer, who pledged his land to cotton. This temporary restoration of King Cotton saddled the farmer with debt and delayed agricul- tural diversification and industrial beginnings. 71 In coming to the directly personal factor, the part of pro- moters and projectors in the building of the mills, it is well to bear in mind the caution that " it is . . . not unnatural that most of us should fall into the error of attributing to the influence of prominent individuals or organizations the events and conditions which the superficial observer regards as the creation of the hour, but which are in reality the out- come of a slow and continuous process of evolution." 72 In certain cases where it would seem plain that mills were due exclusively to one man, it is necessary only to ask where and why he received his impulse, to show that he was really an exponent of a prevailing tendency, just as the commu- nity upon which he relied for assistance, in its response to his appeal, answered a little later to the same social stirring. 73 71 Cf. Grady, p. 175 ff. For other references, see Tompkins, Cul- tivation, Picking, Baling and Manufacturing of Cotton, pp. 5-6; History of Mecklenburg, vol. i, pp. 150-151 ; Thompson, p. 59 ff. 72 Plunkett, p. 27. The mistake has often been made : " You might write volumes, but you would never be able to get beyond the fact that the cotton mill development in Gastonia, Gaston County, North Carolina, and the whole South, is the result of the fact that a few men had a vision " (Joseph H. Separk, int., Gastonia, N. C, Sept. 14, 1916). Cf. Southern Cotton Spinners' Assn., proceed. 7th An- nual Convention, address of Edward Atkinson, p. 89; Cannan, p. 23. 73 One of the sincerest men talked with said : " The Gaffney people never thought of having a mill before I came back from the Clifton village, where I was putting up buildings, and got them stirred up. You get an idea in another place where you happen to be, and you say to yourself : ' Why won't that work in our little town ? ' Well, you've got to do a lot of talking after you get home with the idea, but they'll catch on in the end. The people of Union asked some Gaffney men to come there and tell them about the business. The professor at the high school and I went down to Union, and I recol- lect I made them a right smart good talk down there, and they caught on to it and built the mills they've got now. And that was a dead town" (L. Baker, int., Gaffney, S. C, Sept. 13, 1916). An interview with a mill official whom Mr. Baker persuaded to come to Gaffney showed that he had acted so completely under Mr. Baker's enthusiasm that he accepted the factory as a matter of course. In 102 THE RISE OF COTTON MILLS IN THE SOUTH [208 It has been seen how Murphy pointed out that the New- South was the child of the Old South, fathered in large de- gree by the same leaders who in less happy days had bred only economic deformities. " The old South was the real nucleus of the new nationalism. The old South . . . was the true basis of an enduring peace between the sections. . . ." And everyone must share his regret that "a doubt was put upon its word given at Appomattox. . . . Power was struck from its hands. Its sense of responsibility was wounded and confused." 74 Nothing stands out more prominently than that the South- ern mills were conceived and brought into existence by Southerners. The impulse was furnished almost exclu- sively from within the South, against much discouragement from selfish interests at the North, and capital was supplied by the South to the limit of its ability. 75 Coming now to the part of ex-Confederates in the indus- trial regeneration of their people, it is apparent with what speed they embraced their new duty and how the promise of their participation was welcomed by the wisest heads in no instance did one personality stand out as an almost exclusive influence more than in the development of mills at Columbia through Mr. Whaley (Washington Clark, William Banks, W. W. Ball, inter- views, Columbia, S. C, Jan. I, 2, 3, respectively, 1917). "If I had at my disposal the history of Major Thos. L. Emry, who was the founder and father of Roanoke Rapids, I would simply tear a few pages from it and spread them across this space and you would have the whole story of the pioneering of this wonderful industrial . . . center" (Charlotte News, Textile Ed., 1917). 74 Present South, pp. 10-11. Such men as E. M. Holt, Francis Fries, J. M. Morehead and William Gregg, who years before had seen the wisdom of industrial development along with agriculture and, besides the usual activities of farmer and legislator, were en- gaged in building railroads and mills, could not have their way with the South. Men of opposite faith, later converted to new courses, did no more than adopt a program which earlier had been spurned. On these unfollowed leaders, see South in Building of Nation, vols, xi and xii; Cyclopedia of Eminent and Representative Men of the Carolinas ; Jerome Dowd, Sketches of Prominent Living North Carolinians (1888) ; Biographical History of North Carolina; Tomp- kins, Cotton Mill, Commercial Features, pp. 181, 185, 187-188; Clark, in South in Building of Nation, vol. v, p. 323 ; Copeland, pp. 32-33 ; Goldsmith, p. 4; Southern Cotton Spinners' Assn., proceed. 7th Annual Convention, p. 168. 75 Cf. Grady, pp. 182-184, 197-198; Edmonds, p. 32; Charlotte News, Textile Ed., 1917. 20g] THE RISE OF THE MILLS 103 the North. It has been observed that just as citizens of Salem, Massachusetts, established a club where descend- ants of the witches and of those who hanged them toast one another, so " the same people that turn out, by the city- full, to build Lee's monument and to bury Davis, are taxing themselves for the schooling of negro children. . . ." Each of these Southerners, "devoutly remembering the old; understanding as no one else can why he remembers it; but all the time looking for something not only better and larger than he has known, but grander than any one ever dared to hope for this side of heaven," showed a divine versatility that is the very stuff of civilization. 76 James L. Orr, soon to be governor of his State, was a type man, and he appeared with others of the same persua- sion in South Carolina as early as the constitutional con- vention of 1865. There was nothing sullen about him; what he did, he did whole-heartedly. " He was considered one of the coolest-headed men in the State five years ago this summer; but, for all that, he was one of the leading members in the Secession Convention, and in the Rebel Senate during the whole existence of the Confederate gov- ernment. Now he is one of the leading reconstructionists. . . . He . . . carries himself with a very democratic air." 77 Often reprobated at the North, this was as normal as it was fortunate; Southerners would choose those in whom they had rested old confidences because people and spokes- men had made a mental readjustment which, however unbe- lievable to enemies, was easy and natural. 78 A Northern observer not over-disposed to find good in the beaten South, disagreed with those who wished to antagonize and hinder 76 A. D. Mayo, "Is There a New South?", in Social Economist, Oct., 1893, pp. 201-202; cf. ibid., p. 207. _ 77 Andrews, p. 50. The promptness with which distinguished par- ticipants in the Confederate cause came forward after the war was an indication of the consistency of Southern leadership. Lee was as much the general at the head of a college as at the head of an army. 78 Cf. Dunning, pp. 44-45. Twelve members of the South Caro- lina constitutional convention of 1865 had been members of the secession convention (Andrews, pp. 38-39). Cf. Thompson, pp. 57-58. 104 THE RISE OF COTTON MILLS IN THE SOUTH ^2IO rather than help these inevitable leaders. " For my part," said Sidney Andrews, " I wish every office in the State [South Carolina] could be filled with ex-Confederate sol- diers. It is the universal testimony of every officer of our own troops . . . that the late Rebel soldiers are of better disposition toward the government, toward Northerners, toward progression, than any other class of citizens."' 79 The year 1880 was reached before these men could really assert themselves. Their training in politics stood them in good stead when they came to organize public sentiment in a new campaign, that of industrial awakening. 80 Their old mastery, with even increased power, sprang forward to the evident task. The pity is that they had not longer time left them in which to work for the South. 81 tcdJU When the student of Southern industry meets one of the few surviving members of this company, he at once feels himself in touch with the spirit that was the South's salva- tion. Far-seeing, public-minded, generous-natured leaders because lovers and servers, these have proved themselves true patriots. 82 79 South since the War, p. 95; cf. ibid., pp. 393, 371-372; Dunning, pp. 185-186; Tompkins, History of Mecklenburg, vol. i, pp. 151-152. Andrews had more faith in a "conquered Rebel" than in "most of these North Carolina Unionists" (ibid., p. 167). 80 Cf. Punkett, pp. 72-73. 81 The Industrial South, of Richmond, in 1882 was asking "when will . . . prosperity come" and declared this especially "the impa- tient utterance of the surviving veterans of the war . . . — the men who were crushed to the earth by the loss of all their worldly pos- sessions, who have ever since been struggling for a footing in life again, and who are looking longingly for some assurance . . . that their children and their children's children will have large oppor- tunities for improvement of their fortunes through the exercise of energy in utilizing the bounties of nature around them " (quoted in Baltimore Journal of Commerce and Manufacturers' Record, June 17, 1882). 82 It is worth while setting down impressions of one or two. A ruddy, white-haired old gentleman, cordial, cultivated, and a little shyly if gladly reminiscent, received me in the office of his ship chandlering store in Bay Street, Charleston. He showed by look and phrase and occasionally by direct remark that he had been nicely familiar with all the details of the important mill project in which he had taken a chief part, and that they had been obscured, never obliterated, by the years. Almost with a child's embarrass- ment he explained he did not know why he had kept a packet of 2Il[] THE RISE OF THE MILLS I OS It was natural, at the evening of the period, that many cotton factors should head mill enterprises. They had some money, business connections and a knowledge of the staple that was important. Frequently they had been buyers for, and stockholders and directors in, some of the first enterprises. Charleston cotton merchants played a leading role. Captain F. T. Pelzer is a case in point; he made money in cotton right after the war, was a director in Hammett's Piedmont Factory, became interested in several other ventures and ended by founding the mill bearing his name. 83 Sometimes factors were already executives of mills established before 1880, and went into manufacturing more deeply when industrial development became a fixed policy. 84 A cotton buyer for Hammett's Piedmont mill, intimate memoranda concerning a devoted but unsuccessful venture, and handed them over to me with charges for their safe return. Another, nearly ninety-eight years old when interviewed, sat in the office of the mill which he built and of which he had long been president. He wore a greenish-black, threadbare overcoat, and clutched a bulging umbrella of the same sort. Clear-brained, almost excessively direct, of dominating personality, one felt he had always been equal to the tasks confronting him. As he looked about at the bookkeepers, he spoke with much emphasis, to soften the too piti- fully evident chagrin at being dispossessed. His successor, exulting in what has been called "juvenile capitalism," had little of the affection of the old man for the enterprise. Here was the South of slavery, agriculture and aristocracy, that made the South of free labor, industry and democracy. A writer on the mills has said : " These little personal things will creep into my story and break the continuity of dry developments, but the human element pulsates so frequently through the proposition that I must be excused if I fly off at a tangent at almost any word and mix up the material and the psychological" (Charlotte News, Textile Ed., 1917, concerning Roa- noke Mills). Cf. ibid., concerning Edenton Cotton Mill; Southern Cotton Spinners' Assn., proceed. 7th Annual Convention, p. 44 ff. 83 Frank Pelzer and William Banks, interviews, Charleston, Dec. 28, 1916, and Columbia, Jan. 2, 1917. Col. W. G. Smith was a cotton buyer in Orangeburg County, S. C., before organizing a mill there. Leroy Springs, in his mercantile business at Lancaster, took cotton in exchange for goods. A little later it will be noticed how he became a cotton manufacturer (Banks, ibid.). The same is true of John H. Montgomery (cf. Columbia Record, Textile Ed., Oct., 1916). 84 Thus " William C. Sibley, the president of this company, has had a long experience in cotton, and is one of the largest cotton buyers of the Southern States. . . . He is at present handling, and has for many years . . . efficiently handled, the corporate affairs 0$ the Langley mills . . ." (Boston Journal of Commerce, July 29, 1882). 106 THE RISE OF COTTON MILLS IN THE SOUTH [212 after plans to invite him to Anderson had fallen through, started a factory on his own account. By no means all of the mill builders had direct or even indirect connection with cotton. How thoroughly local most enterprises were and how general in communities was the desire for mills, is seen in the callings from which men came to cotton manufacturing. Lawyers, bankers, farmers, merchants, teachers, preachers, doctors, public officials — any man who stood out among his neighbors, or whose eco- nomic position allowed him a little freedom of action, was likely to be requisitioned into service or to venture for him- self. 85 Neither did the South rely only upon those socially prominent, or upon intellectuals. There was no authorita- tive leading exposition of the problems facing the section. Measures were hit upon by intuition, by force of circum- stance, because of pressing necessity and first-apparent op- portunity. It was a movement of the whole South. 86 Especially did merchants become mill builders. When large plantations broke down into small farms and tenant holdings, factors at the ports could no longer market the whole cotton crop or supply needed credit, because they did not have knowledge of local conditions. Merchants, many of them mere country storekeepers, found themselves more than ever drawn into the buying and selling of the staple, and lending money on growing crops. Supplying every material want of the farmer and taking his incoming cotton in surety, the merchant was the pivot of the economic sys- tem. These merchants, more than anyone else in their com- munities, had credit relations at the North, the importance of which in their manufacturing enterprises will be ob- served in the chapter on capital. 87 They had an interest 85 Cf . Goldsmith, pp. 7-8. " All that was necessary was that the promoter of the mill should have succeeded in the business in which he had been formerly engaged." 86 Cf. Plunkett, p. 133 ff. 87 Cf. Hammond, p. 144 ff. The description of usury in sales on "time" applies over much of the South today. I heard of a North Carolina farmer who, in 1920, with cotton prices unprecedentedly high, was asked half of his crop in payment for fertilizer. 213]] THE RISE 0F THE MILL s 107 in the prosperity of their localities that was none the less effective because not always academic or sentimental. 88 The man who later had capital to head cotton mills at Clinton, South Carolina, set up the only store in the place immediately following the War. The town was then in a poor way, and 1 not being helped by several barrooms. 89 Leroy Springs operated a general mercantile business in Lancaster, South Carolina. A new railway came through the town, but some thriving young places sprang up along it and drew business from Lancaster, which came to a standstill, was " dead." Realizing that something must be done to keep business going in Lancaster, the merchant, with a small capital, built a cotton mill. It had the desired effect. 90 In a conspicuous instance a mercantile firm outside of the South entered the industry in a similar way. Ap- pealed to by Southern customers to take stock in local mill undertakings, a Baltimore groceries house came to have large manufacturing interests, and ultimately changed over to making and selling cotton goods. The moving man in the Charleston Manufacturing Com- pany was half merchant a9 head of the ice company. 91 Saying that Massachusetts mills created only in quantity values that Rhode Island manufacturers produced through quality of goods, William Gregg years before saw that the Massachusetts method would first be introduced in the South, and said : " Cotton manufacturing will not, probably, be speedily introduced into this State [South Carolina], unless our business men of capital take hold of it. Mer- chants and retired men of capital may erect factories, . . . 88 Mr. Clark is hardly justified in asserting that " The conditions were no longer those that attract a few hardy adventurers into a new field of business, but such as draw conservative capital, in large units and in the hands of trained administrators, to assured spheres of enterprise" (cf. in South in Building of Nation, vol. vi, pp. 266-267). Administrators were rarely trained and large investments of capital were rarely for dividends only. 89 Columbia Record, Textile Ed., 1916. Cf. ibid, respecting John R. Barron's Manchester Cotton Mills. 90 Columbia Record, Textile Ed., 1916. 91 Cf. advertisement of Alva Gage & Co., Deutsche Zeitung, Charleston, 1881. 108 THE RISE OF COTTON MILLS IN THE SOUTH ^214 our wealthy planters may engage in this business . . . but it will be long before the Southern States shall have a set of 'manufacturers similar to those in Rhode Island ; they must grow up among us. . . ," 92 When at length Gregg's advice was being followed, and entrepreneurs had to be recruited quickly, it is surprising how few failures ap- peared. True, they were not manufacturers in Gregg's sense, but they worked under natural advantages which well-nigh insured success. 93 Mr. Estes' entry into manufacturing was typical. " I was first in the dry goods business, then the grocery business. I was mayor here for six years. I was successful. They got me to get up the mill. Old Judge King took first $50,000 and then $100,000 of stock, with the idea that I was to be president." 94 92 Domestic Industry, p. 27. 93 Some mistakes there were bound to be, when an industry was being built overnight. " Gen. Irving Walker, a stationery man, was the first president of the Charleston Manufacturing Company. He was a nice man, but he knew nothing about the business. That was at the bottom of nine-tenths of the failures of cotton mills in this State — the presidents were popular, you know, everybody liked them, but they were incompetent, with no technical knowledge." The founder of one mill, mayor of his city, was denominated " a hot air politician." The type of man who could succeed in the eighties usually fails now. Mills at Bessemer City, North Carolina, are illustrative. There is no longer the leeway. Cf. Thompson, p. 272. " Looking back on them," one informant said, " I can see that the first mill men were a set of blundering children, some a little more apt than others." Cf. Ga. Indus. Assn., proceed. 4th Annual Con- vention, address of J. J. Spalding, pp. 46-47, and the writer's " Some Factors in Future of Cotton Manufacture in South," in Manufac- turers' Record, Baltimore, May 10, 1917. Newer manufacturers, though still bearing marks of neglected training, are supplementing " a deep desire to succeed, faith in the soundness of the task and in one's own self, and business and social imagination " with intimate knowledge of detail. Cf. Charlotte News, Textile Ed., 1917, for many instances, especially that of C. B. Armstrong. 94 Charles Estes, int., Augusta, Ga., Dec. 29, 1916. " When this enterprise was inaugurated there were those who doubted whether the mill would ever be built, but with Mr. Charles Estes, to whom, by universal consent, the work of organizing the company was in- trusted, there is no such word as fail . . ." (newspaper clipping in Raworth Scrapbook). "Mr. A. Scheurman, a leading merchant of Griffin, Ga., is now closing out his business with the intention of engaging in cotton manufacturing . . . (Manufacturers' Record, Baltimore, Dec. 14, 1882). Cf. Grady, p. 181 ; Southern Cotton Spin- ners' Assn., proceed. 7th Annual Convention, pp. 111-1x2. 2I5J THE RISE OF THE MILLS IO9 If space .permitted, a review of the histories of H. P. Hammett, G. A. Gray, R. C. G. Love, Daniel Rhyne, and others would show interestingly the channels by which chief leaders came to build mills. Hammett and Gray were both linked with the ante-bellum industry. The former grew up on a farm, taught school, married the daughter of William Bates and was taken into his cotton manufacturing com- pany ; he entered the Civil War and was given duty in the Confederate tax office. After the war he represented his native county in the State legislature, was mayor of Green- ville, 'made president of a rundown railroad and, knowing men of influence and being acquainted with an excellent water power, built his mill in the seventies. 95 Gray at the. age of eight entered the old Stowe Mill, at Pinhook, Gaston County, N. C, as a doffer boy, at ten cents a day. He at- tended school hardly at all. He became an overseer in this mill and later was in charge of the installation of machinery in various' new plants. He was superintendent of several factories, and -moved to Gastonia with a small capital and built the first 'mill in the place. At the time of his death the town had eleven mills, nine of which he had been largely instrumental in projecting. 98 D. A. Tompkins, less than thirty at the opening of the cotton mill era, was late enough to profit by the pioneering work of others. There was enough industry in the South to make the mill engineer's profession profitable. Tomp- kins was one of the first men in whose career it was evident that the South was becoming a real seat of the cotton manu- 95 Cf . Tompkins, Cotton Mill, Commercial Features, pp. 180-190. 96 Cf. Gastonia Gazette, Feb. 9, 1912, and C. W. Patman in Knit Goods, N. Y., March, 1912. Men drawn into the business as a result, directly or indirectly, of Gray's influence, had been teachers, public officials, bankers and farmers. ' Cf. Patman, ibid. One of the most successful manufacturers in Gastonia was raised on a farm and ran away from home when a young man and went through the country peddling clocks and quilts. He said that as a boy he believed the treasurer of a cotton mill the biggest man in the country, and that he thought this over while tramping about later with his wares. He finally set up an instalment furniture business in Gastonia, was made sheriff and was elected to the vice-presidency of a cotton mill. He is now president of many factories (C. B. Armstrong, int., Gastonia, N. G, Sept. 14, 1916). 110 THE RISE OF COTTON MILLS IN THE SOUTH ^2l6 facture, with facilities for machine design and repair and all that general guidance which new companies needed. Tompkins, unlike his predecessors, had technical training. Partly in furtherance of his engineering business, partly from the broadest social motives, he almost raised by hand many cotton mills — inspiring the idea, supervising con- struction, assisting first steps in production. 97 " When manrifactures have become well established a new mill is sometimes organized by a number of men who per- ceive that some one man is a promising manufacturer." 98 Though frequently exemplified at present, notably at Gas- tonia, this was rarely true in the early years. Where an executive was not himself the projector of a mill, the thought that created it was not dependent upon any indi- vidual. Thus a nearby manufacturer was persuaded to become president of a factory at Albemarle, North Caro- lina; but the enterprise had its root in the local pride of an old farmer of the place. 99 There has been an erroneous notion that many promoters of mills in the South in the eighties were Northern men and firms. In the beginning cotton manufacturers of New Eng- land did much to discourage establishment of the industry at the South, and have never sought to realize Southern ad- vantages in a large way. In later days there was never such an incoming of Northerners as that of the Hills, Bates, 97 As an exhibit in Southern economic history, his two little offices in Charlotte ought really to be moved intact to a place where they would be kept for the public. Glimpses revealing Tompkins' per- sonality may be found in his own writings : Nursing and Nurses, p. 3 ; History of Mecklenburg, vol, i, p. vii ; Cotton and Cotton Oil, preface; Road Building and Repairs, p. 26; Building and Loan Asso- ciations, preface; a notion of the character of his service may be gained from : Water Power on the Catawba River, p. 20 ff . ; A Plan to Raise Capital for Manufacturing, p. 18 ff . ; the backs of some of his many pamphlets give the plan of organization of his company. 98 Tompkins, Coton Mill, Commercial Features, p. 30. 99 J. L. Hartsell, int., Concord. This was true of other mills headed by the same man. A merchant fathered a mill at China Grove quite irrespective of its later foster parent (W. R. Odell, int., Concord, N. C, Sept. 2, 1916). At Salisbury the matter of an executive was an afterthought. 217] THE RISE OF THE MILLS III Shelden, Clark and Weaver. 100 The Cotton Mill Campaign, so far from showing any of the antagonism of former years toward Northern men and money, 101 developed the most enthusiastic desire for the cooperation of any outsiders. This led to delighted acclamation of any reported design of Northerners to set up manufacturing in the South, but there were undoubtedly more reports than performances. The Northern observer who said of Charleston directly after the war that the city had not sufficient recuperative power for its own rebuilding, and that New Englanders, if anyone, must make it over, would have been surprised fif- teen years later to see Charlestonians supplying impulse to their own city and really to the whole South. 102 A Southern writer in 1882 said : " Capital and skill are the only things needed to make the South preeminently a manufacturing country and shrewd, energetic men from the East and from Europe are rapidly supplying the defi- ciency." 103 The Southern Land, Emigration and Improve- ment Company, a New York organization, had as one of it's purposes bringing investment opportunities to the attention of Northern capitalists. 104 The leading railroads travers- ing the South Atlantic States combined on a similar plan in 100 Cf. J. B. O. Landrum, History of Spartanburg County, p. 58. Cf. Dehn, German Cotton Industry, pp. 4-5. 101 Andrews, p. 378. 102 Ibid., p. 3. 103 Baltimore Journal of Commerce and Manufacturers' Record, June 3, 1882. " The Barnett Shoals have at last been purchased by Davenport, Johnson & Co., of New York, for $22,000, and it is thought that work will begin here in a short time. Athens is yet destined to be the Lowell of the South" (Athens Banner, quoted in ibid., June 24, 1882). "A party of New York capitalists . . . visited the Peach Stone Shoals in Henry County, Ga., a few days ago, and were so much pleased with the property . . . that they purchased it with the intention of at once erecting a cotton mill, to rank as one of the largest and best equipped in the South" (Manufacturers' Record, Baltimore, March 1, 1883). Cf. Fredericksburg correspond- ence in Daily Dispatch, Richmond, Feb. 23, 1880; Charlotte News, Textile Ed., 1917, concerning Wayne Mill. The considerable North- ern participation in Southern mill development, as in the Chadwick- Hoskins group and the Pacific Mills, came much later. Cf. Char- lotte News, ibid. 104 Baltimore Journal of Commerce and Manufacturers' Record, Aug. 5, 1882. 112 THE RISE OF COTTON MILLS IN THE SOUTH \_2 1 8 1880. 105 Not a few of the Northerners that undertook in- dustry in the South did so at directly Southern solicitation. A mill man of Alabama in 1881 was consummating a con- tract " by which a New England company of capitalists will revive cotton manufacturing in a factory building at Cor- inth, Miss.," and he expected to induce a Connecticut spin- ner who wished to come South to remove to Huntsville, Alabama. 106 The pages of this study bear ample testimony to the power of a different group of promoters from those here men- tioned, namely, the editors of Southern newspapers. Being peculiarly a public movement, the cotton mill development sprang in large part from the activity of the press. The " Federalist " did not fight harder for union than Southern papers, big and little, strove for industrial awakening in the eighties. By 1880 most editors knew that they were to follow DeBow and not Bledsoe, working for understanding between the sections and not separatism, for diversity and not narrowness of economic pursuit. 107 County weeklies were stout followers in a campaign in which city dailies were leaders. No paper was more influential than the News and Courier, of Charleston. The philosophy of Gregg's " Essays on Domestic Industry," published in the Charleston Courier in 1845, was made concrete in the News and Courier's exhortation, " Bring the mills to the cotton," which rang throughout the South and was taken up as the rallying cry of every mover for industry. 108 105 Ibid., July is, 1882. 106 Huntsville Democrat, quoted in News and Courier, Charleston, July 30, 1881. Similar bidding for location of a business designed to be set up in the South has been seen later in the case of a Phila- delphia carpet manufacturer being brought to Gaffney, S. C. 107 Cf . Dodd, in South in Building of Nation, vol. vi, p. 546. 108 Mr. Hemphill declares this sentence " resulted in the conver- sion of South Carolina in less than the life of a generation into the second cotton manufacturing state of the nation. . . ." It is " by its statesmanship and largely through the work of its press" that the South has achieved progress since the war (James C. Hemphill, " The Influence of the Press in Southern Economic Development," in South in Building of Nation, vol. v, pp. 548-549. _ See this whole paper for interesting material). In 1880, in presenting a survey of 219] THE RISE 0F THE MILLS IJ 3 The very genius of the News and Courier was its editor, F. W. Dawson. The power he exerted could never be duplicated under any other circumstances than those of the South in the eighties. His inspiriting force can scarcely be overestimated. Born in England in 1840, he was drawn to the Southern cause and enlisted as a sailor on the Confed- erate vessel Nashville at Southampton the opening year of the Civil War. He entered the army and was promoted to the rank of captain. After the war he worked on Rich- mond newspapers before becoming one of the proprietors of the Charleston News ; his editorship came to full strength when thi9 paper was consolidated with the Courier. 109 He saw the truth of the South's problem and saw it whole; that the people must drop conceits and go to work. Be- sides his major effort for cotton 'manufactures, he effec- tively urged tobacco cultivation in South Carolina and preached against duelling. Doubtless his foreign birth and knowledge of English industrialism was of great assistance to him. Of fine physique, handsome, imperious, brilliant, level-headed, "he had full confidence in himself, with good reason. He was a godsend to South Carolina — the leader in bringing the State back into its own." 110 He met a tragic the mills of South Carolina, Dawson wrote: "Ten years ago The News and Courier formulated what is now an accepted truth, in declaring that the remedy for commercial distress in the North and the secret of sure fortune in the South was to bring the mills to the cotton. . . . The belief was that the manufacture of cotton in a cotton producing State must necessarily pay well, by reason of the saving in the cost of the raw material, by the saving in commissions, and charges for transportation, by the saving in waste, in the rental of land . . . and in the wages to be paid to operatives . . ." (Lead- ing article, published in Blackman). For an instance of how this idea was acted upon in Texas, see Manufacturers' Record, Balti- more, Nov. 23, 1882. Dawson stressed the slogan " Bring the cotton mills to the cotton fields" and associated ideas, "and kept hammer- ing them, until some fellows caught the point and began to build mills" (W. W. Ball", int., Columbia, Jan. 1, 1917). Interviews with Messrs. James Simons and F. Q. O'Neill of Charleston and Tracy I. Hickman, of Augusta, bore out the same point. Cf. Kohn, Cotton Mills of S. C, p. 20. 109 Cf. South in Building of Nation, vol. xi, p. 271. 110 Yates Snowden, int., Columbia, S. C, Jan. 1, 1917. I am in- debted to Professor Snowden, of Columbia, and to Messrs. W. H. Parker, W. P. Carrington, F. W. Wagener and William M. Bird, of I 14 THE RISE OF COTTON MILLS IN THE SOUTH ^220 death, being murdered in 1889 by a physician whose office Dawson entered, it is said, in order to resent an affront to an Irish servant girl in his home. 111 Dawson was not an orator, and had none of the flourish of Grady. Also because his attack was more direct and concentrated than that of his Georgia contemporary, he is not so well known now. Of Dawson as of Grady it may be said : " His influence in exciting hope and inspiring confi- dence in the ability of the South to cope successfully with her difficulties was immeasurable. . . . ' He did not tamely promote enterprise and encourage industry ; he vehemently fomented enterprise and provoked industry until they stalked through the land. . . .'" 112 The old Baltimore Journal of Commerce in February, 1882, began to devote some pages to industrial development in the South, changing its name to the Journal of Commerce and Manufacturers' Record. In November of the same year the Manufacturers' Record became a separate paper, because " it has been demonstrated that there was an actual need for a paper which would adequately represent the Charleston, for descriptions of Dawson. His picture shows rather curly dark hair, fine, searching eyes, fullish lips, long, somewhat irregular nose and a strong jaw — the face of a thoroughbred. 111 If Professor Hart has reason for a late statement that the News and Courier " has for its stock-in-trade, ultra and Bourbon sentiments " and " represents an age that is past," such a comment would not apply to the paper's earlier history (p. 70). As a single illustration of its position in the South's critical years, articles writ- ten by a member of its staff and reprinted under the title " The Cotton Mills of South Carolina," in this study referred to as " Black- man," with a striking editorial from Dawson, should be examined. How dynamic was the paper's advocacy of industry comes out in the complaint of a manufacturer opposed to the use of the Clement Attachment (a machine which represented the extreme of the doc- trine of " Bring Mills to Cotton " in that it accomplished both gin- ning and spinning), that "the newspapers are assuming a great deal of responsibility in giving it so much notoriety" (Blackman, p. 12). 112 Dyer, in The New South, pp. 78-79; cf. ibid., p. 128. Grady gave as the text of the Atlanta Constitution: "If the South can keep at home the $400,000,000 it gets annually for its cotton crop, it will soon be rich beyond comprehension. As long as she sends it out for the supplies that make the crop, she will remain poor " (pp. 219-220). In this thinking, however, the News and Courier pre- ceded the Constitution by ten years. 22 i] THE RISE OF THE MILLS I 15 manufacturing interests and would keep abreast of the rapid improvement in the material affairs of the South." 113 The Manufacturers' Record caught a spirit which had its birth in the heart of the South; as its own words show, it was to " represent " and " keep abreast of " industrial de- velopment in the South rather than originate this in the first instance. In its best years it was a useful popularizer of Southern opportunities. 114 A fundamental cause of the building of cotton mills in the South, really self-evident, was an awakening to the ad- vantage of adding the profits of manufacture to those of production of raw material. 115 Dawson in 1880 put the matter in simple terms : The point on which we lay the most stress is that, to the extent in which cotton . . . produced in South Carolina is manufactured in the State, the whole of the profit upon that cotton, from the first stage to the last, remains in some form within the State for the benefit of its people. Where the cotton is produced here and manu- 113 Cf. Baltimore Journal of Commerce and Manufacturers' Rec- ord, Nov. 18, 23, 1882. 114 Cf. Hemphill, in South in Building of Nation, vol. vi, p. 539. 115 This primary consideration had been explained to an earlier unheeding generation. Gregg said that coarse goods mills in Massa- chusetts presented " a fact that cannot but strike a cotton planter with great force, viz : that 174 hands in 12 months, convert 4,329 bales of cotton . . . into cloth . . . thus adding over $40 to the value of each bale" (Domestic Industry, p. 27 ff.). "Have we not the raw material on the spot, thus saving the freight of a double transportation?" (cf. ibid., pp. 21, 24-25). It was shown a little later that Tennessee cotton planters made only uj4 per cent profit, while manufacturers of the same crop made 24 per cent, and it was asked : " Are there any so blind as not to see the advantages of the system?" (DeBow, vol. i, p. 126). A writer in 1866 quoted an advocate of the " cotton-field system " of manufacture of seven- teen years before, who declared that " the spindles and looms must be brought to the cotton fields. This is the true location of this powerful assistant to the grower," and that to bring mills to cotton " is but one move, whilst sending the cotton to the mills is a heavy annual, perpetual tax," and proceeded to estimate how this could be cheaply accomplished (Barbee, The Cotton Question, p. 138 ff.). In 1878 total costs of manufacture in Lowell, Mass., and Augusta, Ga., were shown, leading to the conclusion that in freight, commis- sions, insurance and exchange the Augusta manufacturer saved $6.62 per bale over his New England competitor on goods shipped to New York, and $10.23 on those sent direct to the West. For Montgomery to double the value of its cotton in this way " is our right, and our duty . . ." (Berney, Handbook of Alabama, p. 271). Il6 THE RISE OF COTTON MILLS IN THE SOUTH [222 factured elsewhere South Carolina is in the position of furnishing the elements which make other communities rich; ... we know that the wealth of New England is due to the profit made upon the manu- facture of the raw material which the South supplies, and which the South . . . buys back from New England at a high price in its manu- factured state. 116 A Southerner speaking at the Atlanta Exposition asked "by what rule of political economy should the Southern people send their cotton at an expense always deducted from its price, to distant sections and foreign countries to the spun and woven . . . ," and told his listeners that " Here the cotton grows up to the doorsteps of your mills, and supply and demand clasp hands. . . ." 117 116 Leading article, in Blackman. Besides many collateral benefits, the eighteen mills of the State converted cotton worth $1,631,820 into manufactured goods worth $3,932,150. An editorial headed " The Gold in Cotton," said : " At present Charleston does nothing to increase the value of the cotton which comes here for sale. It leaves us as it finds us. The city lives on the pickings and scrapings. . . . Cotton mills change all this. A bale of raw cotton, worth forty dollars, is spun into yarns or cloth worth eighty dollars. There is the usual profit in buying and selling the cotton, and, in addition to this, Charleston gets forty dollars a bale, which goes into our purses and comes out of the pockets of the persons who consume the goods" (News and Courier, Charleston, Feb. 2, 1881). 117 Ibid., Oct. 10, 1881. "... every Southern man is sure to prove to you that it is a dead waste to ship raw cotton to a mill 1,500 miles away when it could be made into yarns and fabrics much cheaper in factories distant from the cotton field only a short half- day's journey for a mule" (Atlanta correspondence of New York Times, quoted in ibid., Nov. 5, 1881). Cf. Richmond Dispatch, quoted in ibid., March 25, 1881. " We have the raw material — New England takes it and augments its value by her labor . . . we, too, must endeavor to mix skill and labor with our raw material before letting it pass from our hands . . ." (Observer, Raleigh, March 2, 1880). Cf. ibid., June 6, 1880; News and Courier, Charleston, April 25, 1881. "Freights were high then; it was a great argument that we saved by manufacturing the cotton here and shipping the goods to the North for just what it cost to send the cotton there " (Charles Estes, int., Augusta). Hammett told a meeting of the South Caro- lina Agricultural and Mechanical Society : " The South is fitted for the cotton manufacture, which adds profits and value of labor to value of raw material," showed that the South had an advantage of 10 to 20 per cent over New England in the business, and counted the benefit to Southern communities through establishment of mills (News and Courier, Charleston, Aug. 1, 1881). "There is no rea- son why the South should lose the entire profit upon manufacturing cotton and be content to gain only the beggarly profit of producing it, while England and the North grow rich upon handling it . . ." (Baltimore Journal of Commerce and Manufacturers' Record, Sept. 223] THE RISE OF THE MILLS 117 No one did more to impress this idea upon the South than Tompkins ; he presented it in primer-like plainness and from every angle. It is difficult to realize how far this was from a truism to the South even by the time he wrote and spoke. 118 The question whether the South should manufacture cot- ton or be content with cultivation of the raw material was made vivid by the opposition to Southern mills on the part of Edward Atkinson, of Boston. It may almost be said that he conducted a propaganda to show that the South should devote itself to raising, ginning and preparing the staple to be spun and woven elsewhere. A talented organizer of business, a not unkindly egotist, officious without being patronizing, gifted in social imagination, and one of the first New Englanders to concern himself actively in a public way with Southern economic affairs after Reconstruction, Atkinson sought, sometimes with semi-private purpose, to mirror the South to itself. The image he furnished, by its very distortion, assisted Southerners to a clearer view of their task. At that peculiar juncture in the South he was listened to attentively, and negatively and positively exerted a striking influence. 119 2, 1882). Cf. ibid, for quotation of an expressive illustration of this thought; a labored explanation is given in ibid., June 3, 1882. 118 An ordinary county producing 10,000 bales would get, at 6 cents a pound, $300,000 for its cotton; if sold as cloth at 18 cents, this cotton would bring $900,000. " Assume that this cloth was shipped to China instead of shipping the raw cotton to England and it becomes evident that the English cotton buyer sends here' $300,000 while the Chinaman would send $900,000," and he showed how this $600,000 increment would be distributed ; that, also, fac- tories would bring other benefits by increasing the value of raw cotton and of farms, creating a market for perishable produce and affording diversity of employment to members of the community (Cotton Mill, Commercial Features, p. 16 ff.). Cf. Observer, Raleigh, May 19, 1880, in comment upon Winston and Durham. Cf. Tompkins, ibid., pp. 23-24, 177-178; History of Mecklenburg, vol. i, p. 24. Tompkins' ingenious little book, " Cotton Values in Textile Fabrics," is an object lesson on this point. 119 See Who's Who in America, 1906-1907, p. 54. The range of his writings, from the science of nutrition to the cost of war, indi- cates his ready versatility. He was frequently disingenuous, under- standing more than he expressed, but his blunt force compelled notice. Il8 THE RISE OF COTTON MILLS IN THE SOUTH [224 He enters this story when, at the invitation of leading men of city and State, he delivered an address in the senate chamber of the capitol in Atlanta in October, 1880, espe- cially to explain his proposal for the holding of a cotton ex- position. 120 Shortly before he had expressed himself as unable to recommend to the North investments in Southern cotton mills, but entered into no details. This had roused a storm of protest and discussion. 121 The incident showed, with many others, that Atkinson interested himself in the South a little too late to suit his purpose; the people had already formed a desire for cotton mills that was not easily dispelled. In the speech he had to advert to this, and in its printed form more references were included. He was frank to say, however, that if he were wrong, the proposed exhi- bition could have no more urgent reason than to demon- strate him mistaken. In judging his statements it must be remembered that as head of factory mutual fire insurance companies, he was constructively representing New England cotton manufac- turers. He said at Atlanta : " The true diversity of employ- ment which makes self-sustaining communities consists of occupations that do not appeal to the imagination like the great cotton factory; but the artisans . . . who work in iron or wood, the stove-maker and the like, the furniture- maker and the tinman, the house-wright, the wagon-builder, the blacksmith, and the whitesmith are the most valuable citizens. The hundred arts that require but little capital and support many men are the ones that, next to the farmer, form the bone and sinew of society. When these are es- tablished, the textile factory may well follow, but ought not to precede in any large degree." 122 Rather than cotton mills, the South should put up shops to make implements, and the manufacture of clothing would give work to women in their own homes. " On the other hand, the most impor- 120 Cf. Address at Atlanta, preface, p. 3. 121 Ibid., p. 27, and preface, p. 4. 122 Ibid., p. 28. If factories there must be, then shoe factories required only one-third as much capital per operative as cotton mills. 225]] THE RISE OF THE MILLS I 19 tant branch of the cotton manufacture — that of ginning, packing, and pressing cotton for the use of the factory — must continue to be done in the South, and every million dollars spent in the right manner in this department will ... do more to build up the cotton States than any million expended in cotton factories. It is in order that these op- portunities for immediate profit may be made apparent that the cotton exhibition should be held." 123 The cotton crop, he declared, was depreciated 1 10 per cent by careless han- dling in preparation for shipment to the spinner at the North or abroad. The cotton manufacture is a unit, be- ginning in the field and ending in the cloth room of the fac- tory, and " if the South desires to enter upon the safest, surest, and most profitable branch of cotton manufactur- ing" it should confine itself to the initial processes. 12 * He said that Southern spindles could not keep pace with Southern demand, and so Northern manufacturers did not fear Southern competition ; he did not see that this demand constituted an encouragement to establishment of Southern mills. He tried to scare the South by enlarging on the supremacy of the New England manufacture and contem- plated extensions that were imminent. 125 123 Ibid., preface, p. 5 ff. Interdependence was the foundation of union between the sections. " The railroad has almost eliminated distance ; and each section that serves the other best, serves itself also" (ibid., p. 8). 124 Ibid., appendix, p. 34 ff. 125 Ibid., preface, p. 7 ff. He naively said it would be greatly to the advantage of New England manufactures " to have a solid body of men in the South interested like themselves in promoting better ginning, baling, and handling cotton as it comes from the field " (ibid.). He made suggestive allusions to possible unsuitableness of Southern climate for cotton spinning (ibid., p. 4ff.). One would like to attribute to Atkinson nothing less than a national viewpoint in advocating Atlanta as the place for the exposition because it was in the cotton country where the preparation of the staple could best be urged; manufacturing machinery needed no encouragement (ibid., pp. Q-10). He tried to interest the South in the use of ensilage (ibid., p. 28), was working on employment of then wasted by- products of the cotton plant (Bradstreet's, quoted in Baltimore Journal of Commerce and Manufacturers' Record, June 3, 1882), carried attention to the soya bean (Southern Cotton Spinners' Assn., proceed. 7th Annual Convention, p. 102), said there would not be sufficient labor for cotton mills, and that people would prefer out- 120 THE RISE OF COTTON MILLS IN THE SOUTH [^226 When Edward Atkinson and a committee from the New England Cotton Manufacturers' Association visited the Atlanta Exposition the next year, an official statement of their impressions showed that they appreciated most those exhibits having to do with "ginning and preparing" the cotton, and declared the identity of interest between cotton grower and manufacturer were here demonstrated. 126 In an interview with press representatives he led away from manufacturing, and sought to arouse enthusiasm over the roller as opposed to the saw gin. 127 In a set address in the exposition building he reiterated these points, feared the real reason why cotton manufactures would not succeed in the South was that most enterprisers did not know how to work on a close margin, did not "know the difference be- tween a cent and a nickel." 128 He urged rather the build- ing of railroads, the opening of schools and savings banks, development of dairying, and even the importation of Pon- gee, Tussah or Cheefoo silk worms. 129 But the purpose of the South had solidified too much to be dissolved by such discouragement or neglect, and, as will presently be seen, an exhibit which was planned by Mr. Atkinson to be primarily agricultural, gave tremendous im- petus to the manufacturing of cotton. Commenting on his door employment anyway (Address at Atlanta, preface, p. 5 ff.), and that even coarse yarn mills involved risks the South could not take (ibid., pp. 27-28). He asserted that the approaching exposition " should be rather with a view to the development of tools and im- plements for the cultivation and for conversion of the plant into its' primary forms of fibre, seed, oil, oil-cake, paper stock, and wool, than with a view to the manufacture of cotton fabrics" (ibid., p. 22). Cf. Southern Cotton Spinner's Assn., proceed. 7th Annual Convention, p. 85 ff. The seal of the exposition bore a cotton boll but no spindle. 126 News and Courier, Charleston, Nov. 1, 1881. The advantage of sending cotton north in the raw state was implied in frequent assertions of Atkinson (cf. Address at Second Annual Fair of New England Manufacturers' and Mechanics' Institute, Boston, 1882, pp. 2, 27-28). 127 News and Courier, Charleston, Nov. 8, 1881. Cf. Tompkins, Storing and Manufacturing of Cotton, p. 14. 128 News and Courier, Charleston, Dec. 5, 1881. Cf. the writer's " Factors in Future of Cotton Manufacture in South," in Manufac- turers' Record, Baltimore, May 10, 1917. 129 News and Courier, Charleston, Dec. 5, 1881. 2273 THE RISE 0F THE MILLS I21 exposition speech, an editor said : " Mr. Atkinson is mis- leading only when invincible prejudice keeps him from see- ing clearly, and even Northern newspapers admit that he is wrong in his belief that cotton manufacturing, on a large scale, will not pay in the South." 130 H. P. Hammett had Observed a few months earlier : " It is said the South should plant and prepare the cotton for market, and increase its value by improved cleaning and ginning appliances (which in themselves are proper and commendable), and then send it to the North to be manufactured there, to be returned to us in goods. ... I do not impute any . . . selfish motives to the parties who have thus . . . given their advice, but ... I am of opinion that good earned dividends by South- ern mills are much more convincing arguments to stock- holders than fine spun theories. . . ." 131 130 News and Courier, Charleston, Dec. 5, 1881. 131 Ibid., Aug. 1, 1881. It is said that in connection with the found- ing of Hamraett's Piedmont Factory, Atkinson wrote a notice show- ing how cotton manufacturing in the South could never pay. This came under the eye of Hammett, who pinned to the clipping his annual balance sheet, showing a profit of 20 per cent, and sent them to Atkinson (W. J. Thackston, int., Greenville). Atkinson never did really give up his campaign. In the section on cotton manufac- tures of the United States census of manufactures of 1880, written by him, and transmitted not until 1883, he devoted 6 of the 16 pages to the preparation of the staple, inveighed against bad ginning and urged upon the South opportunities for improvement. Twenty years later he was still on the subject of ginning in talking to the Southern Cotton Spinners' convention, but the revival of one of his Atlanta ideas (Address, pp. 18-25), namely, the folding of sheep upon worn- out cotton uplands, met now with the retort : " Let Massachusetts successfully grow our ' fleecy staple ' in her New England meadows before she advises us to raise Northern sheep in a Southern cotton patch" (proceed. 7th Annual Convention, remarks of B. W. Hunt). Some Southern manufacturers remember well Atkinson's position. "Edward Atkinson?" rejoined one of these, "He was the man that didn't believe in Southern cotton mills. He was one of the most prominent authorities on cotton in his day. He made himself very obnoxious to our folks by the way he opposed cotton manufacturing in the South. He just took the wrong turn on it" (W. R. Odell, int., Concord). And another, who had heard the Bostonian in At- lanta : " Edward Atkinson tried to have an influence in deterring Southerners from founding cotton mills, but we had our own ideas. When he talked to a reporter here against Southern mills, I replied to him_ in the paper" (Charles Estes, int., Augusta). The very exposition building, which Atkinson suggested might be taken away in sections to be used for ginneries or oil mills, was used on the 122 THE RISE OF COTTON MILLS IN THE SOUTH ^228 The International Cotton Exposition, held in Atlanta in the closing months of 1881, occupies a significant place in the history of Southern cotton mills. It accomplished two things : first, it drew together the South's apostles of a new industrial order into confirmatory exchange of views and plans, and afforded concrete, tangible encouragement to al- ready forming aspirations; second, it opened the eyes of the North to the field of investment that lay in the South, breaking down intersectional economic and political bar- riers of prejudice. From commencement of practical or- ganization of the exposition in December, 1880, it was ap- parent, in prospectus and executive personnel, that it was to be a more comprehensive undertaking than Edward At- kinson had suggested. Having origin in his mind, it ex- panded and developed in the hands of others. New England cotton manufacturing machinery makers and mill engineers were included with Southern industrialists and publicists in choice of officials. Not only raising and preparing of raw cotton, but production of cotton goods, was to receive emphasis. The secretary said : " Machinery of all the classes demanded in cultivation . . . and ... in ginning, baling, packing, and compressing raw cotton, belongs to the first division of machinery exhibits. The machinery requi- site for manufacture of cotton, with the best form of mills, the most economical applications of power, and all the details of subsequent manufacture, constitute a great de- partment in which there is a world of interest." The ex- position would demonstrate generally that the South had a great future before it, and that, with assistance, it would become " prosperous in its own right through a liberal de- velopment of its own resources." 132 spot as a cotton factory. On the gratuitous advice offered to the South " by those interested in preventing manufacturing develop- ment," of which Atkinson's must serve here as typical, cf . Thompson, pp. 62-63. 132 John W. Ryckman, in author's preface of Atkinson's Address at Atlanta, p. 4. Another connected with the exposition gave as part of its purpose : " To exhibit to the Southern people and to visi- tors from America and Europe the different processes in the manu- facture of cotton from the boll to the complete fabric, and by the 229] THE RISE OF THE MILLS 12$ That South and North were both ripe for the undertak- ing is shown by the rapidity with which it was accomplished. The exposition was opened in less than a year after first mention of it, in less than six months after real steps began to be made toward it, and in just 108 days after actual work of erection was begun. 133 The Atlanta Exposition was not the inception of the in- dustrial idea in the South, but rather its manifestation. It augmented rather than initiated a purpose. Had the South not known its own mind already, Atkinson's attitude might easily have narrowed the exhibits and diminished their use- fulness. 134. The timeliness of the exposition being apparent all along, its influence in stimulating cotton manufacturing and all friction of competition ascertain the best methods and find the best machinery. . . . We people of the South should embrace every opportunity which . . . will bring among us intelligent and inter- ested observers of our industrial condition, resources and aptitudes. We have in the midst of us the raw material ... of a magnificent prosperity. We lack knowledge, population and capital. These may be slowly accumulated in the course of years, or they may be rap- idly by well directed efforts to obtain them from beyond our own borders. We advocate the latter plan" (News and Courier, Charles- ton, March 14, 1881). 133 News and Courier, Charleston, Dec. 5, 1881. An Atlanta cotton manufacturer headed the executive committee, a Vermont engineer was made chief of machinery, and agents made tours of investiga- tion through the North and Europe. Subscriptions came simulta- neously from North and South ; General Sherman started Northern subscriptions with $2000 (ibid., March 8, May 3, 1881). On the opening day, Daniel W. Voorhees, of Indiana, spoke against free trade (ibid., Oct. 6, 1881). 134 A correspondent in a new mill community wrote : " It is to be hoped the Atlanta Exposition will not take all the enthusiasm out of .our capitalists and enterprising men, but that it will only tend to a greater and more speedy development of our resources " (ibid., Oct. 21, 1881). "A good work has been done, the benefits of which will be felt in every part of the country. The New South takes a fresh start at the Atlanta Exposition" (ibid., Oct. 7, 1881). The secretary declared the exposition was pushed through hurriedly be- cause_ " a knowledge of the South's resources was demanded . . ." (Baltimore Journal of Commerce and Manufacturers' Record, June 24, 1882). Cf. ibid., Oct. 7, 1882. "The Atlanta Exposition . . . was the hopeful and conscious expression of the opening of a new era for Southern industry . . ." (see Clark, in South in Building of Nation, vol. vi, pp. 280-281). Visitors to the exposition "were con- vinced that 'an industrial revolution had actually been effected in the South . . .'" (see Hammond, pp. 328-329). Cf. Copeland, pp. 32-33 ; Goldsmith, pp. 4-5. 124 THE RISE OF COTTON MILLS IN THE SOUTH [23O industrial development was quickly evidenced. 135 The new statesmen of the South, industrially and not politically minded, found voice. 136 Hint9 and hopes became certain- ties. "When the Atlanta Exposition closed ... it began to be realized that the South was awakened to a new life^ . . . Intelligence was to take the place of ignorance in methods of cultivation; machinery was to take the place of hand labor ; manufacturing was to take the place of export- ing raw material and bringing back the manufactured ar- ticle. . . . Capital began to see the rich rewards waiting to be won, and prepared to occupy the vantage ground." 137 Many of the exhibits were sold during the exposition, and orders taken to the amount of $2,ooo,ooo. 138 When 135 A manufacturer remembers that operatives from his mill who visited the exposition brought back small-pox, four hundred cases resulting. But Atlanta spread other and more salutary infection as well. Another's dominant recollection is that " they had a great deal of eastern machinery there, with men sent along to operate it" (Charles Estes, int., Augusta). 136 Cf. letter of A. J. Russell in Baltimore Journal of Commerce and Manufacturers' Record, July 15, 1882. 137 " vv. B. C," in Baltimore Journal of Commerce and Manufac- turers' Record, Oct. 7, 1882. David R. Francis ("The Influence of Agricultural and Industrial Fairs and Expositions on the Economic Development of the South since 1865," in South in Building of Na- tion, vol. vi, p. 568 ff.), does not mention the Atlanta Exposition of 1881, and apparently is unacquainted with its meaning. He attributes to the New Orleans Exposition of 1884-1885 a significance that be- longs to the earlier effort. " Not all the books and papers and speeches that man can produce would do the South as much good in half a century as the single event of the Atlanta Exposition did last year. . . . The cotton spindles of the south will increase year by year until the river cities will resound with the music . . . and the old battle-fields are the scenes of a great industrial revival" (Boston Economist, quoted in Baltimore Journal of Commerce and Manufac- turers' Record, Sept. 30, 1882). The whole lesson of the exposition was expressed when " the governor of Georgia appeared on the grounds dressed in a comfortable suit of cottonade manufactured on the premises from cotton picked from the bolls the same day in sight of the spectators." Cf. Goldsmith, pp. 4-5, on this episode and the influence of the exposition generally; U. S. Census of Manu- factures, 1890, " Cotton Manufactures," by Edward Stanwood, pp. 28-29. 138 Baltimore Journal of Commerce and Manufacturers' Record, Sept. 23, 1882. In Atlanta itself in the year following the exposition two cotton mills began operations, one in the exposition building itself ; plow works were greatly enlarged ; a cotton seed cleaner com- pany increased output; bridge builders extended their business; a cotton compress was erected; a company to manufacture a cotton 23 I J THE RISE OF THE MILLS 125 the Atlanta Exposition closed, some of its exhibits were moved to Charleston and formed the nucleus of an indus- trial display there. 139 Other fairs were projected more widely than achieved, but the North Carolina industrial exhibit, at Raleigh, in 1884, carried on the Atlanta spirit and made it local to the State in a way that assisted cotton mill growth. Northern machinery manufactures sent equipment that was manned by North Carolina opera- tives. 140 An exposition unsuccessfully urged for Balti- more, to have been held the same year, borrowed incentive from benefits derived by the city of Atlanta; there was the idea of capturing leadership of an advance which had been born to the South in a more generous impulse. 141 William Gregg in 1845 had instanced for Charleston the appropriate lesson of the way in which leading propertyholders of New- buryport, Massachusetts, when the shipping of the place deserted in favor of Boston and the town was going to ruin, determined to make an effort to resuscitate its pros- perity by establishing cotton manufactures with steam power. " It acted like a charm. The three or four estab- lishments put in operation, have all done well and produced a new state of things." So it might be, he showed, not only with Charleston, but with Augusta, Columbia and planter commenced building; a cotton seed oil mill was erected and other enterprises went forward (ibid., Sept. 30, 1882). "In six months after the exhibition closed, $2,000,000 had been invested in manufacturing enterprises in that city of only 40,000 inhabitants, all of which was directly traceable to the exhibition " (ibid., Oct. 21, 1882). Cf. ibid., June 24, 1882. 139 Cf. News and Courier, Charleston, Jan. 1, and for month of February, 1882. 140 W. R. Odell, int., Concord. 141 " The rapid development of the South in all her material in- terests has been the wonder of the age, and yet the past is but the harbinger of the future. Baltimore now has the opportunity of placing herself at the head of this grand Southern movement, and thus so closely allying herself with the South as to be ever after- wards the recognized centre of the commercial and manufacturing affairs of that section. Will she do it? The answer must come from our business community and upon it will depend the future of Baltimore " (Baltimore Journal of Commerce and Manufacturers' Record, Sept. 23, 1882). Cf. ibid., June 10, 17, July 1, 22, Oct. 1, 14, Nov. 11, 1882; Feb. 1, 1883. 126 THE RISE OF COTTON MILLS IN THE SOUTH [[232 other points at the South. 142 More than any other man of his time, he understood the public benefits resulting from industry, especially cotton manufactures, and held these to constitute a prime reason for building mills. 143 An advo- cate of rural cooperative credit associations in the South, believed that prosperous men, though not themselves need- ing aid, would take hold of the scheme from "philanthropic motives which always animate the minds of a large propor- tion of the well-to-do citizens of any country, stimulating them to efforts in behalf of the community in which they live." 144 And one who knew the South said of Southerners 142 Domestic Industry, p. 30. He lamented that Charleston's large surplus of dormant wealth was not directed to internal improve- ments in South Carolina instead of seeking Wall Street (Speech on Blue Ridge Railroad, pp. 6-7, 29), and urged that limited liability- be granted to industrial corporations which might thus lay small investors under tribute for the building up of the State (Propriety of Granting Charters of Incorporation, pp. 4-11). 143 An earlier manifestation has been alluded to. " About 1833, following the agitation against the tariff, several companies for manufacturing cotton were organized from patriotic and political rather than from purely commercial motives" (see Clark, in South in Building of Nation, vol. v, pp. 321-322). 144 Hammond, pp. 203-204. " Probably no better field for the exercise of such motives could be found than among the large planters of the South. Long accustomed to leadership in all the political, business and social affairs of the community, imbued with a spirit of helpfulness which their control over the . . . earthly destiny of others taught them to exercise during slavery days, taught finally by their own discouragements during the years of reconstruc- tion how bitter is the curse of poverty, these men would not lack . . . the willingness to help their poorer neighbors along the road to . . . industrial independence." Murphy wrote that the Old South in the New South "chiefly . . . has maintained . . . the old sense of responsibility toward the unprivileged," and that it is this " quick sense of social obligation," this " local conscience," which has given " distinction and beauty to the allegiance between the aristocracy and the common people" (p. 16 ff.). "Cooperation ... is the very spirit of democracy — concern for the common good, not only feeling that I am my brother's keeper, but more — I am my brother's brother. We have at last awakened to the fact that the whole is greater than the part. Too often heretofore we have thought of a social class, a segment of interests. . . . But a better day is dawning when we are alike embracing in our affections the whole people, the lowly.no less than the lofty . . ." (S. C. Mitchell, in South Mobilizing for Social Service, pp. 50-51). Sir Horace Plunkett has recognized that the pioneers of England's industrial preeminence " have been often actuated as much by patriotic motives as by the desire for gain " (pp. 153-154)- 233j THE RISE 0F THE MILLS I2 7 that " they are not only demonstrative ; they really care for one another in most affectionate ways. Helpfulness is not an act of conscience : it is an impulse." 145 Understanding the straits of the South at the opening of the cotton mill era, the readiness of Southern men to realize and assume responsibility in public matters, and the spirit of social service which characterized the awakening to a program of " Real Reconstruction," one accepts as natural the fact that cotton manufactories were frequently moti- vated by the desire to help a community to its feet. Often this wish was joined, and very properly so, with usual com- mercial promptings, but sometimes it controlled alone. The organization of the Charleston Manufacturing Com- pany with a purpose to give work to poor people of the city will be spoken of presently; this company gives admirable illustration of conception of a cotton mill with a plan of general civic betterment. It typified the concern of Char- leston for the welfare of the whole State, a concern which, when finally manifest, answered to Gregg's utmost solici- tude. 146 A notice supplementing an advertisement of the Charleston Manufacturing Company at the time it was soliciting subscriptions concluded : " The advantages, direct and incidental, accruing to every citizen of Charleston from this industry about to be started in our city are so manifest that those who have inaugurated the enterprise have every reason to feel confident of a ready response to the call for capital and of abundant success." 147 145 Page, pp. 111-112. Cf. John Skelton Williams, The Billion Arrives, pp. 16-17; Manufacturers' Record, Baltimore, April 19, 1917, suggestion for non-interest-bearing bonds to meet war ex- penditure. 146 Such cities, in " the heroism with which they meet the daily and the extraordinary crises that time brings . . . leaven the nation of which they are a part " Hemphill, quoted in Kohn, Charleston : Condensation of Jubilee Industrial Edition of News and Courier, p. 15). Charleston invested in South Carolina cotton mills that surplus of bank capital which Gregg had seen going to other quar- ters, and was largely responsible for incitement to an industrial movement that witnessed the purchase of used machinery from mills at Newburyport, Mass., perhaps the very spindles that Gregg had pointed to as building up the New England city. 147 N ews and Courier, Charleston, Jan. 27, 1881 ; cf . ibid., Jan. 28, 128 THE RISE OF COTTON MILLS IN THE SOUTH [234 In 1868, Messrs. Sprague, Rhode Island manufacturers, undertook to develop the water power at Columbia, but failed; the property passed to the State Canal commission, and some Columbians contributed to the employment of an engineer to push the work. In February, 1880, the de- velopment was taken over by a firm of Providence engi- neers with a liberal State franchise, but this scheme also failed. When capitalists of Columbia bought the rights they set forth that " The work ... is one of great magni- tude and involves expenditure beyond the ability of this community. Nor is the interest merely local, but reaches out to every part of the State. We call, therefore, upon all ... to take part in this . . . central development. . . ," 148 The inception of the first mill at Gaffney has been men- tioned. This was distinctly a community enterprise, inspired and pushed through principally by one man with the object of the good of the little town. A Tennessee mountaineer, he had come to Gaffney working on the railroad, and stayed. There was little enough in the place to attract anyone, but he held high hopes for its development. His spirituel face with fine eyes, a dreaminess in his easy movements, a vigor that resides nowhere and everywhere in him, indicate how in spite of the most restricted resources, he possesses ca- pacity that built cotton mills out of hand. As a contractor he was working in a mill village near his town. " At Clif- ton I'd see the hands paid off, the amount of money they spent. I was convinced that stockholders wouldn't go into 1881. One of the chief movers in this mill, when it had failed and manufacturing was to be revived in the old plant by a new company, received from a fellow citizen a note thanking him " in the name of the public generally, for being instrumental ... in directing Mr. Montgomery's attention to the Mill. It means much for Charleston, and is only another of your constant and inspiring efforts for the public material advancement of our city. A hundred men like your- self would ' save the city.' " 148 News and Courier, Charleston, March 25, 1881. Cf. ibid., March 18, 1881 ; Blackman, p. 9. " The capital, because it was the capital, was laid in ashes by Sherman's troops. In the person of Columbia, all South Carolina was ravaged. . . . The city which suf- fered so sorely may reasonably expect the just assistance of the State . . ." (News and Courier, Jan. 25, 1882). 235] THE RISE 0F THE MILLS I2 9 such a large thing unless it paid them to. The first week- end I could get away, I went back to Gaffney and had a talk with some of the leading citizens, and tried to and finally did persuade them that to establish a mill here would build up the town and pay good dividends." He was not discour- aged that Gaffney had no water power like Clifton, and re- solved to make steam answer. The head of a little bank was elected president of the mill company, $50,000 was sub- scribed to stock and a charter applied for. The local banker visited a New York bank to ask for cooperation, but re- turned deeply discouraged. Others lost interest, but the original promoter would not. He sought to interest the president of the mill at Clifton in the Gaffney enterprise, and received confirmation of his beliefs that he could suc- ceed, but no active support. He next attacked the super- intendent of the mill at Clifton, sat with him many nights to persuade him to come to Gaffney with money and expe- rience and head the venture, and finally succeeded. 140 Notices of ceremonies held when a mill commenced opera- tion convey sometimes touchingly the pride of a community in the plant and the public character of the enterprise. Townspeople were like children with a very precious new toy ; newspapers described the arrangement of the machin- ery in the factory with the keenest interest. 150 The potency of associative effort, so marked in Southern cotton mill building in this period, overcame timidity that might have been prompted by a frank and individual can- vass of attending economic facilities. "The mill at Albe- marle, North Carolina, had its origin in the desire of the Efirds to have a mill at the town. Whether there existed real advantages or not, the people would make it appear that there were advantages for that particular location. Many mills were located at places where there was the spirit for them, rather than where they would be, economically, 149 L. Baker, int., Gaffney. 150 Cf. Chronicle and Constitutionalist, Augusta, Feb. 23, 1882; Chronicle, Augusta, Nov. 11, 1883. 9 130 THE RISE OF COTTON MILLS IN THE SOUTH £236 most successful." 151 A Marylander knowing the industry thoroughly said there was little community interest in his State, but that "down South the community interest was very strong. Every little town wanted a mill. If it couldn't get a big one, it would take a small one; if not a sheeting, then a spinning mill." 152 151 J. L. Hartsell, int., Concord. " But with any kind of manage- ment in the first years of their rise they made money, because there was no competition to require close figuring." Cf.Plunkett, p. 186. 152 Summerfield Baldwin, St., int., Baltimore. A mill investor of long experience believes that " usually community good played a larger part than monetary gain in the founding of a cotton mill" (Theodore Klutz, int., Salisbury). "One mill would encourage another, but the greatest factor in the growth of cotton mills in the South was community pride" (C. S. Morris, int., Salisbury, N. C, Sept. 1, 1916). The story of the building of a mill in South Carolina, told by a participant, is typical. " The town had a popu- lation of about 2500. It was stagnant, on no trunk line of railroad. Perhaps only one man in the place was worth as much as $100,000. There had been talk of building a mill; a retired business man, with no manufacturing experience, had tried and failed. Mr. X., living in Spartanburg, had been in charge of a small iron concern. He was an experienced cotton buyer and, though not wealthy, had great ability. He came to our town and announced to gentlemen there that if the local people would take $75,000 in stock he would get up the rest of the money for a 15,000-spindle mill. This offered a ray of hope. This was throwing out a rope to us. Many men saw a chance of getting a job out of it. But in the town and county generally a tremendous effort was made. The largest subscription was $2000. By raking with a fine-tooth comb they got the pledges for $75,000. The average man at first didn't give a thought to divi- dends. He was thinking of building up the town. I was running a country newspaper, and took $300 in stock because I thought it would give me increased circulation and job work. Every merchant thought he would get some trade by it. There were some who hadn't even an indirect motive, who just wanted to see the town grow." And again, another said : " Captain S. E. White was about as near the type of the old plantation head as South Carolina has had since the war. He had 4000 acres under cultivation, under his direct supervision. Fort Mill was just a hamlet in 1887. He wanted to see it become a town, so he started a cotton mill in it" (William Banks, int., Columbia, S. C, Jan. 2, 1917). "Colonel R. L. Mc- Caughril, a banker in Newberry, was the leading spirit in the town. He wanted to see the place grow, so he started a mill" (ibid.). The same was true of the Orr Mill at Anderson. Cf. Charlotte News, Textile Ed., 1917, respecting McAden Mills. " Town pride played an important role. The cotton mill was looked upon as a dynamo to effect changes in all departments of life in a community" (Ster- ling Graydon, int., Charlotte, N. C, Sept. 4, 1916). A commission merchant said : " As a rule the starters of mills got all classes of people to take stock. Usually eight or ten of the leading men of 237] THE RISE 0F THE MILLS I 3 I "A good deal of patriotism developed," said a not im- pressionable mill man, "and every town would vie with others in building mills. Some people took stock and sold it at a discount when it was apparent that the mill would be operated. They were willing to give so much to secure the mill for the town." 153 There is no stronger indication of the different spirit characterizing the building of mills in the eighties as contrasted with earlier periods than the fact that after 1880 many plants were located within the corporate limits of towns and cities. In the earlier enter- prises community spirit had not counted, and even the mills of the seventies, such as Piedmont, were taken to the water powers. 154 Eager discussion as to the comparative advan- tages of water and steam power marked this transition. From being an excuse for the town, the cotton mill came to be erected to invigorate a place that was languishing. It has been said that at least half the South Carolina mills were community enterprises. Later, when the commercial spirit was more pronounced, factories were built just out- side the corporation to escape town taxes. 155 In the case of some investors with whom assistance to the town was an indirect motive, the creation of a payroll, the town could be got to serve on the board — doctors, merchants, lawyers, planters. There would be one leading man who would take the thing up and push it through. He would come to see us. Everybody would want the mill" (Summerfield Baldwin, Sr., int., Baltimore). "What did the lawyer, doctor or fertilizer man know about running a mill? Yet it got to the point where, if he were prominent in the town and did not become a cotton mill president, he lost his social position. Of course, he couldn't do that" (W. J. Thackston, int.,' Greenville). George A. Gray, as a mill expert, organized and built some factories and managed them only until they were running smoothly, having been drawn in by an inex- perienced community (G. A. Gray, Jr., and J. Lander Gray, int., Gastonia, N. C, Sept. 14, 1916). 153 E. A. Smyth, int., Greenville, S. C, Sept. 12, 1916. 154 In 1880 Camperdown was the only factory in South Carolina within the corporate limits of a city (Blackman, p. 13). But this, like the Enterprise Factory at Augusta, was on a water power (Au- gusta Trade Review, Oct., 1884). 155 Cf. News and Courier, Charleston, Jan. 28, 1881. For the pros and cons of county versus town location, cf. Tompkins, Cotton Mill, Commercial Features, pp. 34-35. The building of cotton mills to help towns was entirely sincere; contrast Clark, in South in Build- ing of Nation, vol. vi, pp. 273-274. 132 THE RISE OF COTTON MILLS IN THE SOUTH [238 putting more money in circulation, was the causal stimulus. An editorial recommended the Charleston Manufacturing Company " as a means of enlarging 'the common income. . . . The employment given to hundreds of persons . . . will increase the value of house-property at once. They who earn nothing can't spend much. It was calculated last year that every $228 invested in cotton manufactures in South Carolina supported one person. ... It is evident that the building of half-a-dozen cotton factories would revolu- tionize Charleston. Two or three million dollars additional poured annually into the pockets of the shopkeepers . . . would make them think that the commercial millennium had come." 156 To give employment to the necessitous masses of poor whites, for the sake of the people themselves, was an object animating the minds of many mill builders. One does not have to go outside the ranks of cotton manufacturers to find denials of this, but a study of the facts shows how fre- quent and normal was the philanthropic incentive. 157 It 156 News and Courier, Charleston, Jan. 28, 1881. Leroy Springs wanted a payroll at Lancaster, so built a mill (William Banks, int., Columbia). "The thing that built most mills was the fact that the business men of the town wanted the increased payroll. There is an annual payroll of $2,000,000 in Greenville today, and it was this result to which the town looked in the establishment of mills " (W. J. Thackston, int., Greenville). A textile editor went so far as to say that " the principal cause of the cotton mills of the South was that the people had to be given something to do; it was desired to create a payroll" (David Clark, int., Charlotte). 157 The genuineness of altruism as a motive in the Cotton Mill Campaign is supported by observation of Southern character in other particulars and especially as operative in this period. "It is only when a people, united by a common suffering and bearing a common burden, are overheard in their converse with one another, it is only when the South speaks freely to the South, that one may catch that real spirit of noblesse, oblige which has so largely domi- nated the development of Southern life" (Murphy, p. 7). Answer- ing the statement that North Carolinians were very conservative, an acquainted speaker recalled how one enthusiastic New England woman induced the State to spend for an asylum for the insane at one time a larger sum than the whole annual resources of the Com- monwealth. " Our whole history is full of such incidents. Almost every noteworthy thing that we have done has been done in obe- dience to an impulse. Conservative? We are the most impulsive people imaginable" (Page, pp. 9-10). The South had recently gone 239^ THE RISE 0F THE MILLS J 33 will be noticed in another chapter how important with Gregg had been the plan to afford work to natives des- perately needing support. 158 The South might have learned its duty, too, from the kindly admonitions of a Rhode Islander, Senator James. He was thirty years in advance of the section when he wrote: But it is not only the benefit to be derived in a direct manner to the individual manufacturer, that holds out a strong inducement to the South to go largely into the business — nor yet, alone, the pros- pect of enriching a community as a body. Motives of philanthropy and humanity enter into the calculation, and these should not be dis- regarded. This is a subject on which, though it demands attention, we would speak with delicacy. It is not to be disguised . . . that a degree and extent of poverty and destitution exist in the southern states, among a certain class of people, almost unknown in the manu- facturing districts of the North. . . . The writer has no disposition to reproach the wealthy for the existence of such a state of things. He is well aware that it is the result of circumstances which have to them been unavoidable. But he cannot resist the conviction that, when a fitting opportunity presents itself to the wealthy men of the South to obviate these evils . . . and that even in a way to benefit themselves, they can hardly be held guiltless in case of refusal or neglect to apply the remedy. 159 Hammett, in his Piedmont mill of the seventies, very regardful of hi9 responsibility toward his unfortunate fel- lows, anticipated by a few years the action of many factory projectors. 100 Sentiment must be strong to find place in an through so much misery that the body politic was closely knit; cal- culations of commerce were for the time relaxed, and leaders were thinking for the whole people. Cf . Lewis G. Janes, " The Economic Value of Altruism," in Social Economist, July, 1893, p. 16. As to the effect of the Civil War in rousing the South to extraordinary measures, cf. Andrews, pp. 340-341. 158 To the stockholders of his Graniteville Mill he said : " We may really regard ourselves as the pioneers in developing the character of the poor people of South Carolina," and he called the factory village an asylum for widows and orphans and families brought to ruin (see Kohn, Cotton Mills of S. C, p. 21). 159 "I ._. .appeal to the planter of the South, as well as to every other capitalist. Let your attachment to your interest and the in- terests of the community, united with love for your species, combine to stimulate you to enter, with resolution, this field of enterprise . . ." (quoted in DeBow, vol. i, p. 241). 160 Samuel Stradley, int., Greenville, S. C, Sept. 12, 1916. It has been pertinently said of the years following 1880: "There was no thought ... in those times, with regard to who should work or how many hours they should work. The problem was not one of seeking 134 THE RI SE OF COTTON MILLS IN THE SOUTH [24O advertisement soliciting subscriptions to stock, yet the Charleston Manufacturing Company frankly said : " The necessity of establishing manufactures in our city, not only as a profitable means of utilizing capital, but more espe- cially for furnishing employment to many in our midst, has been long felt. To put this matter into practical operation, a few gentlemen applied to the last Legislature and ob- tained a most favorable charter. . . ," 161 A committee of the State Agricultural Society of Georgia recommended the Clement Attachment to planters with capital as " fur- nishing means of support to needy and worthy people, to wit, women and children principally," and as keeping at home money " to give comfort and support to the planting community." 162 No undertaking was born more emphatically in the im- pulse to furnish work than the Salisbury Cotton Mills. All the circumstances of the founding of this factory were or creating wealth ; it was essentially one of employment, of human welfare in the sense of providing instrumentalities by the use of which men, women and children could earn a livelihood. The exi- gent demand for the bare necessaries of life, which could be gotten in the cotton mills of that period only by the combined toil of the whole family, overshadowed all other considerations. Literally it was a question of 'bread and meat,' and the mills provided work for thousands who could not otherwise subsist" (R. Charlton Wright, in Columbia Record, Textile Ed., 1917). Cf. the writer's " End of Child Labor," in Survey, Aug. 23, 1919. " There was much \ in the humanitarian movement. People saw that the cotton mill man [ was a benefactor. Unlike the profit of the bank, his money went to feed the poor people. This contagion spread and had a great deal to do with the building of mills" (G. W. Ragan, int., Gastonia, N. C, Sept. 14, 1916). ~*~] — 'J 161 News and Courier, Charleston, Jan. 27, 1881. One inquiring among surviving incorporators of this enterprise is told today that " our idea in starting the company was that there were many people here who wanted work, needed it " (W. P. Carrington, int., Charles- ton, Dec. 27, 1916). 162 Observer, Raleigh, Aug. 24, 1880. "Aside from purely mer- cenary considerations," said an appeal to Charlestonians to take stock in mills at Columbia, "... is the incalculable benefit to be derived from the employment of thousands of unwilling idlers . . . in the State, the women and girls for whom it is so hard to find healthful and profitable work" (News and Courier, Charleston, April 13, 1881). It must be remembered that whites, particularly women, could not compete with negroes in certain occupations, and in " servile " ones would not. 24l] THE RISE OF THE MILLS 135 singularly in keeping with the philanthropic prompting. The town of Salisbury, North Carolina, in 1887 had done nothing to recover from the war. It was full of saloons, wretched, unkempt. It happened that an evangelistic cam- paign was conducted; Mr. Pearson, remembered as a lean, intense Tennesseean, preached powerfully. A tabernacle was erected for the meeting, which lasted a month and, being undenominational, drew from the whole town and countryside. The evangelist declared that the great moral- ity in Salisbury was to go to work, and that corruption, idleness and misery could not be dispelled until the poor people were given an opportunity to become productive. The establishment of a cotton mill would be the most Chris- tian act his hearers could perform. " He gave Salisbury a moral dredging which made the people feel their respon- sibilities as they had not before, and made them do some- thing for these folks. There had been little talk of manu- facturing before Pearson came ; there had been some tobacco factories in the town, but they had failed. The Salisbury Cotton Mills grew out of a moral movement to help the lower classes, largely inspired by this campaign. Without the moral issue, the financial interest would have come out in the long run, but the moral considerations brought the matter to a focus." 163 163 O. D. Davis, int., Salisbury, N. C, Sept. 1, 1916. Cf. Page, p. 12 ff. ; U. S. Census of Manufactures, 1880, " Factory System," by Carroll D. Wright, pp. 4-5. The spirit of that evangelistic campaign still rests upon those all along connected with the enterprise. Mr. Davis remarked the fact that three ministers of Salisbury* were prominently connected with the inception of the mill. One of them, Mr. Murdock, was its secretary and treasurer and later president. The first minute-book shows how closely connected were preacher and manufacturer, even in point of time. An account copied into it from the North Carolina Herald (the local paper) of Nov. 9, 1887, headed " The Cotton Factory," says : " Mr. Pearson, in a lecture yesterday afternoon, dwelt upon the fact that the great many poor . _. . people we have here ought to be and must be helped not by gifts and alms but by a chance to make an honest living. That a cotton factory would be the remedy. Pursuant to these urgent appeals a large number of citizens gathered this morning in the Warehouse and organized by calling upon Rev. F. J. Murdock to act as chairman. . . . Mr. Murdock, in strong, eloquent, and earnest words pointed out that it had almost become a necessity to build a I36 THE RISE OF COTTON MILLS IN THE SOUTH [242 Mr. Murdock seems to have been the chief local inspir- ator of the mill at Salisbury ; before the factory was built he had established a building and loan association. A very similar case is that of Dr. Jacobs at Clinton, South Caro- lina. He found that the sodden little town needed to have industry preached to it. He inspired a merchant to build a cotton mill, took the lead in urging improvements for the community, and succeeded in founding an orphanage, funds of which were invested in manufactories of Clinton. On the whole, North Carolina was probably later in re- sponding to the philanthropic impulse than South Carolina. The local Democratic press censured a North Carolina congressman, an Independent, in 1886 for a speech urging mills as means of employment of poor people, because this was opposed to the interest of the farmer. 164 Yet a factory was built in the suburbs of Raleigh the next year partly with this purpose. 165 As late as 1902 a representative manufacturer declared that although negro labor was feasible, abundant, and would be cheapest, the managements " have recognized the cotton mill here to help the poor whites, quoting the Hon. J. S. Hen- derson's words — that next to religion Salisbury needed a cotton fac- tory. Rev. J. Rumple, D.D., seconded Mr. Murdock's appeal. He said that he knew so well the appealing condition of the poor whites of our town and that a cotton factory would be a sufficient remedy. Mr. I. H. Faust urged three reasons for the building of a mill. 1. Increased general prosperity of the town. 2. Benevolence and char- ity in giving the poor a chance to earn a living. 3. Cotton mills pay a handsome interest to investors." Others spoke of the profits of all Southern mills, of the health of Salisbury as an asset, and " Maj. S. W. Cole, the veteran advocate of cotton mills, spoke earnestly and fervently in favor of the undertaking." A committee appointed to solicit subscriptions met the same afternoon. Subsequent items show that by Dec. 15 organization was complete, some $60,000 having been locally subscribed, and a successful manufacturer in Concord, nearby, who was consulted in the enterprise, being elected president. One director was a minister; the others were pillars in Salisbury churches. " The mill was religion-pervaded from the outset." It was decided at the start not to have a company store, thrift has been consistently encouraged in the operatives, the mill has never run at night (Theodore Klutz, int., Salisbury). Especially through Mr. Murdock's influence, several boys growing up in the mill have be- come ministers (Charlotte News, Textile Ed., 1917). 164 John Nichols, int., Raleigh. 165 A. A. Thompson, int., Raleigh. 243] THE RISE 0F THE MILL s 137 fact that the mill life is the only avenue open today to our poor whites, and we have with earnestness and practically without exception kept that avenue open to the white man alone " to provide an escape from competition with the blacks. 166 It has been seen that the spirit for manufacture in the South was born pretty much irrespective of the direction which activity was to take. Bearing this in mind, if one were asked what inspirited cotton mills, he would probably answer first, " Presence of the raw material." There is everything to commend this reply. Tn the beginning South- erners did not reason out all the implications of their thus setting up cotton factories in cotton fields. If success at- tended the pressing present, this was enough. Moreover, New Englanders, as noticed in the case of Edward Atkin- son, more able to calculate upon the future, sought often to discourage a 'movement which they realized portended danger for their section as the principal American seat of the industry, and in this way the outlook of the South was clouded. Ten years after the opening of the period, how- ever, a writer could put the matter plainly, justifying the South's best hopes and rebuking New England's dissimula- tion by saying: "The ultimate transfer of the cotton indus- try from New England to the South may be regarded as an inevitable consequence of industrial development, which should be neither feared nor prevented. . . . There is no more reason why cotton cloth should be manufactured in Lancashire than why cucumbers should be raised in Ice- land." 167 166 See testimony of Lewis W. Parker, Hearing before Committee of Judiciary, House of Representatives, April 29, 1002, p. 11 ff. This statement would bear some modification today. Perhaps at the out- set some saw in the cotton mills not just the means of immediate employment, but the first step toward a better grade of work. Until the present these have been disappointed ( W. W. Ball, int., Columbia, Jan. 1, 1917). These well-wishers of the operatives have not been willing to accept continued evidences of philanthropy in welfare work for the more wholesome self-help to be gained when Southern mill hands, like successive generations in New England, assisted by a greater diversity of industry in the section, reach out to more skilled employments. 167 Social Economist, May, 1891, p. 152 ff. On the purpose of I38 THE RISE OF COTTON MILLS IN THE SOUTH [244 From the outset, though, convinced of the strength of its position, the South put by hypocritical gratuities : " Sir, it matters not what anyone may say to the contrary, common sense tells us that other things — machinery, skilled labor, motive power, and facilities of shipment — being equal, a cotton factory in the midst of cotton fields must prove more profitable than the same concern a thousand miles from the base of supply could possibly be." 168 "Leave it to the North to make the finer, lighter and fancy goods," Ham- mett counselled. " Their manufacture will come South in due time if it should be desirable to make them. . . . We need have no fear of competition in making the heavy goods from the North. They will never build another mill there to make them." 169 English manufacturers to build mills at the South, cf. C. C. Baldwin, quoted in News and Courier, Charleston, July II, 1881. The South was not entirely without similar penetration much earlier. Of E. M. Holt, manufacturing in North Carolina long before the war, it is said that " To him it seemed a geographical and economical incon- sistency and perversity that this' staple should be carried thousands of miles from the place of its growth to be made into cloth, much of which was to be brought back ... to clothe the very people who had produced it; ... he foresaw that not Manchester, not New England, but the South was to control the cotton industry of the world" (Martin H. Holt, in Biographical Hist, of N. C, vol. vii, pp. 182-183). A New Englander said of the South, also before the war : " As respects all raw materials, especially that of a bulky char- acter, economy dictates that, all other things being equal, they should be wrought on the spot on which they are produced. . . . There may be some exceptions to this rule, but . . . there is none in favor of the transportation of cotton to a distant market" (Charles T. James, in DeBow, vol. 11, p. 236 ff.). Cf. Olmsted, pp. 165, 542-543. Atkin- son in 1880, though speaking especially for New England, really put the case for the South when he said that " the supremacy in the art of converting cotton into cloth must ultimately fall to that country or section which possesses . . . proximity to the source of raw mate- rial" (U. S. Census of Manufactures, Cotton Manufacture, p. 8). For clear statements in the South in 1880, cf. Blackman, p. 14, and prefactory leading article. 168 See Gannon, p. 6 ff. Later, Grady declared. " The industries of other sections — distant from the source of supply — may be based on artificial conditions that may in time be broken. But the industrial system of the South is built on a rock — and it cannot be shaken ! " (pp. 206-207). Cf. ibid., p. 80 ff. 169 Quoted in Manufacturers' Record, Baltimore, Feb. 1, 1883. " The water powers are located in the midst of the cotton fields, from which a large part of the cotton consumed may be purchased direct from the producer and delivered at the mills. ... A very 245] THE RISE 0F THE MILL s 139 Nor did some Northern papers at this time fail to recog- nize the superiority of Southern manufactories in posses- sion of the raw material. " They have the advantage of cotton location, and, when they have secured new and im- proved machinery, will do an unrivalled business. " 170 The pertinence of such recognition was admitted by New Eng- land manufacturers in deed if not in word. Their appeal for lower freight rates " on account of the growing opposition of Southern cotton mills . . . was a plea of weakness. . . . The manufacturers of New England would do well to heed the advice of the New York Times . . . and give up the attempt to compete with Southern mills on coarse goods." 171 Many factories were built right in the cotton fields, just material advantage is that it comes direct from the gins, is clean, has not been compressed for shipment . . . and as a consequence works here infinitely . '. . easier . . ." (ibid., quoted in News and Courier, Charleston, Aug. I, 1881). "Among the public enterprises which have been started in Memphis during the past twelve months none have attracted more . . . interest than the ' Pioneer Cotton Mill.' .... With the great staple at our doors it does seem strange that it should be sent to the Eastern States or to Europe to be manufactured into goods that will be sent back here for sale at a handsome profit" (Memphis Avalanche, quoted in Manufacturers' Record, Baltimore, Dec. 28, 1882). Cf. ibid., Dec. 14, 1882, March 8, 1883. 170 Manufacturer and Industrial Gazette, Springfield, Mass., quoted in News and Courier, Charleston, Feb. 3, 1881. " They can save freights, buy cheaper and hire cheaper labor. They save buyer's commission, and warehouse delivery and cartage, sampling, classing, pressing, shipping, marine risks, and freight and carriage to interior towns, which amounts in all to some seven dollars per bale. . . . This makes a tax of eighteen per cent which Fall River pays in competition with Columbus. ... As yet the South manufactures principally coarser goods . . . but the time is not far distant when it will come to make prints, cambrics, laces, and all the finer quali- ties of staple goods." Cf. Philadelphia Record, quoted in News and Observer, Raleigh, Dec. 16, 1880. By 1882 it was being said that Northern mills must make fabrics of higher grade or go out of existence. " Much invested capital will have to be sunk, much good machinery cast aside, and much acquired skill regarded as useless ; but there can be no wisdom in hesitating to make the sacrifice when the refusal to make it means ruin at any rate" (Textile Record, Philadelphia, quoted in Baltimore Journal of Commerce and Manu- facturers' Record, Oct. 28, 1882). Cf. Boston Commercial Bulletin, quoted in ibid., Sept. 23, 1882; April 5, 1883. 171 Manufacturers' Record, Baltimore, March 29, 1883. Cf. a ref- erence to a protest to the Massachusetts legislature against a 58-hour bill in 1890, in Social Economist, May, 1891, p. 159. I4O THE RISE OF COTTON MILLS IN THE SOUTH [[246 as saw mills are placed in the woods. The Woodlawn and Lawrence mills, at Lowell, North Carolina, even conducted their own cotton plantation. 172 Although a water power mill at Cedar Falls, in the same State, had the disadvantage that its product must be hauled twenty-seven miles to High Point, most of the raw cotton was bought loose from the field. 173 A cotton planter built a factory at Enterprise, Mis- sissippi, which took cotton loose from the gin. 174 Founders of the industry and others expressed the pre- eminence in the mind of mill builders of proximity to cot- ton. "There seemed nothing else in the South for manu- facturing to turn to but cotton." 175 " Their whole purpose and idea was to build mills right in the heart of the cotton fields." 176 " In establishing cotton mills the chief advan- tage, in the minds of Southern people, was proximity to the raw cotton." 177 172 Baltimore Journal of Commerce and Manufacturers' Record, June 3, 1882. Several mills owned cotton lands. 173 W. R. Odell, int., Concord. Tompkins built a plant at Edge- field, S. C, for which cotton was secured unpacked from the field (J. H. M. Beatty, int., Jan. 3, 1917, Columbia). Many mills are to be seen today standing in cotton fields (cf. Columbia Record, Tex- tile Ed., 1916. Cf. ibid., as to Lancaster Mill's). The Proximity mill, Greensboro, was named with reference to nearness to raw material (cf. James A. Greer, in Textile Manufacturer, Charlotte, Aug. 19, 1915). The treasurer of the company thinks proximity to cotton was the prime cause of the Southern industry (Bernard Cone, int., Greensboro, N. C, Aug. 30, 1916). Cf. Charlotte News, Textile Ed., 1917, advertisement recommending Monroe, North Carolina, as a location for mills because of excellent and abun- dant cotton of Union County; cf. advertisement of P. H. Hanes Knitting Co., in Every Week, Nov. 12, 1917, p. 15. 174 Mississippi Beacon, quoted in News and Courier, Charleston, June 18, 1881. 175 James W. Cannon, int., Concord, N. C, Jan. 6, 1917. 176 Tracy I. Hickman, int., Augusta, Ga., Dec. 27, 1916. 177 Theodore Klutz, int., Salisbury. " The whole development was the result of the desire of the people to use their raw product" (William Banks, int., Columbia). "They had in mind all over the South the fact that the cotton was on the ground " (James Simons, int., Charleston, S. C, Dec. 27, 1916). "There came a different viewpoint. The old South was done away with. The problem was to utilize the thing nearest at hand to support a large portion of our people" (Henry E. Fries, int., Winston-Salem). "Other things were side issues. Proximity to raw cotton was the great advantage, as it appeared to us " (A. B. Murray, int., Charleston, S. C, Dec. 28, 1916). Some helps to development through this proximity were not 2473 THE RISE 0F THE MILLS H 1 The causes of manufacturing development reviewed and others to be touched upon, sometimes exerted a secondary influence through example of factories already in opera- tion, or even of old mills which had gone out of existence. The stimulus lent by the older establishments, those founded before 1870, was largely through individuals or families, wa9 personal, not inspiring new erections at the hands of men not in some way connected with the original ventures ; on the other hand, a mill built after 1880 often had a social bearing, attracting to the industry enterprisers and commu- nities with no manufacturing tradition. Of course, there foreseen by the first mill builders. Actual spinning tests of the staple may be made, instead of relying upon conventional grading (cf. Tompkins, Cotton Mill Processes and Calculations, pp. 4-5)- Atkinson did not realize that in this way only mills in the fields could improve preparation of cotton for manufacture. Southern mills, moreover, may rely upon a reserve in the hands of farmers, and not stock up in the picking season as heavily as Northern fac- tories. A few smaller mills even buy cotton as they receive orders for goods (cf. Copeland, pp. 182-183). Nor did the founders guess that supposed benefits of contiguity to cotton would vanish and actually turn out as hindrances. Where mills have concentrated, local cotton does not satisfy the demand. The local price is some- times driven above that of spot in New York. Cotton brought from the Delta or other distant points bears a relatively or absolutely higher freight charge than staple shipped to New England or Liver- pool. Also, the product must be sent north to market and, in most cases, to be finished. Any saving in purchase of raw material locally, amounting hardly ever to more than half a cent a pound, is about counterbalanced by freight on goods. When Southern mills were few and small, presence of cotton was a real asset, and product was often sold locally. Unless all forecasts are futile, the present is a " period of transition " for the Southern mills which will give way to more widespread distribution of plants (overcoming the singular disadvantage of some factories, such as those at Gastonia which can use no local cotton for their manufacture of fine yarns), to finishing of product at the South and the development of a South- ern goods market, when old superiorities of location will reappear and prove greater than ever. (These points were substantiated by interviews with John W. Fries, Winston-Salem; George W. Wil- liams, Charleston; Charles Estes and Tracy I. Hickman, Augusta; J. B. Cleveland, Spartanburg; Benjamin Gossett, Anderson; Joseph H. Separk, Gastonia. For fuller discussions see Copeland, pp. 36- 37; Uttley, p. 39 ff.; Thompson, p. 271; the writer's Factors in Future of Cotton Manufacture in South, in Manufacturers' Record, Baltimore, May 10, 1917; an excellently detailed illustration of draw- backs in regard to freight charges is contained in the petition of certain up-country South Carolina mills to the State Railroad Com- mission, Feb. 24, 1903). 142 THE RISE OF COTTON MILLS IN THE SOUTH ^248 were exceptions in both cases. Graniteville, more than other ante-bellum manufactories, possessed! public signifi- cance; it is difficult to tell how far promoters of mills at Augusta and elsewhere knew Gregg or were trained in his factory, and how far they were inspired simply by the ex- ample of Graniteville. 178 The factory is said to have had a fifty-year record of dividends. 179 It is likely very true that its success had an influence in the Cotton Mill Campaign of the eighties through Dawson of The News and Courier, who frequently referred to it. 180 Other old factories furnished more exclusively personal incentive. George Makepeace founded little mills on Deep River in North Carolina. Others, such as the Fries family at Salem, learned from him. Ante-bellum manufacturing of the Fries' was the forerunner of their post-bellum activi- ties. The Pattersons at Roanoke Rapids were connected with the Fries family. The grandfather of a mill president of Raleigh bad been a stockholder in two small mills at Cedar Falls, and knew Makepeace. The pioneer cotton manufacturer of Durham had clerked in the store at Cedar Falls. 181 William Bates, who came from Slater's mill at Paw- tucket, Rhode Island, was important because he influenced his son-in-law, Hammett, as has been noticed earlier. Wil- liam Entwistle, an Englishman with textile training in Lan- cashire, worked in Lawrence, Massachusetts. He came South with the intention of farming, but entered Granite- 178 Cf. Clark, History of Manufactures, p. 553 ff. The influence of Graniteville has been discussed more fully in the first chapter. 179 Tracy I. Hickman, int., Augusta. 180 H. R. Buist, int., Charleston, S. C, Dec. 28, 1916. Local advo- cates of mills sometimes harked back to successes at Graniteville and Augusta (cf. Society Hill correspondence, News and Courier, Charleston, Feb. 23, 1881). Graniteville had personal ties with many later establishments. The grandfather of LeRoy Springs, pioneer manufacturer of Lancaster, was one of the organizers of Gregg's company (Columbia Record, Textile Ed., 1916). The projector of the Rock Hill Factory was the son of a Graniteville founder (Wil- liam Banks, int., Columbia). 181 A. A. Thompson, int., Raleigh. 249] THE RISE 0F THE MILLS *43 ville as a section hand in 1869, then was at Langley and removed to the Great Falls mill at Rockingham, North Carolina, to become overseer of weaving. Great Falls was itself ibuilt on the site of the much older Richmond Manu- facturing Company's factory. Mr. Entwistle has been re- sponsible for much mill building at Rockingham and has given technical advice to other projectors, such as Mr. Cooper at Henderson. The Leak family, owning mills at Rockingham, two generations ago had the Richmond Manu- facturing Company. 182 Coming to mills which were patterns to communities rather than individual enterprisers, it is clear that Ham- mett's Piedmont Factory, projected in 1873 but delayed in commencing operation until 1876, was "a crucial experi- ment " ; that in a real sense " the success of the mills of the South depended upon Piedmont, the initial business." 183 It may almost be said that Hammett belonged to the develop- ment of the eighties; he anticipated the South's duty and opportunity by seven years. His mill was so excellent and complete, he was so able an advocate of manufactures and his public attitude was so constructive that his venture was really "the kindergarten for the industry in the up-country for twenty years." 184 182 William Entwistle and T. C. Leak, int., Rockingham, N. C, Aug. 14, 1920. The Holt mills in Alamance represent distinctly a family development. Gray's apprenticeship served in the old " Pin- hook Factory " has been remarked. The industry at Columbus owes much to the fact that before and during the war the place was " a miniature Lowell" (Observer, Raleigh, Sept. 10, 1880). The Lawrence (1878) and enlarged Woodlawn (1880) mills, at Lowell, N. C, grew out of the original plant of the company built in 1851 (Baltimore Journal of Commerce and Manufacturers' Record, June 3, 1882). Clifton was descended from the older Bivingsville and Glendale factories (Blackman, pp. 10-11; William Banks, int., Columbia). 183 \y. J. Thackston, int., Greenville. 184 " Th e m iiis built in this locality about 1880-1885 were simply results of the great success made by the Piedmont Manufacturing Company. The projectors of these mills used no arguments differ- ent from those of H. P. Hammett " (James D. Hammett, int., Ander- son, S. C, Sept. 11, 1916). Pelzer was an outgrowth of Piedmont, its founder driving over to look at the water power after an annual meeting at Piedmont (W. J. Thackston, int., Greenville). Follow- ing Hammett, Charlestonians had built mills in the Piedmont dis- .. 144 THE RISE OF COTTON MILLS IN THE SOUTH [25O When mills were erected in numbers, experience in these was shared with intending projectors. "In the Trenton mill we made a big success. It got into the papers, and^I had letters from all over the country, even from Texas, inquiring about it." 185 It seems plain that " the success of the Salisbury mill built the Advance mill. A good many who had held back from the first venture went into the second." 186 Most extensions of plants were of course outgrowths of successful experience. 187 Depressed condition of agriculture during and preceding the early eighties was in a large way a cause of cotton manufacture. Unremunerative farming led to industry in two main ways: by putting those able to initiate enterprise on the search for new investments, and by throwing out of a livelihood those unable to make new opportunities for themselves. In North Carolina, a poor agricultural State anyway, the process was especially clear. Water powers were more profitable than land. 188 The same was true of the upper part of South Carolina. Before the war there was little fertilizer used, and this district could not grow cotton. " The State was forced to appropriate $5000 one year to enable Spartanburg County to meet expenses. There was simply not enough property in the county of value." 189 This agricultural poverty reflected itself in a trict. Explaining the causes back of the Charleston Manufacturing Company, one of its incorporators said : " We thought that if a mill could pay in the up-country, it would pay to build a mill in a large center like Charleston " (William M. Bird, int., Charleston, S. C, Dec. 28, 1916). And speaking of this enterprise, a local paper urged : " Let us realize that what is good for Charleston in this respect, is better for us" (Kershaw Gazette, quoted in News and Courier, Charleston, Jan. 31, 1881). Cf. ibid., Feb. 26, 1881, regard- ing mills at Augusta. 185 Q_ \v\ Ragan, int., Gastonia. 186 Theodore F. Klutz, int., Salisbury. " The Salisbury mill showed what could be done in the field" (O. D. Davis, int., Salisbury). Cf. Daily Constitution, Atlanta, Jan. 2, 1880, editorial "Atlanta's New Year." 187 Cf. Charlotte News, Textile Ed., 1917, respecting Erwin Cot- ton Mill Company. 188 John Nichols, int., Raleigh. 189 J. B. Cleveland, int., Spartanburg, Cf. Hammond, p. 80. 25l]] THE RISE OF THE MILLS 145 supply of surplus labor that had been of long standing. Low ebb of agriculture was inevitably expressed in low price of cotton, which directly and indirectly encouraged manufacture of the staple. Generally speaking, the number of mills erected has varied inversely with the price of the raw material. 190 Just before the war a bale of cotton was worth $40 to $50, and the cost of constructing an average spinning and weaving mill was $16 to $20 per spindle. With war, paper money and scarcity of cotton, the value of the bale went to $900, and soon afterwards mills were costing $30 to $40 per spindle. By 1880 cotton and mill construction had returned to the i860 levels. 191 With crops constantly larger, it was seen that the South had reached the maximum quantity of cotton that could be produced profitably until world demand increased, 192 and that American manufacturers needd to expand and extend their export trade. 193 " For a few years after the war, when the price of cotton was so high that anyone could live by a small amount of farming, the land was cultivated extensively; but when the cultivation reached its limit, and the price of cotton became lower, the farmers and home capitalists realized that the only way their condition could 190 " Low cotton meant an increase in the number of failed white farmers. This meant an enlarged labor supply. Low cotton also increased the feeling in the community that the town should be kept going by something else than bankrupt cotton farmers " (W. W. Ball, int.. Columbia, Jan. 3, 1917). Cf. Columbia Record, Textile Ed., 1916, regarding Oakland Mills. 191 U. S. Census of Manufactures, 1880, " Cotton Manufacture," by Edward Atkinson, p. 8. The average annual price for middling upland cotton at New York, gold value, was 30.76 cents in 1865- 1866, and fell, with irregular recoveries, to 11.24 in 1880-1881. Though bales were increasingly heavier, production of bales trebled in these years (cf. table from Bradstreet's, quoted in Baltimore Jour- nal of Commerce and Manufacturers' Record, Nov. 4, 1882). As to alleged serious turning to manufactures in the South consequent upon low prices of cotton from 1830-1844, see Brooks, pp. 148-149; on the increase of spindles in the country in the twenties, similarly caused, see Hammond, p. 246. 192 Baltimore Journal of Commerce and Manufacturers' Record, Nov. 4, 1882. 193 News and Courier, Charleston, Sept. 12, 1881. Cf. Observer, Raleigh, June 12, 24; Aug. 3, 14, 1880. 10 I46 THE RISE OF COTTON MILLS IN THE SOUTH ^252 be bettered was by manufacturing the raw product at home/' 184 Not only were cotton manufactures made a likely field of investment by low price of material through increased pro- duction, but mills rose with the wave of recuperation of business after the panic of 1873 and its following years of depression. Return to specie payments lent assurance, and the demand for cotton goods was brisk. The year 1880 opened very hopefully. 195 The testimony of the president of Graniteville was matched by that of South Carolina manufacturers generally : " We have . . . been running 194 See Tompkins, History of Mecklenburg, vol. i, pp. 181-182. Another writer " remembers seeing five bales of cotton bring the owner only $104. Then the cry went up, ' Take the mills to the cot- ton fields,' and the people from the farms flocked to tend the ma- chinery" (L. P. Hollis, in Columbia Record, Textile Ed., 1916). Cf. Brooks, p. 203 ff. An old ledger of the Sibley mill at Augusta con- tains memoranda of cotton bought at 4 cents a pound. For the benefits conferred on the cotton farmer, see an illustrative but not quite accurate statement in Tompkins, " Marketing Cotton," in Tex- tile World Record, Boston, Sept., 1908. Cf. Sioussat, p. 228. 195 p or the country it was said that " following the resumption of specie payments, which inspired confidence on all sides, and after the last of the United States called bonds matured . . . and when the out-turn of the harvest was pretty well ascertained, the whole scene changed : gold began to pour into the country, business in- creased with wonderful rapidity, prices of bonds, stocks and mer- chandise advanced by jumps, and the whole field of commercial and financial transactions was marked by a great rebound from former depression, which will be remembered ... as the great ' boom ' of the Fall of 1879. In 1877 the country appeared as an insolvent debtor . . . ; in October, 1879, it appeared as the same party with every matured obligation paid up in full, and with abun- dant capital in hand, rousing himself to engage in a new career of industrial prosperity" (Commercial and Financial Chronicle, Jan. 10, 1880). Cf. Baltimore Journal of Commerce and Manufacturers' Record, Sept. 9, 1882. For the South it was stated : " The year that is just finished will be to the present generation a red-letter one; for it brought to an end the long and weary period of enforced economy and restricted business that followed the panic of 1873, and put every branch of industry at work. Agriculture was encour- aged in the west and south . . . the factories received more orders than they could fill, the railroads were blocked with freight, the mines were pushed to a greater extent than ever, and all other in- terests were quickened towards the end of the old year in a way that was full of promise" (Daily Constitution, Atlanta, Jan. 7, 1880). Cf. Observer, Raleigh, Jan. 2, 8, 15, April 24; Daily Dispatch, Rich- mond, Jan. 1, 1880; for a similar statement for 1882, cf. Manufac- turers' Record, Baltimore, Dec. 7, 1882. 253] THE RISE 0F THE MILL s 147 since 1873 between two fires, but we seem to have emerged from that trouble now, and we are at present making hand- some profits. If this condition of affairs continues for five years ... we will make a heap of money. Everything has conspired during the last twelve months to help this coun- try." 196 It is scarcely necessary to say that expectation of profits stimulated the erection of mills. While always considered, the prospect of money gain in dividends was not always most important in the minds of factory builders. Some- times projectors were able to estimate from proven expe- rience of mills running in the South, but more often profits were argued from believed advantages of the section for textile manufacture. Most advocates shared their hopes openly with community or State; few followed a course of communicating a secret to hand-picked investors. Profits realized in these years will be discussed in a later chapter. Dividends of mills were regularly brought to public atten- tion and calculations were printed to show how any prop- erly managed mill could make money. 197 The demand for goods in 1880 allowed sale ahead at value; prices of product advanced faster in proportion than those of raw material; mills could not fill their orders ; some Southern factories ran day and night. All of this tended to draw the attention of the North and of the world to Southern mills, helped up their standards, enlarged their outlook, gave established and prospective plants a springboard for the great impending leap forward. 198 Charleston, the only lending community 196 See Blackman, pp. 4-5 ; cf. ibid., p. 10. Some foresaw New England seizing fine goods manufacture from England to protect itself against Southern coarse product, but ultimately surrendering the whole industry more and more to the factories in the fields (ibid., p. 14, and leading article). Hammett was resolutely hopeful when leaner times began to be feared (cf. Daily Constitution, At- lanta, quoted in Manufacturers' Record, Baltimore, Feb. 1, 1883). As to gain of American exports to China at the expense of English mills, cf. Observer, Raleigh, Feb. 14, June 19, July 25, 1880. 197 Cf. News and Courier, Charleston, Sept. 13, 1881 ; Observer, Raleigh, Aug. 26, 1880; Daily Constitution, Atlanta, March 18, 1880. 198 Cf. Baltimore Sun, Jan. 8, 20, 28; Observer, Raleigh, March 6, April 24, 1880; Blackman, p. 15. By the end of 1884 less favorable I48 THE RISE OF COTTON MILLS IN THE SOUTH [254 in South Carolina, putting money in mills at a distance, showed more investment primarily for profit than did local districts. In some instances, ten years and more after the cotton manufacturing development commenced, mills were estab- lished partly to take advantage of cheap labor. This motive of exploitation was very different from the earlier desire to give the people supporting employment. It has been pointed out that there is a distinction between arguments used in pro- moting factories and the factors which have contributed to the success of the industry. When commentators on the mills say that their rise has been chiefly due to inexpensive labor, it is usually meant that this has turned out to be their chief asset. 199 In estimating the influence of water powers in mill building it must be remembered that while repre- sentative plants were located on streams right at first, there came a time when communities without this facility wanted factories and utilized steam. So far as they go, statements explaining the causal character of water powers are proper. The industry at Augiista and Columbus prior to 1880 was attributable chiefly to falls in the Chattahoochee and Savan- nah rivers, and plants erected after this date owed much to the presence of this asset. 200 conditions were at hand, but the Southern industry had received its impetus by this time. 199 Cf. Copeland, pp. 143-144; Murphy, p. 103; and the writer's "End of Child Labor," in Survey, Aug. 23, 1919. Of course, pro- posals by Northerners to erect factories in the South considered from the outset the advantage in, not any advantage to, labor. Cf. Manufacturers' Record, Baltimore, Feb. 1, 1883; Baltimore Journal of Commerce and Manufacturers' Record, July 15, 1882. 200 Cf. ibid., Sept. 9, 1882, as to Columbus ; Manufacturers' Record, Baltimore, Jan. 4, 1883, as to Augusta. When it is said that " With- out the canal Columbia would have had no mills" (Washington Clark, int., Columbia, Jan. I, 1917), correction must be inserted that without the desire for mills there would have been no canal; it was constructed in the main after 1880. Communities wishing outside assistance frequently advertised their water powers (cf. News and Courier, Charleston, Aug. 17, 1880, assets of Oconee County). Com- paratively late in the development, as in the case of labor, exploita- tion of water powers came to a leading place. Cf. Charlotte News, Textile Ed., 1917, regarding Roanoke Rapids ; Columbia Record, ibid., 1916, regarding Ware Shoals. Speaking broadly, railroads have been responsible for the extension of the industry and the 255] THE RISE 0F THE MILL s 149 Like some other causes, purpose of promoters to provide themselves with salaries did not appear in the 'beginning. Later, the practice is said to have been common, applying particularly to extensions with accompanying salary in- creases, or to projection of plants in new communities by an established manufacturer who wished money to come prin- cipally from local investors. The man who subscribed heavily to make positions for himself and members of his family had little in common with the founders of the South- ern industry. 201 A few mills were started because of desire to use idle land and buildings. Commencement of manufacturing in thlTbuilding of the Atlanta Exposition is a case in point. 202 Mills were regularly erected to help stagnant towns; it was exceedingly rare that one was proposed to create a town or to benefit land speculation. 203 Exemption of factories or of new machinery from State or local taxation made more appeal to the investor as such than to promoters and shareholders participating in com- munity enterprises ; it was believed to encourage assistance from the North and counted with Southern founders who owned most of the stock in their ventures. 204 From time to time reference has been made to reported location of plants rather than for the inception of mills (cf. Co- lumbia Record, Textile Ed., 1916, regarding Glenn-Lowry mill). 201 A. N. Wood, Gaffney, Sept. 13 ; Clement F. Haynsworth, Greenville, Sept. 9, 1916; August Kohn, Columbia, S. C, Jan. 5, 1917, interviews. 202 See News and Courier, Charleston, Jan. 14, 1882. The Arista mill' at Winston-Salem put_ idle land in use (John W. Fries, int., Winston-Salem). Cf. Baltimore Journal of Commerce and Manu- facturers' Record, Nov. 11, 1882, as to a project at Gainesville, Ga. When the development was well begun, plants of various sorts were converted for cotton manufacture. 203 The case of the Region of the Savannah Colonization Assn. is noticed elsewhere. Bessemer City, North Carolina, was an in- stance (S. N. Boyce and J. Lee Robinson, int., Gastonia, N. C, Sept. 14, 1916). 204 Cf. quotation from Bradstreet's in Baltimore Journal of Com- merce and Manufacturers' Record, Nov. 4, 1882; Observer, Raleigh, Feb. 13, 1880; Clark, in South in Building of Nation, vol. vi, p. 282; Blackman, pp. 6-7 ; Manufacturers' Record, Baltimore, Dec. 7, 1882 ; Baltimore Sun, March 4, 1880; News and Observer, Raleigh, Nov, 2, 1880; Kohn, Cotton Mills of S. C, pp. 99, 101. 150 THE RISE OF COTTON MILLS IN THE SOUTH £256 intent of English enterprisers to exploit Southern cotton manufacturing facilities. An Englishman who, from being an operative in Lancashire, Massachusetts, and South Caro- lina, has become important in the Southern industry, said that while there has always been much talk of this, nothing ever resulted. 205 As will be seen in another chapter, North- ern participation was principally by commission and ma- chinery firms and through investment. Before the cotton mill era properly opened some Northern manufacturers came to the South, and after the movement had demon- strated its success New England companies opened branch plants. 206 It is said that in Lancashire machinery manufacturers, commission houses and supply men have established mills with speculative purpose. 207 Equipment firms may even teach operatives in English and Japanese mills to run the machinery. Dull times in the American textile machinery manufacture have prompted makers to encourage erection and enlargement of factories by several means. 208 It is doubtful whether their motive in this policy followed in the South has been in any large degree speculative. It was not such in the eighties ; their desire was to profit from sale of machinery, not from sale of stock taken in payment for machinery. They furnished a facility rather than supply- 205 William Entwistle, int., Rockingham. 206 George Putnam, a member of a commission firm in Boston, established Camperdown at Greenville in 1873; through its example this mill had some influence, and, with Batesville, taken over by Putnam in 1879, had only Northern capital (Mrs. M. P. Gridley, int., Greenville, S. C, Sept. 9, 1916). Converse came from New England to join the Confederate forces and was assigned to opera- tion of the Glendale Factory; after the war, he continued to manage the mill, and conceived the idea of the influential Clifton enterprise (J. A. Chapman, int., Spartanburg, S. C, Sept. 5, 1916). Makepeace and Entwistle are other cases in point (A. A. Thompson, int., Raleigh). It is proper, also, to consider the services of A. D. Lock- wood, mill engineer of Providence, who was employed by enter- prises at the opening of the period. Cf. Clark, in South in Building of Nation, vol. vi, pp. 264-265. 207 Copeland, pp. 317-318. a ° 8 J. L. Hartsell, int., Concord. 2573 THE RISE 0F THE MILLS I 5 I ing an impulse. 209 Tompkins, as mill engineer, as head of a repair and supply firm and as Southern agent of machin- ery manufacturers, was instrumental in building many fac- tories, but he was motivated by desire for legitimate profit and by public spirit. 210 Following the war much new machinery was installed in New England. Southern mills with more than a local market, many of them overworked during the war and run down during Reconstruction, had to reequip or build new plants. This circumstance assisted the spirit for cotton manufactures. 211 These, and others, were reasons why the industry came into being. In the subsequent chapters on Labor and on Capital it will be shown how the South carried out its pur- pose. The present pages deal with the actual rise of fac- tories and aim to exhibit attending public interest as it expressed itself in the " Cotton Mill Campaign." The movement, it has been seen, had a definite beginning about 1880. The whole South not joining in right at first, it is difficult to say when the "drive" ended. Certainly by 1895, if not earlier, it had been demonstrated that the in- dustry carried its own excuse for being, and nothing more than economic motives were necessary to its encourage- ment. 212 209 It is charged, however, that an industrial journal represented machinery manufacturers in more than simply an advertising capacity. 210 He would foe invited to speak to citizens of a town contemplat- ing erection of a mill, explaining the broad benefits the factory would bring them and imparting as much technical information as they needed for organization (Sterling Graydon, int., Charlotte). Cf. Tompkins, Cotton Mill, Commercial Features, p. 25 ff. ; Plan to Raise Capital, pp. 13-14. 211 Henry E. Fries, int., Winston-Salem. 212 Better argument than the first appearance of the term is the clear implication of the News and Courier that, economically, the Cotton Mill Campaign began with 1880. It was said that Hammett ranked as one of the pioneers in the Southern industry because his Piedmont Factory was built before the opening of the Cotton Mill Campaign, and, in seconding his authoritative judgment, the paper took satisfaction in the practical undertaking of a program which it had long urged, and exulted that " seen in the cold light of accom- plished facts, the enthusiasm of which some of our friends hav:e 152 THE RISE OF COTTON MILLS IN THE SOUTH [258 Hammett, in 1883, to allay discouragement that had arisen in some quarters, made an explanation that exhibits the Cotton Mill Campaign: "A state of things has devel- oped which many of us expected to see, and which was inevitable. Too many yarn mills have been built in the last two or three years all over the South from Virginia to Mexico, and as a consequence the market for coarse yarn is overstocked. . . . They were built for the most part by inexperienced 'men, taken from other pursuits, without any experience or knowledge of the business, badly built, the cheapest machinery put into them, with no scientific system for doing the work intended, many of them without suffi- cient capital to pay for them when they were completed." 213 It is plain here how suddenly, under what social pressure, the movement was born. " Once the opportunity had been presented to them the chance was eagerly seized, and all who were able to do so contributed to make the new enter- prise successful. The press urged it upon those who had capital to invest, hailed joyfully every manufacturing proj- ect, and made much of every successful establishment. . . . As is commonly the case with enterprises of this nature, it has been attended with not a little public excitement. . . ," 214 complained, as carrying us too far, has not taken us a hair's breadth beyond the confines of solid business truth" (Aug. 1, 1881). Cf. ibid., April 25, 1881. Something as to the closing date may be drawn from the fact that in 1886 South Carolina repealed an act exempting cotton mills from taxation (cf. Clark, in South in Building of Na- tion, vol. vi, p. 282). 213 " They made poor yarn, which they pledged for the money to operate them, which was of course sold to realize, for such prices as were offered, and when the yarn was thus slaughtered it made a price for them and others to sell by, and it is not strange that they made little money." Most of them made more, however, than Northern mills (quoted from Atlanta Constitution, in Manufac- turers' Record, Baltimore, Feb. 1, 1883). 214 " . . . more mills have been projected than have been built; more have been erected which their projectors would not have erected had they studied the matter carefully before entering upon the experiment. But the failures have been few, and upon the whole the return upon investment in Southern cotton mills has exceeded that upon factories in the North" (see U. S. Census of Manufac- tures, 1900, " Cotton Manufacture," by Edward Stanwood, pp. 28- 29). An instructive table shows that Southern spindles increased from 610,000 in 1880 to 1,756,000 in 1890, reached more than 2,000,000 259] THE RISE 0F THE MILL s 153 An impressive interpretation of the English industrial revolution has shown that while it began through invention, invention alone would have taken generations to establish the different regime. The philosophy of Adam Smith and the moral impulses imparted by the Wesleys and Hannah More joined with the work of Watt to speed the process. "It required all the forms — physical, mental, commercial, and philanthropical — working in separate yet convergent lines, to lay the foundation of an entirely new system of manufactures. . . ." 215 In the South all sorts of forces, imore directly and consciously applied than in the case of England, headed up in the Cotton Mill Campaign ; regret for the past, resolution for the future, expressed them- selves here. Economic inertia was overcome with moral incitement, 218 industrial activity was lent momentum by a " passion for rehabilitation " which made erection of cotton mills, as twenty years later of schools, "a form of civic piety." 217 Leaders were mindful of the psychological qual- by 1892 and more than 4,000,000 by 1900. From 1880 to 1883, 450,000 new spindles were put into operation. Taking 10,000 for the average- sized mill, this means that three years saw 45 factories opened (ibid.). Cf. Baltimore Journal of Commerce and Manufacturers' Record, June 10, 1882, introduction to column headed " Manufac- turing." On Aug. 26/ eight items out of thirty-six dealing with manufactures were about cottori mills ; this was typical. Cf . ibid., Sept. 2, 9, 1882; Manufacturers' Record, Baltimore, Jan. II, Feb. 8, 1883 ; News and Courier, Charleston, July 30, 1881 ; Augusta Trade Review, Oct., 1884; Thompson, p. 73. "The South burst into the development; mills grew up like mushrooms" (Summerfield Baldwin, Jr., int., Baltimore, Md., Jan. 8, 1917). 215 See U. S. Census of Manufactures, 1880, Factory System of U. S., pp. 4-5. 216 Cf . Ingle, pp. 72-73. 217 Cf. Murphy, pp. 17-18. The volitional quality of the campaign appears in contemporary references to it as an " experiment." Cf. Plunkett, p. 170. Industrial advantage, arguing from the past, seemed to be on the side of water power; the wish was sometimes father to the thought in reasonings for steam power to be used at towns not on streams but which wanted mills. Cf. News and Courier, Charleston, March 26, April 25, 29, 1881. In many ways the Cotton Mill Campaign was a romantic movement, resulting in spindles instead of sonnets. There had been intense public interest in the Pacific railway, stretching across a desert to guarantee the Union's integrity (cf. Dunning, pp. 144-145). The South felt a homogeneity in making cotton mills rise from an industrial wil- derness. 154 THE RISE 0F COTTON MILLS IN THE SOUTH \j260 ity of the movement and were jealous that it should have no backsets. "The State cannot afford a single failure in her cotton mill campaign . . . ," 218 said one, and another: "A few disasters amongst new mills would be a calamity, the extent and effect of which it would be difficult to estimate or realize, for while one successful mill inspires confidence, the failure of one to succeed would have directly the oppo- site effect. The people should not allow themselves to be tarried into it too rapidly by popular enthusiasm, which now prevails to some extent throughout the South. . . ." 219 Few episodes are more illustrative of the wholehearted- ness and wisdom with which the South entered upon the Cotton Mill Campaign than that of the Clement Attach- ment. This was a device that combined ginning and spin- ning in one process ; it was small, cheap, and made a limited amount of yarn. Recommended for the use of planters, its employment would represent the first step from agricul- ture into industry. When Southerners were beginning to think of cotton manufacturing there was eager, widespread inquiry as to this equipment, and it was put into operation in some places. But it was not tarried over long — it was recognized as a makeshift, a partial solution which did not satisfy the purpose for a real industrial development. 220 The spirit of the movement for factories may best be 218 " Enquirer," in News and Courier, Charleston, April 29, 1881. 219 Hammett, in ibid., Aug. 1, 1881. 220 An enthusiastic forecast missed fire in asserting " we shall have in a half century some scribbling journalist of the future writ- ing the gossips of the invention of the Clement Attachment — which will by that time have worked greater revolutions in the South than the cotton gin has done in the past half century!" (Daily Constitu- tion, Atlanta, Feb. 6, 1880). Cf. ibid., Jan. 2, Feb. 20, 1880. Black- man solicited many opinions about it, and received generally unfa- vorable replies ; cf. especially pp. 17-18, showing to what pains enter- prisers from all parts of the South went to examine the machine ; cf. News and Courier, Charleston, Feb. 26, May 26, 1881 ; Observer, Raleigh, Jan. 31, 1880. Nor was the South, when the Cotton Mill Campaign began to gather momentum, greatly regardful of outside comment; answerable to the faith that was in them, papers printed onlookers' discouraging and heartening references with like com- posure. Cf. letter of Robert P. Parker to New York Sun, quoted in Daily Constitution, Atlanta, Feb. 13, 1880, and quotation of Detroit Free Press in Observer, Raleigh, Aug. 31, 1880. 26 1 J THE RISE OF THE MILLS 155 caught in newspaper items. These appeared constantly and in numbers, in county and city papers, and there was a lively exchange of such information between publications. Any news bearing upon industry, particularly cotton manu- facture, was put to service. The following is a character- istic heading : " The Straws that Show ! Indications of the Way the Wind is Blowing. The Latest Movements in the Cotton-Mill Campaign." And there follow notices of the receipt of machinery by Clifton /mill and praise from Bos- ton of the efficiency and profitableness of factories at Co- lumbus. 221 Correspondence from a little place since become a manufacturing point of consequence gave a typical in- stance : " In conclusion let me say a few words in regard to the ' Pet ' of the town, the Rock Hill Cotton Factory. This factory is owned and controlled by the citizens of the town (except $15,000 in stock owned in Charleston). It has a capital of $100,000, has over 6,000 spindles with 1,500 more to be added in a few days. The best evidence of its success is that not one dollar of its stock can be bought. It is the intention of the company ... to run the factory day and night ... to keep up with its orders." 222 It was reported that "strenuous efforts are being made in Greensboro to establish a cotton factory in that city." 223 In an article on railroads occurred this paragraph : " It is rumored that the Columbia and Greenville railroad car shops at Helena will be removed to Columbia. ... In case the removal is made 221 News and Courier, Charleston, March 22, 1881. 222 News and Courier, Charleston, Jan. 12, 1882. News of mills from a distance, too, was frequent ; it was noticed that enterprises at Wesson, Miss., were paying handsomely, that a mill building was constructing at Natchez, that companies were organizing at Vicksburg and New Orleans ; when a mill at Nashville declared a 14 per cent dividend another was built; mills at Pulaski, Tenn., were anxious to double their capacity; $50,000 was subscribed for a plant at Jack- son, Tenn.; Dallas was starting a $200,000 factory and Sherman wanted a $75,000 mill (ibid., Aug. 12, 1881). 223 Winston Leader, quoted in Observer, Raleigh, June 17, 1880. " The Statesville Landmark, with its characteristic level-headedness, calls for the building of manufactures. With this would come com- mercial strength for our beloved South" (News and Observer, Raleigh, Dec. 12, 1880) I56 THE RISE OF COTTON MILLS IN THE SOUTH [262 the Newberry News suggests that the buildings at Helena might be easily converted into a cotton factory." 224 It was reported that " the ' Cotton Mill Campaign ' is progressing satisfactorily in Yorkville. We heard an old citizen remark some days ago that he had never seen the town so thoroughly aroused and united. . . . Yorkville to all appearances is moving forward with a determined pur- pose to put into successful operation a cotton mill. . . . The shares have been placed at $500 each, and up to this writing about $25,000 have been subscribed. I would state that this amount has been raised within the limits of the town." 225 It was advertised that " We will give to a Cotton Manufacturing Company that will organize and locate at Landsford, S. C, with a capital of $300,000, a site, 20 acres of land and 300 horse water power." 226 There were many items like the following : " The project for establishing a manufactory for cotton near Walhalla is being mooted. An informal meeting of some of the citizens of that place was held last week with this view and stock to the amount of nearly $10,000 was subscribed by the few present. It is believed strongly that as much as $25,000 will be sub- scribed in that neighborhood, and if the people of the county will join in the enterprise as much as $50,000 might be made available." 227 Town pride expressed itself in keen rivalry. " One little place would have a mill, and its neighbors would say : ' Here, 224 News and Courier, Charleston, March 22, 1881. 225 News and Courier, Charleston, March 25, 1881. " The signers to the prospectus of the mill are among the most reliable and respon- sible men in York County" (ibid., March 31, 1881). 226 News and Courier, Charleston, Feb. 23, 1881. " One gentleman at Griffin, Ga., offers to subscribe one-fourth the amount necessary to build a cotton factory" (ibid., March 25, 1881). 227 News and Courier, Charleston, Feb. 26, 1881, Cf. ibid., Jan. 9, 1882, as to Fort Mill. From Marion came this notice : " Our wants: A bank, an academy, a cotton factory, a comfortable room for passengers at the depot, an iron foundry . . ." (ibid., Feb. 22, 1881). "There is not a cotton factory at Raleigh, but there are not less than five large planing mills, two foundries, two boiler fac- tories . . . ," and newspapers and schools are mentioned (ibid., Jan. 26, 1881). Cf. as to Henderson, Baltimore Journal of Commerce and Manufacturers' Record, Oct. 14, 1882. 263] THE RISE OF THE MILLS 157 we can't let that town get ahead of us. We must start a cotton mill.'" 228 "If Belton got a mill, Williamston would want one. The townspeople would go to their leading citi- zen. It made no difference what a man was, so long as he was the leading citizen he had to become a mill presi- dent." 229 It has been said that the Charleston Manufacturing Com- pany, in all but its ill success, was the type enterprise of the Cotton Mill Campaign. It was peculiarly the child of the slogan, " Bring the Mills to the Cotton." 230 Though never really prospering itself, this factory had much to do with encouraging others, not least because it showed that the city practiced what it preached. 231 In Charleston every detail 228 Henry E. Litchford, int., Richmond, Va., Aug. 29, 1916. 229 Benjamin Gossett, int., Anderson, S. C, Sept. 11, 1916. A pro- moter, by visiting other mills, assured himself of the profitableness of an enterprise in his town: "Will a mill pay in Sumter? Why not? Every mill I visited had to pay $2 per cord for wood — it will cost less here in Sumter. . . . Every one of the mills received their cotton in bales ... at a loss of $1.90 to $2 per bale on bagging and ties. A factory in Sumter can use at least one-third of cotton with- out being packed . . ." (quoted from Sumter Southron, in News and Courier, Charleston, March 31, 1881). Many out of the way places came into notice through erection of cotton mills that would never otherwise have been heard of ; ventures in every part of the South, small and large, visionary or likely to mature, were not only chronicled, but were watched in their development from week to week. Interesting references, similar to those given already, may be found in News and Courier, Charleston, Jan. 4, 6, 21, 26, Feb. 3, 24, 26, March 23, April 6, May 21, Sept. 1, Oct. 21, 1881 ; Deutsche Zeitung, Charleston, Feb. 28, 1881 ; Baltimore Journal of Commerce and Manufacturers' Record, June 3, Aug. 26, Sept. 2, 23, 30, Nov. 18, 1882; Observer, Raleigh, Jan. 2, Feb. 20, 1880; News and Ob- server, Raleigh, March 18, Sept. 15, 18, Oct. 12, Dec. 24, 1800; Balti- more Sun, Jan. 20, 1880. 230 Little memorandum books informally kept by officers of the company covering organization, building and operation, show with what inexperience and yet with what genuinely affectionate solicitude this project was undertaken and followed through the seven years of its luckless career. A flyleaf gives : " Facts & Figures relating to the Charleston Mfg. Co. Born March, 1881 ; died Feby., 1888, leav- ing a large circle of disconsolate stockholders to mourn their loss. ' Requiescat in Pace,' " and there is the significant addition : " ' Bring the Mills to the Cotton.' — News and Courier" (Punctuation is the writer's). Dawson, editor of the paper, was one of the incorporators. 231 " Charleston is in a fair way to have two large cotton factories in a short while. . . . Camden is preparing for a cotton factory. Hodges ... is preparing for a cotton factory. Rock Hill has a I58 THE RISE OF COTTON MILLS IN THE SOUTH [264 of the taking of subscriptions and of erection of the plant was watched with the most absorbed interest. 232 The speed with which companies were organized and plants erected was significant of impatience to be at the task that invited. The company that erected the Huguenot Mill at Greenville formed February 10, 1881 ; a charter was ob- tained March 13 ; a lot was bought in the heart of the city and the first brick was laid March 23, the last June 2 ; by July 22 the machinery was in place and the mill was weav- ing cloth. 233 At the same time that new enterprises outright were being undertaken, old mills were being greatly enlarged or cotton factory. Greenville has several cotton factories. Newberry, the best location for a factory in the State, and the place most need- ing one, is not preparing for a cotton factory, and there is no pres- ent likelihood that she ever will. . . . There are numbers of people ready to aid in the enterprise . . . but there is nobody to take the lead" (Newberry Herald, quoted in News and Courier, Charleston, Feb. 8, 1881). It was not long before a citizen of Newberry did take the lead in erecting a cotton mill. " Why does not Fairfield make the experiment? It is said that fifteen thousand dollars will set in motion over five hundred spindles, and continual additions can be made. . . . The way to begin the new era is to erect a small factory in every county, and then to improve as facilities increase. Imagine Fairfield converting her eighteen or twenty thousand bales of cotton into yarn or cloth each year, and realizing a double price. If we can do no better let us spin a hundred bales at first. . . . Shall the effort be made, or shall other counties, once far behind us in wealth, take the lead and rapidly outstrip us?" (Winnsboro News, quoted in ibid.). The Barnwell Sentinel approved Charleston's course, and the Keowee Courier said Charleston had set the entire state an example (ibid.). 232 Cf. News and Courier, Charleston, Feb. 1, March 16, 28, April 9, July 6, Sept. 2, 1881 ; Jan. 14, 1882; Deutsche Zeitung, Charleston, March 21, 1881. At the same time a movement among German citi- zens of Charleston to establish a cotton mill with $100,000 capitali- zation got as far as application for a charter, but apparently no farther. Cf. News and Courier, Charleston, Jan. 27, March 30, May 4, 23, 1881, Deutsche Zeitung, Charleston, March 31, April 21, 1881. 233 Baltimore Journal of Commerce and Manufacturers' Record, Oct. 28, 1882. " Inside of four months from the commencement of the building, the mill was in operation and the capital invested yield- ing returns to its owners." A mill at Rome, Ga., the cornerstone of which was laid in June, was to be in operation in November (ibid., June 17, 1882). From the organization of Pelzer to com- pletion of the initial plant, including development of the water power for two later factories, required fourteen months (E. A. Smyth, int., Greenville, Sept. 12, 1916). Cf. News and Courier, Charleston, March 25, May 18, Sept. 10, 1881. 265] THE RISE OF THE MILLS 159 equipped with new machinery, plants were changing hands, those that chanced to burn were promptly rebuilt, factory projects that had lapsed were revived and pushed to com- pletion, buildings were converted from other uses to be cotton manufactories, places which had previously had mills reestablished them. Low prices brought by some factories early in 1880 contrasted with the profitableness of the in- dustry a few months later and indicate how suddenly cotton manufacturing burst upon the South ; small ventures which had had a chequered career, doing a small business and fre- quently failing, were taken by progressive managements that made them over and put new life into them. 234 234 Cf. Daily Constitution, Atlanta, Jan. 20, Feb. 29, 1880; Balti- more Journal of Commerce and Manufacturers' Record, July 15, Sept. 16, 1882; News and Courier, Charleston, Feb. 18, March 4, Aug. 19, Dec. 14, 1881 ; Augusta Trade Review, Oct., 1884 ; Kohn and Berry, Descriptive Sketch of Orangeburg, 1888, p. 12. CHAPTER III The Labor Factor The story of the rise of cotton mills in the South is a human story. Loyalty, love, purpose, charity, hope and faith are so intertwined with the specifically economic motive as to be inseparable from it. This is true of the narrative in all of its aspects. England may be said to have launched upon her Industrial Revolution unawares. " With the South the movement was conscious, distinctly marked in its commencement in the minds and hearts of the people!' In Britain the human problems came ~asT a consequence of the development; in the South they emerged with it and remained, for a long period at least, coeval with the in- dustrial advance. i! In this view, one would naturally expect the business im- pulse to be less dominant in the labor factor than in other particulars, but it is singularly characteristic of the incep- tion of the Southern cotton mills that other phases of the history, as for example the activities of entrepreneurs and the securing of capital, were as much bound up with the essential aspirations of the section as was the participation of men, women and children as operatives. Even machin- ery was wrapped with idealism and devotion. As the in- dustry has succeeded, with the passing of years there has been a separation of the economic and humanistic elements so intermixed at its beginning ; the opaque solution has been clarified by precipitation. Forces that were unified at the outset have developed contrary directions and have shown unequal power. The story of the workpeople has become less and less the story of the employers. Just as the erection of plants, once the object of close concern on the part of a whole commu- nity, has changed to a technical problem, and just as the monetary operations of the companies, forty years ago part 1 60 267] THE LABOR FACTOR l6l and parcel of the public life, have narrowed to their purely financial qualities, so divergent interests of capital and labor have emerged. In a region as newly industrial as the South, this has brought questions broadly and acutely social. In this study of the infancy of the manufacture, it is not at- tempted, except sketchily, to trace the lines of later develop- ment. The part played by labor in the rise of the mills cannot be understood unless it is recognized that the white popu- lation of the South is homogeneous and has always been so. There is no distinction in blood between employers and em- ployees. The inauguration of the industry, in point of capital and labor alike, took place within the Southern fam- ily. It made for an intimacy which at first rendered impos- sible and which continues to retard division between fac- tory owners and workers according to economic interest. The settlers of the South were of the same strains and pos- sessed the same characteristics. For an initial period' they moved along the same occupational lines. The invention of the cotton gin placed slavery in the ascendant. Cotton cul- tivation became dominant. The healthy industrial impulse which had shown itself gave way before agriculture. The gin, slavery and cotton formed the wedge that pried a uni- fied population apart. Landowners stood separated from the propertyless; as industry could not compete with agri- culture, so those without farming land could not compete with slave labor. The " poor whites " were dispossessed, not only of pro- gressive occupation, but of participation in the larger life of the section. From the time that cotton began to control until after the period of Reconstruction, these people lapsed into the background. 1 1 Cf . Tompkins, in South in Building of Nation, vol. vi, p. 58. " There is no difference in blood or heritage between them [the operatives] and the mill managements. . . ." It is interesting to see how a writer in 1809, regretting the exclusion of propertyless whites through the cultivation of indigo and rice, welcomed the new cotton farming as bringing these people back to economic partici- pation, little knowing how cotton itself would soon work their vaster 1 62 THE RISE OF COTTON MILLS IN THE SOUTH [268 When the "poor whites" entered the mills, they reen- tered the life of the South. As cotton culture had blocked progress, so cotton mills, while not dispelling the certainty of painful readjustments, opened the way. to a rational eco- nomic future. The settlers of the South were mainly English, German, Swiss, French Huguenot and Scotch^-Irish. They were able pioneers — hardy, industrious, independent, self-sufficient. They desired to have their own religions and to maintain their political and economic freedom. Whether from the Barbadoes, from New England, Pennsylvania, Virginia; whether Moravians setting up their churches and indus- tries; whether Highlanders loyal to the Stuarts and fleeing Scotland by shiploads after the battle of Culloden, they blended to make a stock which has no superior. 2 The term "poor white" is not easily defined, although every Southerner knows pretty accurately what it means. Writers, some through carelessness and others after better ruin : " By the introduction of the new staple the poor became of value, for they generally were or at least might be elevated to this middle grade of society. Land suitable for cotton was easily at- tained. . . . The culture of it might be carried on profitably by indi- viduals or white families without slaves, and afforded employment for children whose labor was of little or no account on rice or indigo plantations. . . . The poor having the means of acquiring property without the degradation of working with slaves, had new and strong incitements to industry. From the acquisition of property the transi- tion was easy to that decent pride of character which secures from low vice, and stimulates to seek distinction by deserving it. . . . In estimating the value of cotton, its capacity to incite industry among the lower classes of people, and to fill the country with an independ- ent industrious yeomanry, is of high importance. It has had a large share in moralizing the poor white people of the country'' (Ramsay, History of South Carolina, quoted in Scherer, pp. 170- 171). As it turned out, cotton in another phase, in manufacture, as William Gregg observed nearly fifty years later, was the means of " developing the character of the poor people of South Carolina." The mills, however, have dangers of being harmful in their evolu- tion as they were helpful in their inception, if they are allowed to be an economic pressure instead of stimulus. 2 Material as to the blood-strains in the Southern white popula- tion is plentiful. The following references are convenient ones, and in several instances give illustration of the character and early life of the people : Tompkins, ibid., and History of Mecklenburg, vol. i, pp. 4-6, 14-15, 18-19, 97-98; Thompson, pp. 17-18, 20-22; Hart, p. 32. County and State histories are helpful in this connection. 269I] THE LABOR FACTOR 1 63 consideration, are mainly at variance on three points. The first is whether there was and continues to be a difference in essential character between the indigent classes in the mountains and foothills and in the low country; second, whether the name "poor whites" is applicable to both of these groups ; third, whether there was a middle class in the South, at and before the period of mill building, which was to be distinguished from the lowest stratum of population. Fortunately for purposes of illustration, observations of writers anywhere from about 1840 forward can be used, because the character of the people from whom factory hands were recruited did not change materially from the time that cotton became king until their ranks had become greatly thinned by influx to the mills. One who employed broad terms spoke of the "non- slaveholding white men . . . outside the essential councils of the South," who " stood aloof ; they were supposed to follow where others led," 3 and said it was from this " vague multitude of the unlettered and unskilled . . . from the great army of the non-participants that the population of the factory is chiefly drawn." 4 Mr. Thompson asserts a difference between the indigent whites of the mountains and those nearer the middle por- tion of North Carolina, saying that in the extreme west the inhabitants in i860 lived the same primitive lives as their grandfathers, while unpropertied whites in the Piedmont were not socially distinguished from their more fortunate neighbors until a late date. White men would often assist a landowner whose slaves were insufficient, at such times sleeping in his house and eating at his table. " Indeed, it is not too much to say that the Piedmont section of North Carolina was more nearly a social democracy after 1840 than were the manufacturing sections r ~of New England, where by that date there was a well-defined manufacturing aristocracy." The Civil War, however, marked the com- 3 Murphy, pp. 14-15. 4 Ibid., pp. 104-105. Cf. ibid., p. 103, and Phillips, in South Mo- bilizing for Social Service, p. 567. I64 THE RISE OF COTTON MILLS IN THE SOUTH [27O mencement of the increase of tenant farmers and share- croppers with consequent class cleavage. Those after- wards in very poor circumstances had been closely associated in general estimation with the small traders and profes- sional men. 5 The common origin of mountain whites and tenant whites and the applicability of the term " poor whites " to both groups is noticed by Mr. Hammond, who calls them all quite properly, in view of circumstances in which they found themselves, "parasitic." 6 Along with his character- istic bias and exaggeration is the usual portion of truth in this observation of a Northern newspaper correspondent who traveled through the South in the autumn following Lee's surrender : " Whether the North Carolina ' dirt eater/ or the South Carolina ' sand-hiller,' or the Georgia 'cracker,' is lowest in the scale of human existence would be difficult to say. The ordinary plantation negro seemed to me, when I first saw him in any numbers, at the very bottom of not only probabilities, but also possibilities, so far as they affect human relations; but these specimens of 5 Thompson, p. 09 ff. Early title deeds show the settlers in the Piedmont of North Carolina to have been weavers, joiners, coopers, wheelwrights, wagon makers, tailors, teachers, blacksmiths, hatters, merchants, wine makers, surveyors, fullers and "gentlemen" (Tomp- kins, History of Mecklenburg, vol. i, pp. 24-25). Slavery and cotton had worked their change by 1856, when Olmsted wrote that " the slaveholders have . . . secured the best circumstances for the em- ployment of that slave-labor which is the most valuable part of their capital. They need no assistance from the poor white man : his presence near them is disagreeable and unprofitable. Condemned to the poorest land, and restricted to the labor of merely providing for themselves the simple necessities of life, they are equally indif- ferent and incompetent to materially improve their minds or their wealth" (p. 515). Cf. ibid., p. 296; Tompkins, ibid., p. 88. 6 Speaking of cotton culture before the War, " the majority of the white laborers were of the class of ' poor whites,' many of them descendant's of the ' redemptioners.' . . . these people . . . had be- come the parasites of Southern society. Some of them were forced into the mountain region of eastern Tennessee and Kentucky and western North Carolina, and others were left on the abandoned cot- ton and tobacco lands of the sand hill region of South Carolina and Georgia" (Hammond, p. 97). 27l] THE LABOR FACTOR 1 65 the white race must have reached a yet lower depth of squalid and beastly wretchedness." 7 That the poor whites were the victims of the economic regime and that their laziness was to be attributed in large measure to this prime fact, has been made clear by a keen and sympathetic student of Southern economic history. "All whites who were poor were not 'poor whites,' 1 but many embraced in that term of contempt and pity were poor ... in the ambition to contend against what seemed to be the inevitable." He thinks that, corresponding to the countryman in New England, there were very moderately circumstanced whites in the South that might be taken as constituting a "yeomanry," but that below these were "the neglected people who . . . were but little removed from the status of the settled Indian. . . . They were the de- generates, the children of ancient poverty and wrong, with little or no opportunity to better their condition among sur- roundings of a corrective character. . . . Had they not been too lazy to wander far from their apologies for home,., they would have become American gypsies. . . . The vic- tims of heredity and of institutions in which they had no- interest, placed under laws made for them rather than by them, they were happily removed from the pressure of population that would undoubtedly have reduced them ta- the criminal or the dependent class." 8 7 Andrews, pp. 335-336. " The Georgia ' Cracker ' . . . seems to> me to lack not only all that the negro does, but also even the desire for a better condition and the vague longing for the enlargement of his liberties and his rights." 8 Ingle, p. 22 ff. " John Forsythe of Mobile hit off some of their traits in contrasting the unadulterated ' Cracker ' and an unadul- terated Yankee, born and bred in the country. ' One is slow . . . and the other quick; one takes a minute to rise from his seat, the other never sits at all except in pursuance of a calculation ; one is not without faculties, but they seem to be all asleep, the other with all his wits alive with sagacity, curiosity, invention. The one con- tent to doze away life with as little labor as possible and all the- enjoyment compassable; his log. hut, wool hat, homespun suit, and corn and bacon the limits of his desires . . . ; loving his gun and. his horse, addicted to tobacco and strong drink,' quick to anger, a dangerous enemy, and a fast friend. The other instinct with life . . . never satisfied with the present wellbeing while anything better 1 66 THE RISE OF COTTON MILLS IN THE SOUTH [272 Governor Hammond, of South Carolina, was moderate when he said: "According to the best calculations which, in the absence of statistic facts, can be made, it is believed that, of the 300,000 white inhabitants of South Carolina, there are not less than 50,000, whose industry, such as it is, and compensated as it is, is not, in the present condition of things, and does not promise, hereafter, to be, adequate to procure them, honestly, such a support as every white per- son in this country is and feels himself entitled to." 9 Professor Hart believes that the term " poor whites " means lowlanders, and that the mountaineers belong in a different category. His reason is chiefly that the mountain whites do not have to contend with the universal presence of the Negro. It is to be remembered that this distinction is of later emergence, and that slavery was responsible for is beyond to tempt his longings and his wits.' " A South Carolinian who seemed to be informed gave Olmsted his opinion that com- munities of poor whites on the banks of the Congaree River were in more hopeless plight than the degraded peons of Mexico, and a rice planter described similar people living in the pine barrens near- est the coast : " They seldom have any meat . . . except they steal hogs, which belong to the planters, or their negroes, and their chief diet is rice and milk. They are small, gaunt, and cadaverous, and their skin is just the color of the sandhills they live on. They are quite incapable of applying themselves steadily to any labor, and their habits are very much like those of the old Indians " (p. 505 ff.). A Northerner told Olmsted of stopping once at a sand-hiller's cabin. One of the four grown daughters was weaving, the others seeming to have nothing to do. " ' I asked the girl at the loom how much she could make a day by her work. She did not know, but I ascer- tained that the stuff she wove was bought at a factory in the vicinity, to be used for bagging yarn ; and she was paid in yarn. . . . She traded off the yarn at a store for what she had to buy. ... If she worked steadily from daylight to dark . . . her wages . . . were less than sixteen cents a day, boarding herself. . . . These people are regarded by the better class with as little respect as slaves . . .'" (ibid., p. 507). This was in South Carolina. Twenty-five or thirty years later such establishments as this bagging mill had largely dis- appeared, the bartering of yarn was no longer practiced, and such a family of girls as here described was in all likelihood working immediately in a cotton factory for money wages. 9 Quoted in Olmsted, p. 514. Here again is the thought that they were crowded out of occupations : " Some cannot be said to work at all. They obtain a precarious subsistence by occasional jobs, by hunting, by fishing, sometimes by plundering fields or folds, and, too often, by . . . trading with slaves, or seducing them to plunder for their benefit." 273] THE LABOR FACTOR 167 the history of the class of unfortunate whites, whether they were left in the low-country, stranded upon the sandhills between coastal plain and Piedmont, or driven into the hills. 10 — The pertinence of recent accounts of the poorer moun?- tain and tenant whites in their native surroundings is illus- trated by the fact that the mills very recently were receiving families in just as destitute condition as those which first entered the factory communities. 11 They regularly came with empty hands. An episode recited of a mill at Spar- tanburg is typical, where "one day a covered wagon or mountain schooner drove up to the . . . office. It was full of~family and that wasabou t all. ' Y ou could put uponj a. smalLtabl e all the earthly pos sessions of t hat family/ s aid Mr. Montgomery. T he man asked for wo rk. Mr._Mont- gomery told the superintend ent tofindjhem a_vacant house. ' Buf whaTaibouttne rashun s? 'Inquired 1 the new 'help .' " 12 The most recent historian of the American industry in his description o>f the people who filled the mills of the South does not distinguish between Piedmont, mountain, and lowland (tenant) whites. 13 It has been seen that while many of the Southern mill ventures were undertaken partly with the express purpose 10 Southern South, p. 30. For some account of the middle-country poor whites, with a list of the disparaging names applied to them, see ibid., p. 38; a description of the mountaineers (p. 34 ff.) is most dismal. 11 An admirable recent picture of the life of the poor whites in mountain and lowland sections is contained in a painstaking pam- phlet by Frances Sage Bradley and Margaretta A. Williamson, " Rural Children in Selected Counties of North Carolina," published by the Children's Bureau of the U. S. Department of Labor. "Columbia Record, Textile Ed., 1916. Of a factory at Rock Hill it is reported : " A man who moved to the mill from Union County a few years ago was so poverty stricken that he had not even a bed upon which to sleep. He was in such poverty that it was a matter of jest" (ibid.). Of another mill village it is told that " nineteen families have moved into this community within the last fourteen years, bringing their entire worldly possessions in one wagon load . . . ; none of these . . . families had a stick of fur- niture or a sack of flour or the means to provide for the same" (ibid.). 13 Copeland, pp. 40-41. L*^ 1 68 THE RISE OF COTTON MILLS IN THE SOUTH \_27A of giving work to the poor whites, in a good many cases the opportunity for profitable employment of these people was entirely overlooked, this giving color to the belief that in proportion as the poor whites dropped out of participation in the economic order, they tended to drop out of the mind of the dominant class. The abolition of slavery did not bring the neglected men and women immediately back into the thought and sympathy of the South any more than into the employment of the South. 14 It has been seen that William Gregg, the builder of the Graniteville Factory in South Carolina, was the father, in the sense that he was the anticipator, of a new economic life for the South. His keen consciousness of the poor whites stands out in striking contrast to the state of mind indicated in the preceding paragraphs. It is interesting to notice a statement of Gregg's which shows clearly the con- dition of the lower strata of the white population fifteen years before the war ; it is to be remarked that he was com- bating a tendency not simply to omit the poor whites from consideration, but to place the negroes ahead of these even, as possible industrial workers. " Should we stop," he asked, " at the effort to prove the capacity of blacks for manufac- turing? IShall we pass unnoticed the thousands of poor, ignorant, degraded white people among us, who, in this land of plenty, live in comparative nakedness and starva- tion ? "f A.nd he continued : Many a one is reared in proud South-Carolina, from birth to man- hood, who has never passed a month in which he has not some part of the time, been stinted for meat. Many a mother is there, who will tell you that her children are but scantily supplied with bread. . . . These are startling statements, but they are nevertheless true, and if not believed in Charleston, the members of our Legislature, who have traversed the State, in electioneering campaigns, can attest their truth. 14 A Virginia correspondent of the American Agriculturist before the War asserted that whites could be got to work for less price than blacks, but the slaves were preferred. Newcomers were ad- vised, if they wished to use whites, to bring them with them, since the native white population was inferior to the black (quoted in Olmsted, pp. 211-212). A farmer in the same State who employed only free labor found Irishmen at $120 a year the best workers; native whites were declared worse than free blacks (ibid., p. 99). 2/5^ THE LABOR FACTOR 1 69 It is only necessary to build a manufacturing village of shanties, in a healthy location in any part of the State, to have crowds of these poor people around you, seeking employment at half the com- pensation given to operatives at the North. It is indeed pitiful to be brought in contact with such ignorance and degradation ; but on the other hand, it is pleasant to witness the change, which soon takes place in the condition of those who obtain employment. The emaciated, pale-faced children, soon assume the appearance of robust health. ... It is, perhaps, not generally known, but there are twenty- nine thousand white persons in this State, above the age of twelve years, who can neither read nor write — this is about one in every five of the whole population. 15 A writer already quoted refers to the poor whites of the ante-bellum South as constituting part of the last grade of a class distinguishable from both the unpropertied and the influential landowners, which might be termed a "yeo- manry," but he notices their tendency to sink rather than rise in the social order. 18 Thus again it is indicated how the pressure of slavery, if it worked to bring a small number to the surface, gave to masses an impulse ever downward. There is very little to show the character of the white operatives in the small and scattered factories that existed in the South prior to the great rise of mills about 1880. Many were doubtless immigrants or descendants of recent immigrants. The Graniteville mill had workpeople who did not differ materially in their economic or social aspects from those in later manufacturing communities, and per- haps the same may be said of a few other establishments in the ante-bellum period. But Graniteville was not typical of 15 Domestic Industry, p. 22. It is to be observed that knowledge of the plight of the poor whites gained in electioneering campaigns was passive, and did not awaken a purpose to improve conditions. Gregg himself, as a member of the legislature, was the exception that proved the rule. Despite the difficulty of travel and the absence of " statistic facts," as Governor Hammond said, public ignorance of a 20 per cent illiteracy in the white population is as reprehensible as the fact of the illiteracy itself. When Gregg was working out his philosophy in practice, he reported to his Graniteville stock- , holders in 1855 that 79 in 100 grown girls who came to the mill could neither read not write, adding that " that reproach has long since been removed" (quoted in Kohn, Cotton Mills of S. C., p. 21). Cf. statement of a colporteur in Olmsted, p. 510. 16 Ingle, pp. 20-21. 170 THE RISE OF COTTON MILLS IN THE SOUTH [276 I I its time. Graniteville tapped a class of labor as a class; the 'smaller factories, with all sorts of local limitations in situa- tion, power, machinery and peculiarities of operation, at- tracted only individuals, had no labor objective. It was not recognized that any widespread condition existed that made employment in mills desirable, and no distinctive problems grew out of the collecting of persons in the little villages surrounding the factories. That many negroes were used in these enterprises, alone or with whites, helps to blur the picture of the white operatives. In the matter of labor, these early establishments corresponded roughly with grist mills and saw mills then and today. Nobody bothered about where the employees came from or why. It is probable that in most instances they had been living in the imme- diate localities. It may be concluded that the difference between the mills before the great period and those which followed, with respect to labor, was one of size of the manufacturing unit and of degree of standardization of the industry. 17 The amount that had to be done for the poor whites after they came to the mills (speaking now of the large develop- ment of factories), and their too evident entire newness to the demands of progressive living, reflect a light back upon the years in which they had been pushed aside. The his- tory of the industry since 1880, in the human phase, has been chiefly the effort at reinstatement of a great portion of the population previously neglected. 18 So metimes th e peopkbxou ght with them littl e besjd ej__bjid^ habits and a 17 Glimpses of before-the-war operatives frequently indicate for- eign birth or ancestry, and are not always inspiriting. Cf. Olmsted, PP- 356-357; Buckingham, Slave States of America, vol. i, p. 171. 18 A recent president of the chamber of commerce of a capital city said that while in office he refused to give his especial support to projects to establish cotton mills in the place because of all the people who came to a factory, only five or six families would be composed of desirable citizens, the rest lowering the average of population. " You have to take care of these people when they are sick," he explained, " and you must give them schools and churches. Thousands of dollars, of course, were spent in eradicating the hook 277^ THE LAB0R FACTOR 171 total dependence u pon the management for moral care a nd ^ysicah upbnMTngT® HoweTeTTmich the poor whites had failed of recognition before, instances are rare in which mill men, at the outset of the factory era in the South or later, have complained of the quality of the operatives. It may be said that the work of a cotton mill, certainly a mill on coarse goods, is scarcely skilled at all, and that in the beginning management was as unaccustomed to its task as spinner and weaver to theirs. It may be observed that labor was above all cheap, and that advantage thus conferred silenced all objection. But the fact is not altered that Southern mill owners showed a splendid faith in the capacity of their workpeople. North- ern superintendents in Southern manufactories seemed unanimous in their satisfaction with the labor. One of the most distinguished of Southern mill projectors wrote in reply to some doubting remarks of another South- ern manufacturer : " I do not admit that the Northern peo- ple are any better material out of which to make cotton manufacturers and operatives than our own, and especially in the ' Piedmont belt,' of the South, is the best in the United States, and capable of being educated to as high an order of skill as any other. I have been in most of the best mills at the North . . . and have observed their operations closely, and I challenge that there is as high skill and an equal degree of expertness in the operatives of the Pied- mont Mill, as far as the kind of goods made requires . . . as is to be found in any mill in New England." 20 19 The head of a large establishment told how " ninety per cent of the operatives — kids and all — used to use snuff. We would get from the loom-boxes, where they would leave them, a barrel of snuff boxes a week in cleaning. Now not fifteen per cent use snuff" (T. S. Raworth, int., Augusta, Ga., Dec. 30, 1916). 20 Letter of H. P. Hammett to Atlanta Constitution, quoted in Manufacturers' Record, Baltimore, Feb. 1, 1883. An old man, look- ing back to the starting of his mill forty years ago, said to the writer with a determined look in his eye : "In a speech made in Atlanta at the Exposition [1881] Edward Atkinson told us that we couldn't manufacture goods in the Southern States because we couldn't get help down here; that we should let them manufacture the cotton and we raise the cotton. I saw the help coming in from 1^2 THE RISE OF COTTON MILLS IN THE SOUTH |_2^8 It has been seen that some writers would distinguish be- tween the mountain whites, the poor whites of the Pied- mont belt and the corresponding group in the coastal plain ; that some question exists as to the application of the term " poor whites " ; and that some believe there was a tolerably defined middle class in the South before and following the war. However these facts may be, it is chiefly important to understand that the mills drew from all these divisions of poor whites, and if there was a group between them and the upper whites, it did not work to alter the essential eco- nomic situation. Whatever technical differences existed prior to the opening of the factories, in the willingness to seek mill employment there was a general merger of types of indigent white people. 21 Genenous estimates of the capacities and promise of the poor whites in the mills and out of them are as easy to find as it is natural to give them. Anyone who sees the people in the country or in the industrial communities and who knows anything of their lives, feels a respectful warmth go out to them. With all the marks of their hindrance upon them, he must recognize that they have all the worth which the best blood in America can bestow. 22 dinner at Fall River in the eighties, and it couldn't compare to ours!" (Charles Estes, int., Augusta). Another old man declared: " North Carolina has within its borders more Anglo-Saxon blood than any other State in the Union. There is no better labor in the United States than in the cotton mills of North Carolina" (Charles McDonald, int., Charlotte, N. C, Sept. 3, 1916). It has been prop- erly observed that the term " poor white trash," common in writing about the South, is rarely used by Southern whites. " They are un- progressive, they fail to make the most of their opportunities, but they are not degraded. It is suspended or arrested development rather than degeneracy" (Thompson, p. 113). 21 Cf. Thompson, pp. 69-70. 22 The statement of Mr. Baldwin, principal of the Piedmont Indus- trial School at Charlotte, while very familiar, is worth quoting here. He is speaking especially of the operatives in his own section : " I am satisfied that they are the finest body of people on earth doing similar work. Descended from the early English, Scotch and Ger- mans, they have been sleeping, as it were, while the procession of progress has been passing by. Serious, independent, as all hill and mountain people are; sensitive, because of that independent spirit, for the most part sober, they are a people of untold possibilities, now that they are beginning to arouse themselves from the drowsi- 279] THE LABOR FACTOR 173 The cotton mill operatives came immediately from the soil. The cotton manufacturing' South sprang directly from the cotton growing South. It is probable that never before or since in economic history has an agricultural population been so suddenly drawn into industry. The sharp emergence of manufacturing from farming, the more abrupt because long delayed, is in a large way the theme lof this study. The picture is one with a cotton mill in the foreground and acres of cotton plants in the background, stretching away almost to the horizon. The relation between farm and factory was especially close in the case of labor. In the decision of individual men and families to leave the land for the manufacturing vil- lage it is possible to see, very tangibly, the working of causes that were moving the whole South. In another place the counter pull of the plough against the spindle will be mentioned, when it will be shown how now one and now the other, in the estimation of workers, has gained ascend- ancy. At this point it is important to notice briefly the agricultural conditions prevailing at about the time of the rise of the mills. It has been said of North Carolina that "before 1890 the question of satisfactory labor had not been entirely solved. The better class of labor was not easily drawn from the farm9 to the factories." After 1890 the price of cotton, due to increased production of the domestic staple, to the size of Egyptian and Indian crops, and the depression fal- lowing the panic of 1893, fell lower and lower. The crops of 1894 and 1895 brought for the most part about five cents per pound, and low prices of wheat, corn and tobacco ac- companied the drop in cotton. Fertilizer bills were hard to meet, mortgages were difficult to carry. Cotton mills were running day and night and selling yarns in the markets of Philadelphia, New York, Boston and the Orient. In this ness of generations and to grapple earnestly with the duties of this active, work-a-day world" (quoted in Goldsmith, p. 27). Cf. Kohn, Cotton Mills of S. C, pp. 21-22. " These people are all Americans, and hundreds could qualify as Sons or Daughters of the Revolu- tion" (Thompson, pp. 110-111). 174 THE RISE OF COTTON MILLS IN THE SOUTH [j280 condition of things, farms were sold, rental arrangements were not renewed and industrious and lazy alike flocked to the mill communities. 23 In the case of other Southern States, the development came earlier and more abruptly than in North Carolina, and the abandonment of farming for the factory occupation was not so dependent upon the price of the staple at the particular time. Even in North Carolina, however, the causes back of the migration, if it may be spoken of a9 such, were of much longer standing than the account just given might be taken to indicate. The condition of South Carolina in the decade before the war, in which the average value of the productive industry of the State was declared not to exceed $62 per head of the whole population, omitting the two largest cities, persisted, roughly, down to the years of the rise of the mills. 24 The desperate, almost comical poverty of after-the-war years left on the minds of men who lived through them im- pressions that will not be erased. " In my county," said one of these, " the term ' farmer * applied to a man was a name something very like reproach. Every bull yearling was under chattel mortgage." 25 23 Thompson, pp. 69-70. Cf. H. J. Davenport, Economics of En- terprise, p, 201. 24 From an article on the agriculture of South Carolina, written for The Carolinian by a resident of the State, and printed after- wards in DeBow's Review; quoted by Olmsted, pp. 518-519. "Full one-half, or more, of this amount is consumed on the plantation or farm, as necessary means of subsistence; leaving about $31 as the value of cotton and other marketable produce, per head." 25 Henry E. Litchf ord, int., Richmond. The story of a family brought to a Charlotte factory when the Mountain Island mill was washed away in the summer of 1916 is illustrative of conditions pre- vailing forty years ago. The old woman and her three daughters had recently become operatives, and had nothing. With a fourth daughter, afterwards married, the family had tried to farm in the foothills. They made fairly good crops, the girls working in the field, but, in payment for land, stock, implements and feed, the land- lord took all they made above a bare living and a dress or two a year and a pair of shoes for each occasionally. When the old woman finally left for the mill village she was able to pay herself out of debt and, so "the man" told her, she had $7.50 coming to her in cash, but this she never got. In the mill town they proved thrifty, the mother managing to keep her family going a whole week on $5 advanced by the management (Sterling Graydon, int., Char- lotte). 28l]] THE LABOR FACTOR 175 One who has witnessed the economic awakening of Greenville County, South Carolina, from the commence- ment, rehearsing the evils of the system under which farmers bought on credit, paying once a year, frequently by note, much to the hurt of the agricultural community, spoke with satisfaction of the change since that time. He said that now no merchant in Greenville does a time business with farmers. The latter get small loans at the banks ; one bank has for many years been lending some people regu- larly such small sums as thirty dollars, and it will lend as little as ten dollars. He remembers what may almost be said to have been the beginning of a local money economy. He saw the first whole bale of cotton ever brought to the Greenville market. The man who purchased it was con- sumed with fear as to his wisdom in putting so much money in cotton. Would the county ever need so much? This was about 1870, and gives a notion of the pettiness of farm operations in the up-country region then and later. 26 A system of tenancy in which the farmer contributed little or nothing besides his own labor ; in which, by custom, by pressure of the landlord, by dictate of his creditor mer- chant and by absence of initiative, the tenant raised only cotton ; and by the working of which the proceeds of a crop, on which a lien was held, were consumed before they were realized, could not make agriculture promising. 27 It is re- 26 W. J. Thackston, int., Greenville. The almost total absence of money in rural communities will be noticed later. Mr. Kohn, after reviewing the situation of operatives at the time they were farmers, came to the conclusion that " the attraction of the cotton mills, to those who are in them, in a word, is the cash money" (Cotton Mills of S. C, p. 26). Cf. ibid., pp. 22, 27. 27 Some facts gathered by Mr. Kohn as to recently prevailing tenancy arrangements in South Carolina serve as a fair picture of earlier conditions. In the Pee-Dee section the landlord ordinarily paid for fertilizers, ginning, bagging and ties, and the tenant received half the crop. It was thought good for a tenant to " make " fifteen bales of cotton, his half, at $50 a bale, bringing him $375. The sale of a few bushels of corn not needed to feed the stock, and hauling and other work might net him $150 additional, a total of $525. This family might have one plough and two hoe hands. The same family in a cotton mill, at the time of writing, would have made about $900. A tenant in the Piedmont section, having to share in the cost of fer- i / I76 THE RISE OF COTTON MILLS IN THE SOUTH [282 lated that the help for a mill built as late as 1896, picked up on the neighboring farms, "had no money, no prospects. Cotton was the only money crop and the price, four and one-half cents, was such as to make a year's wages insig- nificant by comparison with what could be earned in the mills. They came to the mills for employment, for relief from the weight that pressed down upon them." 28 Having seen something of the character of the poor whites and the economic situation in which they were before the building of the factories, it is natural next to examine thee xperience of the mills i n recruiting labor. First will be noticed the cases, alimost"uKivefsal, in which applicants for work were plentiful, and afterwards some instances in which, for special reasons, operatives were not so readily obtained. The labor motive for the building of mills has been dealt with in a previous chapter. Plenti fulness of labor is an easy conclusion from the arguments advanced that cotton manufactories should be established in the South because labor was cheap and because the employment would be a benefit to large numbers who had only precarious means of livelihood. In only a few cases in which sufficient labor for a proposed mill was felt assured, did the anticipation prove incorrect. There was little guessing involved; it was a mine the veins of which lay in a net- work on the surface of the earth. The question was whether mills could be built, . not whether they could be filled with workpeople pressing to be admitted. Gregg's recognition that the poor whites would make tilizer, would have very little left after meeting the advances of the merchant and fitting out his family with clothes (Cotton Mills of S. C, pp. 27-28). Cf. News and Courier, Charleston, South Caro- lina in 1884; Bradley and Williamson, pp. 20-21; Charles H. Otken, The Ills of the South, chaps, ii and Hi; Hammond, pp. 144 ff., 155. 28 Columbia Record, Textile Ed., 1916. For a good statement of the reasons why poor whites came to the mills, see Derrick, " The Cotton Mill Population of the South," in Bulletin of Newberry Col- lege (S. C), vol. ii, no. 8, pp. 32-33. Cf. Thompson, p. 114 ff. on this point and for an interesting classification of types that enter the factories ; the same classification might have been made forty years ago with equal truth. 283] THE LABOR FACTOR 177 good cotton mill operatives is matched by. the view of a Northern man made a decade before the war, that if the cotton manufacturing industry should be founded in the South, labor would be in supply. He urged that cotton planters should become cotton manufacturers, showing how the profits from industry were greater than from agricul- ture, and continued : " But, after having admitted all this, the cotton planters and capitalists of the South raise the inquiry : Suppose we wished to go into the manufacturing business, though we had plenty of raw -material, how should we obtain the labor and skill qualified for the work, and of both of which we are deficient?" This conjectured in- quiry, one coming naturally from owners of large planta- tions worked by negro slaves, was answered without hesita- tion : "... a fine supply may at all times be obtained, in New England, to manage and supervise . . . operations . . . and there are thousands of persons at the South, who would gladly and gratefully accept such employment to earn a livelihood, much superior to that which their present means can possibly afford ; and would quickly become qual- ified for the work of operatives, under the charge and direction of good . . . managers. ... In a comparatively short period, hundreds of factories might be erected and started at the South, and fully supplied with every descrip- tion of skill and labor wanted." 29 Impossible as was this proposal for widespread manu- facture of cotton at the South at the time it was made, the prophesies it contained were realized, when finally the mills were built, with remarkable completeness. Thus thirty years after James wrote, the president of the Louisville and Nashville Railroad was able to say in a Northern paper : " Mills for the weaving of the coarser cotton fabrics are now in successful operation in Tennessee, Georgia, Ken- tucky and several of the Atlantic coast States. . . . The labor question in the South, which a few years ago pre- sented many difficulties, is now as practically settled there 29 James, in DeBow, vol. i, p. 233 ff . 12 I78 THE RISE OF COTTON MILLS IN THE SOUTH [284 as in any other portion of the land. The class formerly known as 'poor whites' are mixing and assimilating with their more fortunate neighbors. They are making good workers in mine and field, good operatives in factories. . . ." 30 It was stated in 1880 that within a few months five hundred white North Carolinians had left the State to seek homes in the West, and that the movement was increasing. The number of emigrants with sufficient energy and means to go far away did not need to be large to indicate that there was a surplus of labor. 31 The Atlanta correspondent of the New York Times, de- scribing the cotton mill campaign in the South, said that " there is an abundance of native white labor to be had at from 50 to 60 cents a day " ; explained that while negroes had not been proved entirely unsuitable for the work, — " there are white men and women enough for all present demands," — and continued: "Of the many benefits which the community at large, as distinguished from the capitalist and manufacturer, will enjoy from the extension of manu- factures in the South, the chief one will be the opportunity afforded for the profitable employment of thousands of hands now idle." White labor must yield to black in cotton growing and in the less skilled trades. " Shut out in so many directions the whites, who now find life a bitter strug- gle, will gladly turn to> the spindle and loom as a means of gaining a livelihood. Manufacturing will be their deliver- ance. . . . For girls and women who have hitherto had no Opportunity to earn money the establishment of factories in every town and village will be an incalculable blessing." 32 30 Quoted from New York Herald in News and Courier, Charles- ton, July 11, 1 881. By 1888 the abundant supply of labor in South Carolina was not only recognized in the State itself as an asset, but was advertised as such to manufacturers who might be considering locating there, the State department of agriculture publishing that " the manufacturer of cotton goods finds ... a population willing and anxious for employment, out of which can be made as intelli- gent, skillful and reliable operatives as are to be found anywhere " (South Carolina Department of Agriculture, Sketch of Industrial Resources of S. C, 1888, p. 27). 31 Concord Sun, quoted in Observer, Raleigh, Feb. 24, 1880. 32 Quoted in News and Courier, Charleston, Nov. 5, 1881. How 28 5 ] THE LABOR FACTOR 1 79 From one of the new cotton mill localities in 1881 came the following, which has the distinctive flavor of the times : Not only should there be different kinds of crops, but we ought to have other ways of securing a livelihood besides farming. There ought to be other kinds of work furnished the girls of the State besides housekeeping. The factories that are springing up over the country will help them a great deal. Here is a factory established at Piedmont which will give employment to six hundred persons, half of whom will be girls. But we need others. There is a man here now from Edgefield who has a family of six girls and who has come here to get them work in the Piedmont factory. But he is too late. Every house in the place has been engaged and there are twenty families that have applied for positions, but have been re- fused because they are not needed. Four families of thirty persons have moved in since yesterday. Many who were not idle or even, perhaps, exactly " mar- ginal " producers, came to the mills, thus increasing the vis- ible labor supply. It was said that " as soon as the crops are gathered all the others that have secured places will move here. The population at present is over one thousand and it will be 1,500 in two months. There are more carpenters and mechanics employed here now than at any past time. . . . 240 rooms are being plastered." 33 little conditions in the South varied from one locality to another, how universal were the causes which underlay its economic, plight, are instanced on every hand. Places outside the South were more likely to possess peculiar economic characteristics. Thus a Philadel- phia textile journal remarked that "Baltimore . . . offers some of the best advantages for starting manufacturing establishments of any point in the United States. . . . Labor is plenty and cheap, there being a great number of females who are employed during the pack- ing season, which lasts but a short time; the balance of which they eke out a miserable existence by sewing." Here was a purely local circumstance (Philadelphia Hosiery and Knit Goods Manufacturer, quoted in Baltimore Journal of Commerce and Manufacturers' Record, July 15, 1882). 33 News and Courier, Charleston, Oct. 21, 1881. The son of the founder of this mill told the writer that " there was no opposition among the country people against the mills. At Piedmont in the early days it was impossible to give employment to all that offered themselves" (James D. Hammett, int., Anderson). The rush to hastily constructed mill villages, though from a local region, was much like the lightning growth of gold towns in California and Alaska, and, more recently, at munition plants. Of the Clifton mill in South Carolina it was said that " there are families coming in constantly and the cottages as fast as completed are occupied, and still they come" (News and Courier, Charleston, Oct. 21, 1881). There were many reasons for a large proportion of women and girls 180 THE RISE OF COTTON MILLS IN THE SOUTH [286 Of operatives proper in Southern mills, the census of 1870 showed that women comprised 41.2 per cent. In 1880 the percentage was 49.4, but by 1890 it had receded again to 40.6. In the New England mills, on the other hand, the proportion of women in all classes of employes was a little higher in 1890 than in 1880 — 49.4 per cent as against 49.2 per cent. In Southern mills the percentage of children decreased slightly between 1880 and 1890 — from 24.5 to 23.7, whereas in New England the proportion of children fell away greatly in the decade — from 13.9 to 6.8 per cent. In the South the percentage of men increased from 30.2 to 35.6, and in New England from 36.8 to 43.7. Thus in New England mills, decrease in the proportion of children was accompanied by an increase in percentage of men, but also by some increase in percentage of women. In the South, on the other hand, a slight reduction in propor- tion of children was coincident with an increase in the pro- portion of men and a correspondingly sharp fall in propor- tion of women. In New England there was a relative elimination of children, and in the South of women. 34 in the ranks of those who applied to the cotton mills for employ- ment. Elsewhere the effect of the Civil War in reducing the number of men and boys and in crippling others is noted. It was less easy for females to compete with colored labor than for males, not only from physical but from social causes. The cotton factories offered a field from which negroes were excluded. The work was light and suited to deft fingers. What applied to women and girls was true in slightly less degree of young boys. 34 The percentage of women in Southern mills in 1880 is taken from absolute figures in U. S. Census of Manufactures, 1880 (Cotton Manufacture, by Edward Atkinson, pp. 15-16), and is higher than that given in the census of 1800 (Cotton Manufacture, by Edward Stanwood, p. 173) — 45.3 — in which all classes of employes, and not simply operatives, were included. Obviously, office force and " out- side" help would include few women. Other percentages approxi- mated in the fractions are from U. S. Census of Manufactures, 1890, ibid. The trend pointed out above may be seen more clearly by taking the year 1900 into consideration. Between 1890 and 1900 the number of children in New England mills increased 8.7 per cent ; women 1.89; and men 23.9. In the South the number of children made a gain of 177 per cent; women 125; and men 223 per cent (percentages from absolute figures in U, S. Census of Manufactures, 1900). Of course, it must be remembered that the number of all employes in Southern mills was greatly increasing after 1880. In 287J THE LABOR FACTOR l8l In the South and in New England the cotton industry, in respect to labor, has eased itself at the points of relatively greatest strain. In the South this meant proportionate de- crease in number of women employed; in New England it meant decrease in relative number of children. If it is borne in mind that in the South there was first great pres- sure for employment, changing gradually to insistent de- mand for workers, distinction in alteration of proportions of operatives in the two sections is not difficult to account for. In the South in the beginning everybody was eager for work, and women seemed better suited to take hold of an industrial task than children; later, when the fullest numbers were needed, the nature of factory work was familiar, and more children could go into the mills if mothers worked at home. 35 Census figures are borne out roughly by many references that may be found relative to labor in the mills at the outset of the period. Thus two-thirds of the operatives at Lang- ley were female (girls included with women) in 1880, and it was reported of Graniteville and Vaucluse that "the number of operatives employed is 775 ; two-thirds of whom are females and who range from 11 years up." 36 In a fac- tory at Selma, Alabama, "the operatives number 120, mostly women and children, taken from Selma and vicinity." 37 The prevailing low rate of wages, as also variations in wages between one mill and another, may be taken as indi- cations that labor was in abundant supply. An examination 1850 and i860 the number of women was about the same — 6157 and 6039. In 1870 there were 4190, in 1880 there were 7587, and in 1890 there were 15,083 (ibid.). 35 Edward Stanwood (Cotton Manufacture, p. 33, in U. S. Census of Manufactures, 1900), was mistaken in neglecting these considera- tions. " Whole families in that region," he said, meaning the South, " enter the factories, because in no other way can the demand for labor be satisfied. Consequently the changes in the proportion of men, women and children employed are largely fortuitous." On the face of it, his statement is unfortunate, because taking together great numbers of families entering the mills, a statistical trend would easily show itself; moreover, after a family has been in the mill a while, some members may discontinue the factory employment. 36 Blackman, pp. 7 and 4. 37 News and Courier, Charleston, March 31, 1881. 1 82 THE RISE OF COTTON MILLS IN THE SOUTH £288 of newspaper files covering the opening years of the cotton mill period failed 1 to disclose a single advertisement for operatives. When it is remembered 1 that factories sprang in great numbers and simultaneously from an agricultural regime, this is striking. 38 An article summarizing a newspaper correspondent's study of the South Carolina cotton mills in 1880 declared that " the difficulty in obtaining operatives is not great, it seems. Indeed no new industry has ever been adopted with less difficulty, and with fewer drawbacks and discourage- ment [sic], than the business of manufacturing cotton in South Carolina." 39 There appeared to be no apprehension about getting operatives for the largest plants. Thus the King mill at Augusta in 1883 began production confidently. "The first beam was taken off the slasher Wednesday morning ... at 10.30 o'clock, and the first loom was started Wednesday afternoon at 3 o'clock. Last evening there were fifty-three looms running. Supt. Smith reports that so far he has had no trouble getting hands, and does not anticipate trouble in this direction." 40 38 It is true, of course, that newspapers were then not so widely read as now, and did not reach to very large extent the people who were attracted to the mills. Many of the first mills were from the start operated at night, which required a double force of hands. Thus more mills than were built might have sprung up and had labor to run during the day, without exhausting the labor supply, providing the conclusion reached in this study, that workers were plentiful without respect to locality, is correct. For example : " Quite a num- ber of Mr. Cornelson's new factory hands have already arrived at Orangeburg, and the mill is now being run at night" (News and Courier, Charleston, April 9, 1881). 38 Leading article, in Blackman. The article was probably written by F. W. Dawson, editor of the News and Courier. 40 Chronicle, Augusta, Nov. 11, 1883. The first president of this mill told the writer that the factory " got plenty of help right here locally, all natives " (Charles Estes, int., Augusta). A factory which had its start before the cotton mill campaign was in every sense a local enterprise. Its operatives were described as being " all natives, with one exception, who have been educated to the business. This class of labor is very readily obtained from the surrounding coun- try" (Blackman, p. 10). Speaking of the beginnings of the cotton mill South, a commission merchant who has been intimately iden- tified with the development said that " labor was superabundant and very cheap" (Summerfield Baldwin, int., Baltimore). 289] THE LABOR FACTOR 1 83 The superintendent of the Langley factory stated that "labor was very plentiful and that they could get 20 per cent more than was required to run the mill. The . . . operatives are made up entirely of the people born and raised right in the vicinity." 41 An old man who saw the founding, of the mills 1 said that the availability of a labor supply did not form a strong motive in the locating of fac- tories, for there was never any difficulty about getting operatives. 42 A superintendent in another State gave similar testimony : "Proximity to a labor supply was not considered in the location of (mills early in the period. There was plenty of labor at first." Mr. Tompkins, explaining what he consid- ered the corrective results of manufactures protected by a tariff, gave a little picture of the South that had been famil- iar: "You all know that fifteen and twenty years ago we did have an army of unemployed. . . . Any town in those old days presented a street spectacle of listless loafers, white and black, leaning against the door facings, telegraph poles and sitting on boxes. Even the dogs caught the list- less spirit and didn't get up to bark." 43 41 Blackman, p. 7. 42 Charles McDonald, int., Charlotte. 43 American Cotton Manufacture and the Tariff, p. 9. For the Arista Mills, at Winston-Salem, the attempt was made to get skilled operatives from other factory communities, but this proved expen- sive and unnecessary, because many in an already floating population offered experienced services and others came in sufficient numbers (John W. Fries, int., Winston-Salem). It will be seen later that labor continued in abundance for a good many years following 1880. Tompkins, whose largeness of view is not often to be interpreted as exaggeration, thought the South had enough idle people to fill fac- tories that would drive England and Germany out of world markets (Cotton Mill, Commercial Features, p. 177). He argued that "Those who know the existing conditions will probably not dissent from the opinion that it would be easy to put 1,000,000 people to work manu- facturing cotton, and never miss them from present employments. Estimating 12,000,000 out of the entire population as being white people, even from amongst these, a million could be more than easily spared" (ibid., p. 20 ff. There is much in this reference to show how cotton mills in the South took up slack in the available working force and improved conditions of urban and rural communities). In 1900 Tompkins believed night work was necessary if all of the mill people were to be kept in jobs. " The night work in cotton mills 184 THE RISE OF COTTON MILLS IN THE SOUTH [29O Having seen how generally willing the people were to offer themselves for work in the cotton mills tor a long period after the first establishment of the industry in the South, it may next be shown from what localities labor was drawn, what were the immediate and what the second- ary regions of supply. Before speaking of the migrations to the mills from districts just surrounding them, however, incidental notice might be taken of the fact that farmers' daughters frequently embraced temporary employment in the little neighborhood mills running before the Civil War. They wanted to make money to buy trousseaux or to help their families, but they did not intend to become factory workers. They perhaps walked to and from the mill morn- ing and evening, or, if their homes were at an inconvenient distance, might live with a friend near the factory. These conditions prevailed with respect to five mills on Deep River in a Quaker community in North Carolina prior to 1850. This was not considered menial service, and the young women often married officials in the mills. The custom was roughly that of farmers' daughters in parts of the South today, who work in canneries in their neighborhoods a few weeks in the summer. 4 * This practice had nothing to do with the readiness with which an agricultural population entered factories from 1880 forward. It is difficult, speaking for the majority of cases, to agree with the statement of Mr. Copeland relative to the smaller Southern factories that " frequently a mill was established in an out of the way place so as to employ workmen who were not willing to move but would work for low wages is better than any other work the operatives can get now or they wouldn't take it. It would be a hardship to close all the mills at night and throw all these people at once out of regular employment " (Labor Legislation, p. 4). Mr. Thompson thought it necessary to state in 1906 that " the difference [in wages] in favor of the factory is so great that only the natural inertia of a rural population com- bined with certain social disadvantages of factory labor prevents an over-supply" of operatives (p. 274). 44 Cf. Thompson, p. 51 ff. With reference to similar conditions in cotton mills in New England at about the same period, see Cope- land, p. 12. 291] THE LABOR FACTOR 185 near their homes." 45 There were instances in which the proximity of a labor supply was a factor in determining the location of a mill, but with these comparatively rare estab- lishments, the thought was that the plant would be closer to prospective hands than other mills, would be in the path of an efflux of labor. In hardly any case could the people do otherwise than move their homes to the village provided by the factory, or to the town in which the factory was located. They usually knew that they were divorcing them- selves from the soil. The mills went to the labor only in the sense that they competed for positions convenient to a general labor supply. It is said that cotton manufactures were located at Anderson, South Carolina, partly because the place is only about thirty miles from cheap labor in the mountains, but workers came to this mill first from the close neighborhood, and afterwards from the mountains. 46 It is true that sometimes the prejudices of the people and their local ignorances assisted a mill placed near them. A superintendent who has had experience in soliciting labor for a large mill in a city said that "a new operative from the country naturally goes to a country mill. These people look on Spartanburg as I would look on New York City, as a great big corrupt assemblage of humanity where folks can't raise their children right." 47 But the people who went to the mills had decided to become operatives, and if coun- try families sought country mills, these might have been at a greater or shorter distance from their homes without con- siderably influencing their willingness to seek the industrial employment. Ordinarily, " it was possible then to locate a mill almost anywhere and strike a labor supply." 48 Labor was so abun- dant that it was an advantage, rather than an object. It has been suggested to the writer that the cotton mill era in the South was made possible by the pushing of railroads 45 Cotton Manufacturing Industry of U. S. (p. 143). 46 J. A. Brock, int., Anderson, S. C, Sept. 11, 1916. 47 W. J. Britton, int., Spartanburg, S. C, Sept. 5, 1916. 48 C. S. Morris, int., Salisbury, N. C, Sept. 1, 1916. 1 86 THE RISE OF COTTON MILLS IN THE SOUTH [292 up to the mountains, thus tapping pools of labor that flowed down into the Piedmont and lower country. Perhaps three considerations have prompted the thought: first, that cheap labor certainly contributed largely to the success of the fac- tories at the outset; second, that there was an important period of railroad building in the South Atlantic States just before and during the years in which the cotton fac- tories were erected; third, that many operatives came from the mountains. The number of mountaineers and " hill people" in the mill population of the South is large, but the curiously prevalent impression that all factory opera- tives were drawn from mountainous districts is mistaken. Labor in the years of the rise of cotton mills was scat- tered; it was available in nearly every part of the South; it was not dammed up in the mountains alone. It will be seen that the people came to the mills first from districts immediately surrounding the plants. Wagons carrying the entire household goods of the new help formed the means of conveyance. After a good many mills had supplied foci for the labor of their localities and some operatives had been trained, labor begun to be a little fluid. Workpeople moved from mill to mill. As more factories were estab- lished, the populations of more sections were attracted to industrial life, the total body of operatives became larger, the distance from one plant to another was less, informa- tion as to comparative conditions in mill villages was more easily obtained, and there developed what has been called " the floating element." But this mobile element, it is to be noted, was composed of cotton mill operatives, and not of people just from the land. 49 Not until late in the history 49 " Railroads to the mountains did not tap pools of labor. There was not much floating or flowing of labor until the mills had been long established" (Charles McDonald, int., Charlotte). President Baldwin, of the Louisville and Nashville Railroad, in the interview in the New York Herald already quoted, spoke of the part of the railroads in opening up a future for the South and dwelt at some length upon the poor whites and their entrance into the factories, but did not mention any assistance of railroads in forming an outlet for pent-up labor supplies. Cf. George B. Cowlan, The Undeveloped South. Search has failed to reveal a case in which, among the many 293^ THE LABOR FACTOR 1 87 of the Southern mills, as will be pointed out in more detail presently, did establishments get fresh labor from any dis- tance, and in these cases the stimulus to move came from the mills, not from the people. The iron filings had no greater impulse to move to the magnet than formerly ; more power had to be given to the attractive force. The mills had been building a good many years before it was neces- sary for them to solicit labor, and it proved hard work. 50 Labor from the mountains came a greater distance, per- haps, than that from the farming districts, but this was because there were no mills right in the mountains. It was essentially local, just as much as was the tenant labor. 51 For the Westminster Mill, in South Carolina, a very small affair owned by cotton planters, " the operatives con- sisted of seven young girls of the neighborhood who had never seen a cotton factory and one skilled operator, who trained them and attended to the card." 52 So far from bringing labor to mills, railroads may rather be said to have brought mills to labor. A newspaper correspondent wrote from the Piedmont : " Six years ago the country now trav- ersed by the Air Line Railroad was an almost unbroken wilderness. There were few people, little energy and no progress. Now there are towns and villages 1 all along the reasons urged for extending small up-country branch lines, that of releasing needed labor figured. 50 Cf. U. S. Census of Manufactures, 1900, Cotton Manufacture, p. 30. Even at this time, when the industry was " growing at a won- derful rate," the report was that " the help employed is chiefly local." 51 It is interesting to note that very recently, since hands have become scarce, a tendency to erect mills actually in the mountains has shown itself. 52 Blackman, p. 18. In this instance the operatives must have lived at their fathers' places, but this was unusual. The local character of the labor supply is frequently indicated in the provision made for the operatives' homes and general living — poor people from the vicinity came to and snuggled up against the mills like chicks under the protecting wings of the mother hen. The villages were like medieval hamlets clustered about a fortified castle. The factory was the provider. An officer of a small establishment which commenced operation in the seventies said : " Our labor is composed entirely of natives who have been educated to the business. They are very comfortably located, and have the free use of all the wood they require" (ibid., p. 8). The same had been true of the older Granite- ville factory all along (cf. ibid., p. 55). 1 88 THE RISE OF COTTON MILLS IN THE SOUTH [294 route, and the back country is rapidly being occupied by a thrifty and industrious population. In Pickens County, at Greenville and in Spartanburg, cotton factories have been built. . . . One hundred hands are now employed in the factory [Clifton], and, when the mill is finished, this num- ber will be increased to four hundred. The employes, with the exception of the superintendents in the various depart- ments, are all natives; there are no others on the pay rolls of the company." 53 In an account of the Huguenot factory, in the same State, it was said that " in the operation of the mill home labor is employed, the weavers being principally native women and girls, who with application soon become proficient in the art of operating the looms." 54 One evidence of the local origin of operatives and mill projectors alike was the mutual respect prevailing between management and workpeople. The owners of cotton mills did not look down upon their employes. They might and usually did recognize that the operatives were lacking in education, thrift, energy and property, and they applied themselves to alleviate these conditions, but always there was the knowledge that employer and employe were of the same origin, the same blood, and, not remotely, the same instincts. After-war struggles brought an intimacy through propinquity which in earlier years had been impossible. Men who were active in the opening of the cotton mill era in the South resent any suggestion, recognizing in it a slur somehow upon themselves, that the operatives were in- ferior people. 55 53 News and Courier, Charleston, May 21, 1881. 54 Baltimore Journal of Commerce and Manufacturers' Record, Oct. 28, 1882. 55 A very elderly gentleman, characteristic of the best the old South produced, had no sympathy with writers who are free in forming theories about the South, or who wish to make Southern problems seem distinctive. " Where did the first labor for the Greensboro mills come from?" he was asked. " From the mountain sections?" He replied with scorn: "That's all stuff! Magazine writers and such people, magazine writers, I say, come down here and spread such statements. The people came from right 'round here — some from this county, some from counties adjoining this. They were no paupers, either. They were the best kind of people. 295] THE LABOR FACTOR 1 89 The remarkable story of the Salisbury Mills, born in a religious and philanthropic impulse, has been told in a previous chapter. It goes without saying that this factory " was built for the home people," and it is interesting that the managers " never had anybody else in it." 56 For the Kershaw mill, "the employes came from right around Kershaw and are good citizens." 57 In many mills early conditions are reflected today. The Shelby Cotton Mill, it is recently reported, " employs . . . about three hundred operatives. They are ... in most cases native Cleveland county stock — good old Scotch-Irish and similar blood lines," 58 and it is said of the small Indian Creek Mill : "It gives employment to about sixty operatives and these workmen are native Lincoln county people." 59 Proximity was the chief determining factor in the source of labor. If there were not enough people in the immediate vicinity of a mill in the Piedmont to fill its needs, some operatives would be recruited from the higher country a little distance away. Thus of the Spartanburg mills: They went into the mills because it was a new thing, you know, and looked like a good thing." Asked then, " What did they do before going into the mills," he replied : " Farmed ! [with emphasis, as though anyone should know that]. Worked their farms! 'Course, many of them didn't own their places, were tenants. They helped themselves by going to the mills — got schools now and all that." This statement is mistaken in excluding the attraction of labor from the mountains, and overdraws the propertied character of the first operatives, but is significant in spirit. Though recognition was granted the poor whites belatedly, it was generous when it came (James Moorehead,. int., Greensboro). 56 O. D. Davis, int., Salisbury. Operatives came from within a radius of twenty-five miles (C. S. .Morns, int., Salisbury). Cf. Charlotte News, Textile Ed., 1917, with reference to this estab- lishment. 57 Columbia Record, Textile Ed., 191-6. 58 Charlotte News, Textile Ed., 1917. 59 Ibid. Instances are easily multiplied. Cf., respecting Clyde, Carolina, Great Falls, Raleigh, and Bladenboro mills, ibid. ; for ex- ample, " The greater part of the employes in these mills, particu- larly the older ones, came to the mills from the territory surrounding Rockingham. Many of them came from tenant farms where a year's livelihood was earned by the proverbial sweat of the brow, and much of it" (ibid.). Cf. also Columbia Record, Textile Ed., 1916. I9O THE RISE OF COTTON MILLS IN THE SOUTH [[296 " Labor first came from this immediate section, supple-^ merited by people from the mountains and foothills." 60 A great many workpeople for a South Carolina mill lo- cated in the country came from the four surrounding coun- ties, but " another big body of the new help came from the mountains of western North Carolina." 61 A woman who had been president of the Batesville fac- tory, in South Carolina, gave an interpretative account of the commencement of the mill period. " The section was desperately poor," she said. " The village of Greenville would have been called in the foothills. Farming returned hardly anything to put in the farmers' mouths. There were women and girls — many more women than men, be- cause the war had taken the men — whose lives were empty. The 'mills opened a vista before these; it was like finding a mine, you know. Most of the mills got local labor. In 1880 Camperdown, say, could draw no labor within a radius of half a dozen miles. This was also true of Batesville a few years later, before labor came from the foothills of the mountains. . . . After ten or fifteen years the labor of the localities was exhausted, and it was necessary to send to the mountains." 62 _,. It will be noticed presently that the pull of the field 60 J. A. Chapman, int., Spartanburg. Of a knitting mill at Union: "... fifty per cent of the operatives are natives of Piedmont South Carolina, the others from the mountains of North Carolina and East Tennessee" (Columbia Record, Textile Ed., 1916. Cf. ibid, with reference to the Ninety-Six Mill, and Charlotte News, Textile Ed., 1917, the account of Rhyne's establishments in Lincoln County, and such mills as Marion and Mayworth). 61 Columbia Record, Textile Ed., 1916. A newspaper summary of a survey as late as 1917 said : " Many mills will be found where there is not a man or woman employed except North Carolinians. . . . For the most part these employes come from the territory imme- diately surrounding the mills with additions from the mountains of the State" (Charlotte News, Textile Ed., 1917). 62 Mrs. M. P. Gridley, int., Greenville. Cf. Kohn, Cotton Mills of S. C, pp. 22-23. A Piedmont manufacturer said : " The labor at first was strictly local. Neighboring farm people came, probably from the same township or school district with that in which the mill was situated. Later it was necessary to send for labor from a distance — North Carolina and Tennessee. Labor at first was local- ized and did not move much" (J. B. Cleveland, int., Spartanburg). 297]] THE LABOR FACTOR I9I against the factory has tended to make cotton mill labor in recent years doubly difficult to secure. Not only have those readily willing to do factory work been drawn to the mills, but many who enter the mills return, to the farms. This is true more largely of tenant help than of mountain people. When the family pulls up stakes in the mountains and comes down to a mill village, the temptation to leave again is not so strong as in the case of a family which has moved in from a familiar farm a few miles away. 63 In rare instances mills at considerable distance from the mountains received their labor primarily from the moun- tain regions. It is said that labor did not come to the Char- lotte mills to any great extent from the adjoining country, but almost entirely from the mountains or foothills. Even for tenants, the farming was too good in the vicinity of the city to allow mill wages to tempt them away. 64 Before proceeding to other topics, it is convenient to speak of certain instances in which mills found difficulty in securing operatives. Usually, peculiar local circumstances were responsible for the inability of a factory to provide itself with employes. As has been made clear, the rule was an abundance of help. A writer who lived in the South in the years just preced- ing the first years of mill building assumes that a prime perplexity of the mills was the recruiting of operatives. Thus, speaking of the founding oi the industry in South Carolina, he says : " . . . the money had to be raised, largely with the assistance of the North; the companies formed, property bought, materials secured, homes for the opera- tives constructed, and last and most difficult of all, em- ployees obtained." He quotes approvingly a letter of an 63 It is said that " most of the operatives at Kannapolis (at any rate thirty miles from the mountains) — the permanent ones — come from the mountains. A good many come in from the surrounding farms to work a few months, and then go back to the farms" (H. W. Owen, int., Kannapolis, N. C, Jan. 6, 1917). 64 " If you will trace back through two or three generations, you will find that 75 per cent of the operatives in my mill are descended from people who came from the mountains" (Sterling Graydon, int., Charlotte). 192 THE RISE OF COTTON MILLS IN THE SOUTH [298 Englishman who was in the State during the Civil War, written to a mill president in 1908, saying that the rise of industry from agriculture seemed "all the more extraordi- nary because the State possesses no coal, and there was no superfluous population out of which to evolve mill hands." Paucity of labor was spoken of as an apparently insur- mountable difficulty. 65 Probably both of these writers meant that there were no laboring people accustomed to factory employment, that there was no industrial class from which to draw. 66 It will appear later that in a good many important instances, the projectors of cotton mills in the eighties failed to see the opportunity of utilizing the labor of, the poor whites, and looked for operatives from every other than this most plausible source. Contemporary estimates of facilities for establishing a cotton mill rarely voiced any doubt on the head of labor. 67 If any cause of scarcity of help may be termed general, it was a prejudice against factory work under bosses on the part of persons who had been, in however poor or suppo- sititious a fashion, their own masters. It might be supposed that objection to indoor employment and life in a mill vil- lage would be frequent with people with rural traditions. It must be remembered, however, that, their farming being at lowest ebb, they needed to take desperate remedies, and, moreover, dislike of a mill community could not be very strong in the face of the barrenness of country living. 65 Goldsmith, p. 7. 66 The Englishman concluded with the question : " For how could anyone see that the water power of the Alleghanies [Blue Ridge] could be converted into electric force, or that you could turn the clay-eating Cracker into a self-respecting mill hand?" (ibid.). Per- haps this correspondent's surprise at the success of the mills is ren- dered plainer by recalling the devastated condition of parts of South Carolina during and right after the war. Cf. Andrews, p. 34, as to Columbia in 1865. 67 No particular apprehension can be ascribed to the desire of a correspondent of a newspaper that, lest a single failure should occur in the development of mills in the State, all possible light should be thrown upon comparative costs of steam and water power and ad- vantages of location with respect to freight, health and labor (News and Courier, Charleston, April 29, 1881). 299] THE LABOR FACTOR 1 93 Where unwillingness to accept factory employment actually operated to keep some people out of a mill, the plant was in most cases located in a city and could depend upon the urban population for its help. But there might be difficulty even here. The President of the Atlanta Cotton Factory in 1880 was unable to get hands to run the mill at full capacity. He thought this was due to objection of women and girls to the class of work or to surroundings in the mill. A newspaper editorial, com- menting on the situation, thought that the girls, when they considered the matter, did not mind factory work, but that the absence of cottages for operatives was the cause of the dearth of labor, rents in the vicinity of the mill being high, and the pay being too small to allow of a long trip from home to plant. It was pointed out that if operatives were brought from the North, as was being contemplated, the same housing problem would confront them as the natives. However, if suitable cottages were built near the mill, the president " could obtain in Atlanta and the section of coun- try adjacent any number of women and girls who will not only gladly work, but will be eternally grateful to him for furnishing them the imeans of earning a comfortable and honest livelihood." 68 It was explained that the managers of the Charlotte Cot- ton Mills, employing fifty -five hands, nearly all skilled workers drawn from surrounding factories, " had been anxious to obtain their operatives among home people, but some insuperable prejudice seems to exist to the business, and not more than one or two, so far, have engaged." 69 Difficulty in getting labor was not more hindering in any 68 Daily Constitution, Atlanta, Jan. 2, 1880. In Charleston, which has had a bad reputation on the score of availability of labor, a mill in the last two decades has solved the problem by building an excel- lent village around the factory. Most of the operatives, it is true, have not come from Charleston, but from other parts of South Caro- lina and from other States. "We have always had enough help; we could start another mill right in our village and have labor enough for it" (Julius Koester, int., Charleston, S. C, Dec. 27, 1916). 69 Raleigh Observer, quoted in News and Courier, Charleston, Feb. 26, 1 881. 13 194 THE RI SE OF COTTON MILLS IN THE SOUTH [3OO city than in Charleston. Labor was the bete noire of the Charleston Manufacturing Company. And after the event transpired, it seemed that every one should have recognized that this would be the case. That a plant which was the perfect embodiment of the cotton mill campaign, as has been seen, should be built in Charleston, was natural ; that it could not succeed was almost as inevitable. Founded in idealism, it was not able to prosper in fact. Born in the minds and hearts of Charleston's best, it did not proceed from the determined and more silent cooperation of the whole community in the manner of other ventures which became permanent. It must be remembered, however, that this mill stood at the commencement of the cotton mill period in the South ; it in a sense marked the epoch. There were few traditions, either local, state or sectional, upon which to calculate. One of the leading projectors of the company, explaining that at the time the mill was started there were few females in cotton factories, described the unfortunate experience of this first enterprise in a large seaport : " It was considered belittling — oh ! very bad ! It was considered that for a girl to go into a cotton factory was just a step toward the most vulgar things. They used to talk about the girls working in mills up-country as if they were in places of grossest immorality. It was said to be the same as a bawdy house ; to let a girl go into a cotton factory was to make a prostitute of her." " How was it," he was asked, " that this was not under- stood by you gentlemen in launching the Charleston Manu- facturing Company ; that the women of the laboring class in Charleston would not go into the mill ? " The reply was undoubtedly the plain fact. " It never oc- curred to us," he said. "We canvassed the matter among ourselves." 70 "Our idea in starting the Charleston Manufacturing Company," said another of the original stockholders, "was that there were many people here who wanted work, needed 70 William M. Bird, int., Charleston. 30 1[] THE LABOR FACTOR 1 95 it. We found out they did not want it. They were ashamed to work in a factory. We thought it was going to help the town immensely. We found just the reverse. Instead of people flocking here, we had to take discarded labor from other mills and bring it here. We thought we could get enough people in Charleston to fill the mill, but we found the number here willing to work was very small. " 7X Local help failing, there was difficulty in obtaining hands from the up-country. " Some operatives from the Pied- mont objected to coming to Charleston in the summer time. They had seen many Charlestonians going through the Pied- mont region to the mountains for, they said, their health. This unfounded prejudice operated." 72 " Men were get- ting good pay in fertilizer works, on the wharves and in 71 W. P. Carrington, int., Charleston. Another said : " Young women looked upon factory work as lowering, and thought it was dangerous for young men and young women to work together as they must do. I thought this myself until I saw them working in fac- tories at the North; every girl would have two hundred or three hundred jealous girl eyes watching her; they were safer than in their own homes" (it is useful to remember that country people going to a country mill village found themselves surrounded by persons all in the same situation — there was nothing but the industrial commu- nity. In a city, however, even if the factory has its own cottages, operatives might feel censure of a non-industrial population. Fur- thermore, among city dwellers, however poor, women and girls were less accustomed to work than was true in the country, and would be more regardful of fancied social distinctions). The Charleston Manufacturing Company encountered trouble in recruiting labor that an older and smaller venture in the place did not, partly because it had been so much discussed and stood out in the public mind, im- pressed with a declaratively industrial character. " The Charleston Bagging Manufacturing Company," this informant continued, " mak- ing bagging from jute, used native labor, a hundred operatives or so. The bagging mill had been successful with female labor, and this encouraged us in our company." But the event as it transpired was not a complete surprise : " Still we understood that Charleston having had almost no factories, there would be prejudice against females working. But we thought this would wear off. We did not expect to get our labor force from Charleston at first. We thought the native labor would sift in gradually, and this proved to be true. Lockwood (the New England engineer who designed the mill) told truly when he explained that the first expert operatives to come to a new place were floating, and that it would require two or three years to get a steady, experienced force. In our impatience we looked upon the natural slowness in getting operatives, particu- larly women, as a terrible delay" (A. B. Murray, int., Charleston). " Ibid. I96 THE RISE OF COTTON MILLS IN THE SOUTH [3 02 industries/' said another, "and women did not need to work. There was not the press of life there was in a colder climate." 73 This story of trials seemed to he coming 1 to a bright con- clusion : " We brought the expert labor from the Piedmont, and the native population sifted in later, and took hold very nicely." But it was only passing into its final phase : " The two or three years following 1880 were bad ones for cotton mills. On August 31, 1886, the end of the company's fiscal year, the mill showed a small net profit. On the night of that day, the earthquake occurred. The railroads gave free transportation, and our operatives that had come from the up-country left. You couldn't have held them here with chains. Even the local operatives went away with the up- country operatives. We had a good working force at the time of the earthquake — after the earthquake, the only thing left was overhead charges. The officers were here, but the operatives had all disappeared." The prospects of this mill were never really promising afterwards. 74 Until very recent years, any class consciousness among Southern cotton mill operatives was induced by the prej- udice of the general community against them. The mill village, especially the company-owned town, has crystal- lized this sentiment, and politics and the lack of any other considerable industry in the South have made their unfor- tunate contributions. Dislike of the operatives' station is 73 George W. Williams, int., Charleston. He meant this to apply to the Charleston Manufacturing Company and to a successor, the Vesta Mill. " We were great phosphate people down here, and the laborers were distracted. But the leaders stuck to it [the enter- prise of a big mill in Charleston]. We went through three organi- zations" (F. Q. O'Neill, int., Charleston, S. C, Dec. 27, 1916). 74 A. B. Murray, int., Charleston. " The ground was in a tremor for several years after the earthquake. It took two years to reor- ganize the plant. We had to send to the up-country for skilled operatives." And another concluded : " We thought that if a mill could pay in the up-country, it would pay to build a mill in a large center like Charleston. The labor trouble was the chief reason for the failure." He felt that had the attempt been made fifteen years later, after 10-cent stores and dry goods stores had begun to employ women, the mill might have succeeded (William M. Bird, int., Charleston). 3O3] THE LABOR FACTOR 1 97 undoubtedly greater at present than in the years when the mills were building. A just statement of the facts as they prevailed forty years ago is the following : " There was some prejudice against operatives on the part of others, but it did not show itself. So far as speaking to them cordially, etc., was concerned, they were received." 75 It has been said that mill managements in the eighties showed none of the spirit of neglect of the poor whites that had character- ized the period before the Civil War, and this attitude was not persuaded merely by business motives. It is probable that no great development could have taken place, calling for enlistment of the service of thousands in the population, without some objection against workers in the new industry becoming evident. But in the case of the cotton mill opera- tives this was at a minimum. The South was too much in earnest in its work to question the social status of those who were factors in its accomplishment ; work was too scarce to permit of a choice influenced by popular dislike or esteem; the South of the eighties was twenty years re- moved in time and many more years removed in experience from the older South of an idle class ; and last, the poor whites by entering the mills tended to throw off the atmos- phere of unnoticed destitution in which they had been en- veloped before they had beeri given a useful outlet for their services. If their situation was not envied by some, by the majority it was not despised; if they were looked upon as a class with disfavor, this was not on the surface, and nobody had time to bother with such notions. 76 A part of the prejudice against operatives, if it may be 75 Charles McDonald, int., Charlotte. " Many cf those who be- came operatives had owned their own land, and when misfortune overtook them, in the shape of bad crops and debt, came to the cotton mills" (M. L. Bonham, int., Anderson). 76 The usual sentiment is illustrated by some words of Hammett: " It is clear that what the South needs more than anything else is diversified labor, and to realize that to labor is respectable, and to be idle is not respectable. With all the unemployed water power and other natural facilities one of the main industries should be to con- vert into goods a part of the cotton produced by the soil" (quoted in Manufacturers' Record, Baltimore, Feb. 1, 1881). I98 THE RISE OF COTTON MILLS IN THE SOUTH [3O4 called such, perhaps took rise in objection to mills on the part of rural communities. This was a different thing from the social discrimination spoken of above. It was a con- flict between occupations, not between elements in the popu- lation. " Right at first," said one informant, " there was a good deal of opposition to the mills on the part of farmers, and this made labor hard to get." 77 Another asserted: " Our white people were accustomed to be their own mas- ters. They had not lived in great groups or worked under bosses and that kind of thing." He remembered that this hindered the recruiting of local hands. 78 In the same issue of a South Carolina newspaper saying that " Cedar Creek . . . affords ample water power at this point [Society Hill] for a factory," and that "there is plenty of labor," it was told that the management of the Camperdown Mills, at Greenville, was finding it impossible to get two hundred and fifty extra hands needed to run the plant at night. This was due, it was explained, to the pres- ence of disorderly women in the neighborhood, who were proposed to be used as operatives, and who could not be got rid of. Circulars were distributed all along the line of the Atlanta and Charlotte Railroad, and in other directions. The mill offered free transportation and a dollar a day for all time lost by prospective operatives, but, after an ex- penditure of $500, no more workers were in the factory, and it was regretted that the mills were " receiving a large accumulation of orders it will be impossible for them to fill." 70 The scarcity of labor which was experienced twenty-five years later was of an entirely different character from the 77 Marshall Orr, int., Anderson, S. C, Sept. 10, 1916. 78 M. L. Bonham, int., Anderson. 79 News and Courier, Charleston, Feb. 23, 1881. Without men- tion of the reason for it, this fact of scarcity of help for the Camper- down Mills was recalled to Mrs. Gridley, who confirmed the report of years before : " If Camperdown sought in vain to get 250 opera- tives in the early eighties, it must have been because the mill had a rough class of help. The bad reputation the labor force earned kept away the mountain people the mill was trying to attract" (Mrs. M. P. Gridley, int., Greenville). 305] THE LABOR FACTOR 1 99 scattered instances here noted. After 1900 it became a problem of inore or less general concern ; many mills bad been built, some of them very large, and the condition of the body of the poor whites 1 was somewhat better than in the earlier period, not a little by agency of the cotton manu- facturing industry itself. In spite of a degree of optimism, difficulty was presaged in an address before the Southern Cotton Spinners Association in 1903 : " Now in regard to an insufficient supply of native-born help. This 'may !be true in localities, but it has been the experience of all manu- facturing centres that the building of the mills has eventu- ally drawn, in close proximity, people from the country and outlying districts, and it is not worth while to consider this matter as fatal to the future increase of spindleage here." The speaker thought that " even when our native country help is exhausted ... if it be true that cotton manufac- turing may decline in our sister countries, there will be op- portunities for skilled employes from those countries to be obtained. We should not cross this bridge until we come to it." 80 It has been said that the projectors of cotton mills in the South not only welcomed the native whites as workpeople, 80 Proceedings 7th Annual Convention, address of E. W. Thomas, p. 149 ff. Tompkins in 1900 had foreseen that the objection to night work would take care of itself, for "as mills increase labor will become scarcer until there will be no available labor for night work " (Labor Legislation, p. 4). In 1904 a Georgian speaking to Georgians said : " Why one section — a comparatively old one — is short of labor, is not my province to discuss. It is simply a question and no theory that we have confronting us." He thought that immigration agents ought to draw workers from Italy, in about the same geographical latitude with Georgia, to fill the domestic " vacuum of labor," and wanted Georgia represented at the St. Louis World's Fair by a solicitor who would operate in conjunction with real estate firms and the railroads in bringing home-seekers to the State (Georgia Industrial Assn., proceed. 4th Annual Convention, address of Hon. I. C. Wade, p. 34 ff.). The convention appointed a committee "to urge the establishment of a Department of Immigration by the State of Georgia" (proceed., p. 33). In 1907 Mr. Kohn wrote that " there is plenty of capital, energy, enthusiasm, business ability, water power and cotton for South Carolina to have very many more spindles than she now has. The one difficulty is that of securing additional labor" (Cotton Mills of S. C, p. 60). Cf. ibid., p. 63, and T. W. Uttley, p. 68. 200 THE RISE OF COTTON MILLS IN THE SOUTH [^306 but planned factories in many instances partly with the express purpose of affording them employment. There were some cases, however, in which the possibility of em- ploying the poor whites was curiously overlooked, and operatives were sought or proposed to be sought outside of the South. It cannot be said for this strange neglect of the obvious opportunity of utilizing the Southern population that the poor whites were "out of sight, out of mind." They were very much in evidence everywhere, were mutely appealing for assistance and notice; even asking, if one pleases, to be exploited. The disposition to seek opera- tives outside of the South, so far as it showed itself, was fostered by three circumstances : first, the feeling that ex- perienced workers must be found to start the industry ; second, the desire to weaken the negro by increasing the white population ; third, new and prospective cotton manu- facturers fell in easily with the prevalent plans of agricul- tural interests to secure immigration to the section. How earnestly Senator James, of Rhode Island, plead for the establishment of cotton mills by Southern planters, and how he urged that the needy native white people be em- ployed in the factories, has been noticed. Eager, however, to leave no stone unturned in proving the plausibility of his proposal and in answering especially the question as to how help was to be gotten, he declared that " Even should the planter, who goes into the manufacture of cotton, find it necessary to import his operatives from Europe at his own expense, he would still be a great gainer by the transac- tion." He showed how, by saving one cent per pound on raw cotton — the cost of transporting the staple to a North- ern mill — the Southern manufacturer would be able to de- fray the charges of bringing over English operatives, and have a considerable surplus to his credit. 81 This suggestion seems to have been tried in practice, for it is said that a superintendent of the Augusta Factory, probably in the seventies, brought a boatload of operatives from Scotland. 81 Cf. DeBow, vol. i, p. 238 ff. 307] THE LABOR FACTOR 201 The mills of Augusta still have English and Scotch people in them, likely descendants of these immigrants. 82 Foreign- born operatives transplanted to Augusta supplied many of the mills throughout the South, particularly in the Caro- linas, with skilled superintendents and overseers. A writer at the close of the Civil War would have been right in including the whole South in an observation made as to Charleston, that it wished immigrants from Europe rather than newcomers from the North. " Immigration is held to be the panacea for all present evils and troubles. One of the representatives elect from this city will make strong efforts to secure legislative action at the coming ses- sion of the General Assembly in favor of a bill granting State aid to foreign immigrants. The Yankee is not wanted here, except by the enlightened few ; but Germans who will consent to take a secondary position will be welcomed." 83 The extraordinary scheme of " The Region of the Savan- nah Colonization Association " for bringing New England operatives to cotton mills in the South is worth mention be- cause, with its preposterousness, it shows the thought in the minds of some. It was set forth in 1882, and never got beyond the stage of advocacy. It was a promoters' plan for combining a pastime for rich men of leisure with a health resort and an industrial community. An agency of the Association explained that the Region of the Savannah (the entire States, apparently, of Georgia and South Caro- lina), offered in its piney woods and mild and dry climate the only relief from pulmonary diseases. New England cotton mill operatives who, left in Northern factories, were destined to lose their ability to work or would even die, might come to this salubrious district and regain their health by coupling farm work with factory attendance. In New 82 George T. Lynch, int., Augusta, Ga., Dec. 30, 1916. Compare the proposal of " Hanover " that English operatives be brought to Richmond (Daily Dispatch, Jan. 14, 1880). 83 " ' The only way in which we can control the labor of the free negro is to bring him in competition with the white laborer,' is the language of scores of men." By " the white laborer " the native white was not meant (Andrews, pp. 207-208). 202 THE RISE OF COTTON MILLS IN THE SOUTH [^08 England were found, besides the broken-down mill hands, retired business and professional men of means, for whom the Northern climate from December to May was too severe for comfort. It was proposed that some of these wealthy- invalids should buy a few thousands of acres in the Region of the Savannah, build forty or fifty neat but inexpensive houses on the tract, and rent those not occupied by them- selves (they would be there to give character to the project), to sick New England operatives, and to pleasure-seekers wishing a wintering place in the South. " This would give a nucleus for a permanent settlement, and in a very short time an industrial community would grow up about it." A correspondent of the Savannah Morning News was quoted approvingly, this writer proposing that each family might have a forty-acre farm and divide its labor between agricul- ture and a cotton mill which would be centrally situated. It was asserted by the projectors of this scheme that it would make a return of 100 per cent on the capital invested. 84 84 Gannon, p. 8 ff. " The Region of the Savannah Colonization Association " was built on the constitution of the defunct American Colonizing Company, founded in 1818, for the furtherance of trade, it was declared, between the South Atlantic States and the West Indies, the west coast of Africa and the Brazils, the principal Amer- ican depot being at Charleston, Port Royal or Savannah ; the build- ing of small cotton mills, to be operated by a transplanted New Eng- land industrial population, would be linked with the construction of small ships to carry the product of these factories. Another inspi- ration to the project was President Grevy's system of cooperation; the success of the young Meaux workman who in one year built 250 houses on a tract of land of an old marquis and started his colony at a cost of 240,000 francs with considerable profit to his fellow-enterprisers — he began with only ten 5-franc pieces of his own — was instanced. The aims of the Savannah Association were to be accomplished largely through dissemination of information. The pamphlet was published as propaganda by Gannon and Mayhew, 176 Tremont Street, Boston, who were general agents for the organi- zation. The Southern Land, Emigration and Improvement Com- pany, a New York organization designed to encourage immigration to the South, said in its prospectus : " That the South now offers greater inducements to capital, enterprise and intelligent industry than any other quarter of the globe, is beyond question to those who are informed upon the subject. . . . The Southern people them- selves are thoroughly awakened at last to the fact of the abundance of their resources. They are putting forth every energy to secure their share of the overflowing tides of population from the old 3O9] THE LABOR FACTOR 203 A journal which championed the South's interests, la- menting the failure of immigrants to go to that section, de- clared that " if the South is to be built up, her unoccupied lands turned to the uses of civilization, her streams become the seats of great 'manufacturing enterprises, and all her natural advantages made to bear material development, there must be a systematic effort to induce immigration. Railroads, States, private individuals, are all alike inter- ested in this; and it behooves all to work persistently to accomplish it." 85 It seems likely that immigrants, especially where for- eigners, were not often sought by the South for industrial workers. Agricultural interests were uppermost in the minds of the people, and schemes to supplant the free negro were, for the time being, as natural as they were imprac- ticable. 86 Even where immigrants had been in mechanical pursuits in their own countries, their usefulness in industry might be overlooked. 87 world . . ." (quoted in Baltimore Journal of Commerce and Manu- facturers' Record, Aug. 5, 1882). The alliance of the South Caro- lina Railroad with the Georgia Railroad and the Central Railway was looked upon as bringing to Charleston " increased business, direct trade with Europe and white immigration" (News and Courier, Charleston, April 14, 1881). 85 Baltimore Journal of Commerce and Manufacturers' Record, June 24, 1882. 86 An editorial on the benefits of immigration to South Carolina placed the whole stress upon agriculture. Nothing was said about using immigrants in cotton mills, though Charleston's advantage in being able to get them from German ports at two-thirds of the charge if taken to New York, was mentioned (News and Courier, Charleston, May 20, 1881). A week later the same paper com- mended the commissioner of immigration for steering foreigners away from "towns or cities where they would be a burden to them- selves and those around them" (June 17, 1881). A like omission of cotton manufacturing in stating the reasons for immigration is seen in an address of the Georgia Commissioner of Land and Immi- gration to the State legislature (quoted in ibid., Aug. 5, 1881). Cf. editorial in Observer, Raleigh, April 10, 1881. 87 Sixteen families — Poles, Germans and Austrians — in 1881 passed through Charleston on their way to Columbia. " They have no property, and are uncertain of their final destination. They are generally mechanics, but claim to know something about farming, and are willing to do anything to make a good and honest living." 204 THE RISE OF COTTON MILLS IN THE SOUTH [3 10 In conclusion of these references to advocacy of immi- gration as apart from the needs of cotton mills, it is inter- esting to notice a jeu d'esprit of " Henry LeBiank," written under date of July 13, 1893, twelve years in the future, predicting failure of plans for European immigration to South Carolina, the foreigners being unsuited to the climate, crops and mode of living, and adding in a postscript : " It would do you good to see the immense number of factories at Columbia, down by what was an old ditch, but now a splendid canal. Spartanburg has over 30,000 population, and seven railroads centre there." 88 Instances in which immigrants were looked for as cot- ton mill operatives show the newness of the South to indus- trialism, the suddenness with which an urgent program was embraced. How foreign manufacturing was to the South's past, how novel a departure it represented in the minds of mill projectors, comes out in the rare cases in which native whites were not considered as operatives ; such an opportunity might not even be debated, but it was thought that new wine was to be put into new wineskins. Thus in advocating the building of a mill near Winnsboro, South Carolina, in a county in which poor whites were plentiful, these were overlooked as industrial workers and, for that matter, as agricultural laborers. "If we can do no better let us spin a hundred bales at first. . . . We believe There was no mention of directing them toward cotton mills (News and Courier, Charleston, May 10, 1881). Relatively few immigrants actually came — seventy-four persons colonized in South Carolina in a typical week — and most of them were placed with farmers (see ibid., March 23 and July 1, 1881). Despite every demonstration of failure, projects for bringing in foreigners to become cotton farmers would not die. As late as 1908 Tompkins declared : " Every condi- tion in the cotton growing States is favorable for the European farmer who wants to emigrate. . . . Such a movement would go further than any other to insure a cotton supply adequate to the world's demand and at a reasonable price" (Cotton Growing, p. 7). And earlier he had urged that " the New England Cotton Manufac- turers' Association turn itself into an emigration society pro tern, for the purpose of securing the occupation of the Southern cotton land" (The Storing and Marketing of Cotton, reprint from Trans- actions of New England Cotton Manufacturers' Assn., vol. 77, p. i 9 ff.). 88 News and Courier, Charleston, June 15, 1881. 3 1 1] THE LABOR FACTOR 205 there is money enough in the county, here and there, to make at least a modest beginning, so as to attract outside capital. Shall the effort be made, or shall other counties, once far behind us in wealth, take the lead and rapidly out- strip us? We want white immigrants. Bring the mills here and they will come. Colored labor will raise the cotton, and white immigrants will convert it into yarn." 89 A news- paper in another community concluded that the freed ne- groes had done little to better their condition and had, more- over, kept away skilled immigrants ; " . . . remove at least half the negro labor from the State, then it [skilled immi- grant labor] will come, and with it capital which will seek investment in our manufacturing interests, and at once put us on the highway to wealth, power and happiness." 90 Another editor, in contrast to these less thoughtful con- temporaries, expressed sanely the better judgment in op- posing wholesale immigration on the ground that there were needy people in the South to be thought for first, and because the section was in no position to invite new-comers to share in her uncertain lot : " We have many worthy native people of the more indigent classes who must be pro- vided for in some way before we talk of hurrying those here who, at the best, may take the bread out of the mouths of our own people. We are by no means opposed to legiti- mate immigration, but we are very far indeed from seeing the good sense of bringing upon ourselves or our unhappy visitors the cruel lot of being thrown into Southern com- 89 Winnsboro News, quoted in News and Courier, Charleston, Feb. 8, 1881, 90 Pickens Sentinel, quoted in News and Courier, Charleston, Feb. 3, 1881. An opportunity of securing skilled textile operatives among immigrants from Alsace-Lorraine was evidently received gladly. J. H. Diss DeBar, of New York, directing a movement to bring over foreigners, had written to the president of the Atlanta Factory, for facts as to the employment of any immigrants that might be sent down. Other Georgia mills were urged to communicate with this agent with information as to wages, rent and other conditions affect- ing work in the manufactories (Daily Constitution, Atlanta, March 24, 1880). 206 THE RISE OF COTTON MILLS IN THE SOUTH [312 munities without bread and without any hope of employ- ment." 91 The futility of attempts to attract immigrants began to be seen in South Carolina early in the eighties, a newspaper declaring "We hope the State will abolish the office of superintendent of immigration. It is ... a worse than useless expense." 92 Twenty-five years later, following 91 Columbia Register, quoted in News and Courier, Charleston, Feb. 3, 1881. One long acquainted with the State's politics believed the motive of supplying cotton mill operatives was not important, that "back of the efforts of South Carolina, through Commissioner Boykin's office, to secure immigration, was the desire to get rid of the negro and to bring in whites to take his place." When Boykin left office, another commissioner was appointed. " Then there were some years when there was no commissioner of agriculture or immi- gration. It was largely a matter of politics" (M. L. Bonham, int., Anderson). As to the purpose to oust the negro, the comment of a German-language newspaper is indicative, especially since Germans were particularly sought: "Col. Boykin, the immigration commis- sioner, has returned from New York, and reports that he is able to get in Castle Garden as many immigrants for South Carolina as are wanted. He seems to be intent chiefly upon getting laborers who are able to take the place of the negroes" (Deutsche Zeitung, Charleston, April 25, 1881). Cf. Daily Constitution, Atlanta, Jan. 31, 1880. With the negro question in mind, Henry W. Grady said: " Companies of immigrants sent down from the sturdy settlers at the North will solve the Southern problem . . ." (Dyer, in New South, p. 139). Cf. State of S. C, Fourth Annual Report of Com- missioner of Agriculture, Commerce and Immigration, p. 4, and pre- ceding reports ; DeBow, vol. ii, p. 127. Frequently immigration to the South from other parts of this country was in mind ; cf. DeBow, ibid., and quotation from United States Economist, in Baltimore Journal of Commerce and Manufacturers' Record, Sept. 30, 1882. 92 Abbeville Press and Banner, quoted in News and Courier, Charleston, Nov. 25, 1881. _ Cf. State of S. C, ibid., p. 32. A pam- phlet issued by the immigration commissioner was attacked as sophomoric, unfair in claiming too much, and generally " a disgrace to the State" (News and Courier, ibid.). In 1894 the editor of a publication that had done much to encourage immigration admitted that the South had been " in no condition to invite immigration. . . . All efforts to attract settlers to this section could only prove futile. The time was not ripe" (Edmonds, p. 29 ff.). The most famous effort to recruit foreign immigrants as operatives for Southern mills was the episode of the Wittekind, which, even without hindrance from the federal authorities, was so unsuccessful that it would hardly have been followed up. South Carolina planters were inter- ested in securing farm hands by the venture, and combined with manufacturers in a fund which was utilized through the State immi- gration commissioner. The North German Lloyd steamer Wittekind, in two trips to Charleston, in November, 1906, and February, 1907, brought a few hundred passengers, principally Belgians, Austrians 313] THE LABOR FACTOR 207 South Carolina's unsuccessful effort at importing immi- grants, the Georgia Farmers' Union " unanimously voted against foreign immigration, because it would bring unde- sirable people who would compete with the Georgians for factory labor and would raise so much cotton that it would lower the price." 93 This is a far cry from the disposition remarked in a few of the early mill projectors to overlook the opportunity, not to mention duty, to employ the native whites in the textile industry. The Southern mills have almost no foreigners. Just oc- casionally a few trickle in by chance. " Once in a while," said a superintendent, "we have a spasm of French Cana- dians and Poles. They are not imported, nobody goes after them. They don't stay very long, and come only two or three families together." 94 and Galicians. It is likely that disappointment of disingenuous prospective employers at frustration of their plans by the central government has colored judgment of the results of the experiment, but it appears all in all that the new-comers were not so well con- tent as to form a satisfied nucleus which would automatically attract relatives in succeeding years. Mr. Gadsden, a representative of South Carolina business men, who investigated the matter in Eu- rope, wisely reported : " Our efforts have been almost entirely ex- pended in inducing immigrants to come to the South, and we have thought little or nothing of how the immigrant is to be treated after the immigrant has come in our midst; ... we have entirely over- looked our industrial conditions, namely, that the wage scale through- out the South is based on negro labor . . . our attitude throughout the South toward the white labor will have to be materially altered before we can expect to have the immigrant satisfied to remain as a laborer with us" (quoted in Hart, pp. 52-53). On the whole matter see State of S. C., ibid. ; a good deal of reading between the lines is necessary. Cf. also Goldsmith, p. 10, and Kohn, Cotton Mills of S. C, p. 24. The action of South Carolina was preceded by agitation in manufacturers' associations in other Southern States looking toward immigration. A speaker before the Georgia Indus- trial Association in 1901 asserted : " There is room in Georgia for several hundred thousand competent white foreigners." Three years later it was being urged that practical steps be taken (Proceed. Fourth Annual Convention, p. 13 ff.). 93 Hart, p. 54. A " Southern writer " was quoted as saying that " The temptation of cheap alien labor from abroad is obvious as one of the ways in which a home population may be dispossessed. When it ceases to fill the rank and file with its own sons ... it ceases to be master ... of the country" (ibid., p. 55). 94 George T. Lynch, int., Augusta. Cf. Thompson, p. 30, and Kohn, Cotton Mills of S. C, p. 24. 208 THE RISE OF COTTON MILLS IN THE SOUTH [314 It has been seen that despite shortage of operatives in peculiar individual instances and ill-advised efforts to at- tract immigrants to compensate for an actual or anticipated scarcity of immediately available labor, the rule at the open- ing of the cotton mill era was an abundance of local help. The mere erection of a factory was sufficient inducement to the gathering of a working force. The problem was rather to secure the plant than the operatives. This condi- tion lasted for about twenty years. " Labor for the early factories came from the localities — 90 per cent of it. But after 1900, when there was a madness of mill building, they began to pull labor from a distance of 250 miles. Whereas people had before straggled in at will, the mills now com- menced concentrated efforts to get them out of the moun- tains." 95 In the active years preceding the panic of 1907 this practice became more frequent. A superintendent in the up-country gave his experience: "The first labor for the Spartan Mills came from the surrounding country, and was supplemented soon by people from the mountains of North Carolina and Tennessee. In 1905-6 and 1906-7 there was a scarcity of labor. Spartanburg mills sent agents into the mountains to bring out help, the mill ad- vancing railway fares of operatives. In this way from 1905 to 1907, 171 families were brought to this mill." 08 The advantage of the cotton mill village as contrasted with the mountain farm, which had earlier been too patent to re- quire statement, began to be carefully explained in dodgers distributed through highland districts, or were set forth 95 H. R. Buist, int., Charleston. 96 W. J. Britton, int., Spartanburg. He was much disappointed in the results of the soliciting system, and said that of the 171 families brought to the mill village, only 10 remained. " I would rather have half a dozen families that paid their own way to the mill than fifty families brought here." Commenting on the necessity of scout- ing for labor, Mr. Copeland declares " The growth of the industry has taken away the advantage which was its chief asset." In the period referred to employers bid against each other for help, so that wages were raised nearly one-fourth. Almost all the mills were reported to be short of their full complement of operatives ; " for the time being the South had built more mills than it had labor to operate" (p. 46 ff . ) . X 315] THE LABOR FACTOR 20O, by satisfied operatives taken along by agents as bait. 97 Ex- haustion of the readily available supply of poor whites is further indicated in efforts since the Great War to attract workers from the eastern sections, which lowland tenants, in their full knowledge of the needs and opportunities of the mills, had already been constructively solicited, and in the building of mills actually in the mountain districts. 88 It has been seen how slavery was largely responsible for crushing the early manufactures which arose in the South and prevented recovery of industries in the section. At first blush it seems strange that negroes, in the period before the Civil War in which manufactures were at lowest ebb, should have been employed in cotton mills. It might be objected that slavery, so far from being an enemy to the textile industry, assisted such factories as were in operation. The point, however, is easily cleared up when it is re- membered that the old mills were generally very small, scat- tered, unstandardized, and made the rudest products, and when it is considered that managers of factories, many of them planters, might naturally use slaves whom they owned or could hire cheaply rather than whites who were less de- pendent and who must be better paid and differently treated. There was less difficulty in adapting slaves to the work of the ante-bellum cotton mills than in employing free negroes in later years, because processes were more elementary and because many slaves, especially women and girls, had been taught something of the textile art in domestic industry on the plantations. Thus the finding of negroes in mills which anticipated the real development of cotton manufactures in the South is to be considered rather a proof of the depress- ing effect of slavery upon the industry than as supporting a contrary argument. It may be believed that most of those who before the War advocated the use of negroes in cotton mills held no very- hopeful or plausible economic philosophy. If they really 97 Cf . Kohn, Cotton Mills of S. C, p. 23, and John C. Campbell, From Mountain Cabin to Cotton Mill, especially p. 5. 98 S. N. Boyce and J. Lee Robinson, int., Gastonia. 14 210 THE RISE OF COTTON MILLS IN THE SOUTH [316 understood the situation, the proposal to employ slaves must have been acknowledged as a makeshift; if they did not, it was none the less a fanciful dream. In most cases there must have been no further thought behind the use of negroes than that it was convenient, cheap and sufficient for the limited project in hand. Certainly William Gregg made a sound diagnosis of the South's ailments, and showed more foresight in economic matters, it may be thought, than any other Southerner of his day. Some surprise, therefore, may attach to the state- ment that he advocated the operation of cotton mills with negro labor. The explanation lies in two facts : first, though he had visited the Pennsylvania and New England fac- tories, there was nothing in the South to 1 compare with them, and it would have taken an imagination and faith superior even to his to transcend the numbing effect of his dominantly agricultural surroundings and reach beyond them to visualize the necessary conditions of industry as afterwards proved in history ; and second, seeing the great difficulties in the way of his proposals, statesmanship prompted him to utilize any means that offered to make a beginning. There is something pathetic in the tone of his appeal, born almost of exasperation : Surely there is nothing in cotton spinning that can poison the atmosphere of South-Carolina. Why not spin as well as plant cot- ton? The same hand that attends a gin may work a carding ma- chine. The girl who is capable of making thread, on a country spinning wheel, may do the same with equal facility, on the throstle frame. The woman who can warp the thread and weave it, on a common loom, may soon be taught to do the same, on a power loom; and so with all the departments, from the raw cotton to the cloth, experience has proved that any child, white or black, of ordinary capacity, may be taught, in a few weeks, to be expert in any part of a cotton factory; moreover, all overseers who have experience in the matter, give a decided preference to blacks as operatives." 99 Domestic Industry, p. 21. He had not only the sight of South- ern mills of his time operating with negroes, but he relied upon the judgment of a well-known authority who understood the English and the American industry, quoting James Montgomery to the effect that " If the experiment of slave labor succeed in factories as is confidently expected, the cost of manufacturing the cotton into cloth 317] THE LABOR FACTOR 211 The Saluda Factory, near Columbia, was reported in the early fifties to be operating successfully with slave labor, the negroes being mostly owned by the company. The en- terprise was of $100,000 capital, and employed 128 opera- tives, including children ; there were 5000 spindles and 120 looms, the product being heavy brown shirting and South- ern stripe. " The superintendent is decidedly of the opin- ion that slave labor is cheaper for cotton manufacture than free white labor. The average cost per annum of those employed in this mill, he says, does not exceed $75. Slaves not sufficiently strong to work in the cotton fields can attend to the looms and spindles in the cotton mills. . . .'" The average cost of a white operative per year was said to be $116, so that those using slaves, it was claimed, enjoyed " over thirty per cent saved in the cost of labor alone." 100 will be much less there [in the South] than anywhere else, so that it will not be surprising if in the course of a few years, those Southern factories should manufacture coarse cotton goods, and sell them in the public markets, at one-half the price, at which they are manu- factured in England. There are several cotton factories in Ten- nessee operated entirely by slave labor, there not being a white man in the mill but the superintendent . . ." (ibid.). Montgomery in- stanced other cases of actual or intended use of negro labor at Rich- mond and Petersburg, and went so far as to say " there is every reason to believe that it is better adapted to the manufactory than to the field, and that the negro character is susceptible of a high degree of manufacturing cultivation. . . . This kind of labor will be much cheaper, and far more certain and controllable. The manu- facturer will have nothing to do with strikes, or other interruptions that frequently produce serious delay and loss to the employer " (A Practical Detail of the Cotton Manufacture of the United States of America (1840), p. 192). He estimated the total expense for the services of the best negro workmen for a whole year at $170, females and young men being cheaper. Gregg's quotation from Montgomery is not quite accurate, though perfectly exact in spirit. Further ex- tenuation is brought to Gregg in the statement of Mr. Kohn that " The history of the early efforts of the industry in this State indi- cate that slave labor was very largely used" (Cotton Mills of S. C, p. 24). Buckingham, writing three years before Gregg, implied that it would be the natural thing to use negroes at least equally with whites (Slave States of America, vol. i, p. 171). The Rocky Mount Mill, in North Carolina, employed negroes from 1820 to 1851 on the coarser yarns, most of the product going to country merchants near the factory, but some to the Philadelphia market. In 1849 negroes were the only operatives (Thompson, pp. 250-251). 100 The health of the blacks in the mill was said to be better than that of whites in the same occupations (quoted from New York 212 THE RISE OF COTTON MILLS IN THE SOUTH [3 1 8 DeBow approved the recommendation of a Tennesseean that slave labor be applied to the manufacture of cotton and wool throughout the South, " such labor having been found most advantageous wherever adopted." 101 It may- Herald, in DeBow, vol. ii, p. 127, note). Another observer said that the experienced white overseers from the North, at first prejudiced against the slave labor, testified to its equal efficiency and even supe- riority in many respects as compared with white. The negroes were tested out at spinning, but later learned to weave, and turned out full quantity of cloth. " The resources of the South are great, and it should be gratifying to all who view these facts, with the eye of a statesman and philanthropist, that the sources of profitable em- ployment and support to our rapidly-increasing African labor are illimitable, and must remove all motives for emigration to other countries" (ibid., vol. i, p. 232). In 1847 the plant was declared to have done a "fine business" for three years previous (Columbia Telegraph, quoted in Kohn, Cotton Mills of S. C, p. 18). Other reports of this mill do not paint so bright a picture. The son of a man who relinquished the superintendency in 1838 said the mill was owned by slaveholders who chose to use some of their negroes in this way — they were planters first and manufacturers second. The negro labor was not successful (Charles McDonald, int., Charlotte). One manager of the mill was reported as saying that slave labor failed there because of the malarial condition of the neighborhood and because the negroes' fingers were clumsy (William Banks, int., Columbia). Mr. Kohn states that the factory was operated largely by slave labor until the close of the Civil War, when whites were installed, and quotes Hammond's Handbook to the effect that 90 slaves in charge of a white overseer were " capable of learning within reasonable limits" (Cotton Mills of S. C, p. 16). This fac- tory, perhaps the best known of those employing negro operatives, is said to have been burned by Federal troops entering Columbia. The ruins, across the river, about three miles above the city, are still to be seen, flanked by a grove on a small plateau overlooking the stream. The foundations and maybe one or two stories were of stone. The race, now empty of water, is stone-lined and deep, and huge wooden beams and parts of the rude shafting remain in the wheel-pit. The dam flung across the river seems still in tolerable condition, though the sluice is widened by years of neglect. Mr. Kohn says the establishment was hampered by lack of capital, and quotes Gregg to the effect that the capitalization of the plant was not more than sufficient to pay for the expensive dam (Cotton Mills of S. C, p. 17). Except in the weaving department, blacks were employed in the DeKalb factory in South Carolina for several years, thirty belonging to the company, which thought they compared favorably with white operatives. Wages of negroes were 1854 cents a day and board ; whites who succeeded them, exclusive of the weavers, were given from 13 to 36 cents a day. References to wages in old mills in Georgia and Alabama seem to indicate that there must have been some negro employes (Ingle, pp. 75-76). Cf. Kohn, ibid., p. 16. 101 Industrial Resources, vol. ii, p. 124. 319] THE LABOR FACTOR 213 be concluded that slave operatives in ante-bellum mills were common. The attempt of Alexander and Haskell, both perfectly familiar with the negro and the economic condi- tions prevailing in South Carolina at that time, to employ blacks in the Congaree Mill, the first erected in Columbia after the war, must indicate that they were repeating a familiar practice. 102 It will be seen presently that the rare later efforts to use negroes were considered experimental and watched with doubt by outsiders. It must be observed that, partly as a suggestion from pre- war usages, partly as an evidence of disposition sometimes shown to overlook the labor supply so naturally found in the poor whites, and partly springing from the speculative frame of mind that prevailed just before the industry took its real rise, the possibility of employing negroes in cotton mills was much in the air in 1880, certainly in South Caro- lina. The chief source of information, the Blackman Re- port, contains clear indication of the activity with which the public imagination was working at the time. Few new mills were building; the remoter history of the industry had lapsed for the moment into the background; the new development had not commenced. In contemplating the mills then in operation, there was the feeling that they were not important as types of the past nor as presages of what was to come. That there was to be a new story there was no doubt. Thus in this interval between sterile past and dy- namic future, inquiries might be poorly informed and an- swers afterwards shown to be mistaken — very often the creeping of the chrysalis from the old cocoon was not noticed. But knowledge was being gathered, stock was being taken, resolve was forming to meet the challenge that was rightly guessed to be impending. The Blackman survey of cotton mills operating in South 102 William Banks, int., Columbia. The experiment lasted a year or more, but the negroes were found to be poorly adapted to the work; a fire disabled the plant after this trial, and it was converted into a warehouse. Cf. Kohn, Cotton Mills of S. C, p. 24, and Tompkins, Cotton Mill, Commercial Features, pp. 109-110. 214 TH E RISE OF COTTON MILLS IN THE SOUTH [32O Carolina in 1880, made for the News and Courier, while rarely mentioning the finding of negroes in the factories then, often comments upon the exclusive employment of whites. 103 Probably ante-bellum experience was respon- sible for survival of negro operatives in the Saluda Cotton Factory. Blackman found in this mill a hundred opera- tives, twenty-five of whom were colored, ranging in age from eight years up. Operatives lived in homes owned by the factory. Asked as to the feasibility of employing negro operatives, the superintendent replied that " at his factory he had worked mixed operatives with great advantage. The negro was as capable of instruction in the business as the white male or female, and could afford to work much cheaper, as they lived so much cheaper. The negro labor he found was easily controlled. . . ." 104 Blackman in his visits to the mills had a stock question designed to bring out the pros and cons of negro labor. He received answers from which it must have appeared pretty evident that negroes were not destined to play a progressive part in the history of the industry. These discouraging replies were based on disbelief in the suitability of working negroes and whites together, on the inadaptability of ne- groes to the employment, and on the plenti fulness of whites offering for service in the mills. One of the owners of a large factory said negroes were not apt enough to learn the business properly, whites would not work in the same room with negroes, and as most of the work was paid for by the piece, the labor if mixed must necessarily give unsatisfac- tory results. 105 103 Thus of Glendale: "The factory employs 120 operatives, all of whom are white" (p. 10). Cf. as to Langley, p. 7, and Red Bank, p. 8. 104 Ibid., p. 9. 105 Ibid., p. 10. An officer of a little establishment said: "The whites and blacks will not work together, and we have an abundance of white labor, which is certainly superior to any class of colored labor that we could employ" (ibid., p. 8). Cf. ibid,, p. 11. The summary of the report correctly said : " There is . . . considerable diversity of opinion on the subject of labor, but it appears that the preponderance is in favor of white labor, as more dexterous and trustworthy, and we assume that this difference will become more 32 1] THE LABOR FACTOR 215 The failure of a mill at Concord, promoted and managed by a negro and worked by negro operatives, cannot be taken as a strong argument against the feasibility of using the negro in the textile development of the South. This ven- ture was tried under such adverse circumstances as to make it practically without value as an indication one way or the other. The mill was projected in 1896 by Warren Cole- man, born in slavery and said to have been the illegitimate son of a prominent North Carolinian. Coleman had made money as a merchant in Concord and had built perhaps as many as a hundred "shacks" which he rented out to ne- groes; it was supposed that he was worth as much as $50,000. His natural father is reported to have assisted him to get his start in life, and to have advised about the mill project. The idea was that the factory should be a negro undertaking. The colored press commented enthu- siastically upon the appeals for subscriptions to stock, $50,000 was raised and the company organized in 1897 with Coleman as secretary and treasurer. The enterprise look- ing promising, the capital stock was increased to $100,000, some white people in the community subscribing to encour- age the effort. Negroes all over the State took the small sihares, which could be paid in trifling instalments. But many of the poor negroes who had subscribed could not meet the payments — some of the washerwomen made hardly more in all than their investment obligations amounted to, and many of the artizans who had agreed to work out their subscriptions in assisting with the erection of the plant dis- appointed the management. It took four years to complete the building, and when the mill was ready for operation, Coleman had had to assume much of the forfeited stock. A white superintendent from Easthampton, Massachusetts, was employed. The factory was handicapped by second- hand English machinery; the yarn market was depressed; marked as finer classes of goods are more generally made. The difficulty in obtaining operatives is not great. . . ." One superin- tendent had declared, however, that in his opinion, a mill could be run with negro operatives entirely, directed by skilled whites, at a 40 per cent saving (ibid., pp. 5-6). 2l6 THE RISE OF COTTON MILLS IN THE SOUTH [322 other Concord mills were making little profit. The yarn market pulled up, but Coleman's plant failed to pay, and ran only off and on after the first year. Coleman died in 1904 and a few months later the factory was sold under the mortgage. It was said the negroes made clever enough operatives, learning quickly, and the manager at the last attributed the failure to other factors than labor — poor ma- chinery, insufficient capital, unaccustomed administration. Operation of the plant was loose ; sometimes the mill would stand idle for hours waiting for cotton or fuel. Some of the operatives would be considered good average workers in any Southern mill, though while white spinners at Con- cord were receiving ten to twelve and one half cents per side, the negroes could command only five or six cents, making only about $2.50 per week. Being old, the machin- ery had to be run slowly to give good results; the negro overseers showed favoritism and were harsh in docking operatives. The manager believed that under favorable circumstances, near a city where more intelligent negroes might be gotten, a mill could be run successfully with white overseers and colored operatives. 106 In an attempt to use negro hands in the old plant of the Charleston Manufacturing Company, the direct manage- ment was of the best, capital was sufficient and the machin- ery was new. But because of peculiar local circumstances attendant upon this experiment, it does not reflect much light upon the apparently satisfactory character of the labor in the Concord mill. Before telling of the Charleston expe- rience, however, it is curious to notice that one of the great pioneers in the cotton mill movement in South Carolina, H. P. Hammett, in 1880, answering questions designed to bring out points in which Charleston as a prospective manufac- turing place was interested, predicted success for such a scheme as afterwards proved a failure : " I should think 106 p or the story of this mill I am indebted almost entirely to Mr. Thompson's " From the Cotton Field to the Cotton Mill," amended in minor particulars by an interview with Mr. Charles McDonald at Charlotte. 323] THE LABOR FACTOR 2\"J that a yam mill in Charleston properly constructed to make coarse yarns alone out of the rejected and cheap cottons that could be bought there might be run to great advantage and profit to the owners. I rather think negroes could be used to do coarse yarn work. I think that they could be trained to make very fair goods. I don't think the labor would be much cheaper than with white operatives. We give our operatives good wages and take care of their morals." 107 The plant, which for some time had been idle, was bought through the initiative of a successful up-country manufac- turer in cooperation with Charleston men who had been in- terested in the former company. It operated first with white labor and was a failure. After about a year it was determined to try negro operatives. Enthusiasm of much the same sort as had marked the original enterprise, directed now, however, toward the opportunity for negroes instead of poor whites as factory workers, was evinced. But again the projectors relied upon their own a priori opinions much more, it may be thought, than upon assurances proceeding from a study of the abilities and willingness of the negroes whom they wanted to employ. They felt so keenly that the plan ought to succeed that they did not inquire greatly whether it would succeed. " The superintendent of the mill and myself," said one of the stockholders, got the col- ored preachers and a negro ex-policeman down here at the bank and showed them the opportunity for the colored peo- ple if they would go into the mill and make good operatives. They saw it too, and as far as we know did all they could, but they couldn't make efficiency where it wasn't. The negroes lost a great opening." 108 107 Blackman, ibid., p. 17. 108 George W. Williams, int., Charleston. Another of the in- vestors gave a similar account : " We were assured the colored people would work for low wages, less than whites, and would be faithful, but they turned out to be just the reverse. We had everybody ex- horting them, telling them now was their opportunity, and that if the experiment succeeded here, mills all over the South would be open to them. But when a circus would come, they would all troop 2l8 THE RISE OF COTTON MILLS IN THE SOUTH [324 The superintendent declared that he had educated 3000 negro operatives to the work and made them competent, but that on any one day he could not get 300 of them in the mill. White operatives were used in the picking and card- ing rooms, separated, of course, from the negroes up stairs, employment of some whites being necessary to provide enough workers to run the plant. 109 The mill operated for about a year with negro labor, and the unsuccessful venture was discontinued. " We had the best management and fine machinery, and all the money necessary. It was the labor. I am absolutely convinced it was the labor." 110 The failure at Charleston had the effect on some mill men of confirming their disbelief in negro labor, but with others did not daunt their faith in the theoretical soundness, at least, of the proposal. 111 away to it. It was a sight to see them " ( W. P. Carrington, int., Charleston). The negroes, shunning "the opportunity of their lives, would go for oysters in the oyster season, and then for straw- berries in the strawberry season" (Williams, ibid.). 109 F. Q. O'Neill, int., Charleston. 110 Ibid. One of those who had worked hardest to prove negro operatives suitable, said : " If a white man will get 92 per cent out of a machine, a negro will get 76 per cent only and be satisfied " (George W. Williams, int., Charleston). Mr. Thompson thinks one reason for the unsatisfactory issue of the experiment was that the mill as worked by negroes was expected to pay dividends on a capi- talization enlarged by installation of new machinery. Mr. Mont- gomery, chiefly responsible for the enterprise, is reported to have assigned the distractions of the city as cause for the failure (pp. 251-252). A stockholder attributed non-success to malign influ- ence of selling agents of the mill : " The commission house took every means to show the colored labor unprofitable. Those negro women could tie a knot at a spindle as well as white women could." One of those interested in the company still believed the plan of having negro operatives was abandoned too soon, that the mill was on the eve of making money when the machinery was moved to Gainesville, Georgia (William M. Bird, int., Charleston). Cf. Tomp- kins, Cotton Mill, Commercial Features, pp. 109-110. 111 The superintendent of a large up-country mill said that the superintendency of the Charleston mill was offered him at the time negro labor was to be installed, and that he promptly declined the position. " The negroes' average of intelligence is so low that you cannot organize them. If you could pick them from all over the State, you might accomplish something, but taking them as they come, you cannot accomplish anything" (W. J. Britton, int., Spar- tanburg). A mill president of Augusta, speaking of the prevalent belief that cotton cannot be profitably manufactured in a seaport 325] THE LABOR FACTOR 2I9 Evidence gathered at first hand, even after fifteen years, still bears out Mr. Thompson's observation that the mill men of the South have thought of negro labor in a specu- lative way only, as a remote possibility or necessity. Since Mr. Thompson wrote, however, the South has approached measurably closer, in common conception of manufacturers, to a genuine and widespread shortage of operatives, and has felt this condition in an increase in wages not entirely con- sequent upon the European war. Steadily the question of employing negroes in the mills has gained a place in the minds of manufacturers. The hope of continuing the favorable labor differential, in spite of child labor legisla- in the South, gave his reaction in brisk sentences : " I have no sympathy with this view. If you can command the managerial skill, and you can surely get the machinery, you can run a cotton mill in a seaport as well as anywhere else. At once Manchester and New Bedford and Lowell come into your mind — they have all got spin- ning climates. In the South there is, of course, the labor problem, most of the operatives coming from the up-country. But there are labor troubles in anything. You will have labor troubles in running a shoe factory in New England or in picking prickly apples in the. Zulu Islands. So far as I am concerned, if I wanted to operate a mill at the coast, say Charleston, I would employ negroes. I wouldn't work them as those people worked them. I would not pay them half as much as white labor, but just as much. There is no reason why colored labor will not prove profitable." An expert in cotton mill practice said : " A negro can run a ginning outfit as well as a white man, and is tickled to death with it. The great trouble with negro labor for cotton mills is poor adaptability to organiza- tion. If I was going to run a mill with negroes, I would want to be right on the ground and study them, and not follow the experiment of trying to run the mill in Charleston with the president living in Spartanburg. I don't see why colored operatives cannot be used in cheap mills" (J. H. M. Beatty, int., Columbia, Jan. 3, 1917). A superintendent eminently practical declared : " The only trouble with negro labor is the mixing of the races. If a mill could be run exclu- sively with negro operatives, there would be no difficulty. Why, we have negro bricklayers, tailors, decorators, and these do handsome work ; negro women are good seamstresses ; there are negro dentists and doctors. I don't see why piccaninnies won't make good factory hands, spinning and weaving. There is nothing lacking in their capacity to learn" (George T. Lynch, int., Augusta). And a mill official whose family name is synonymous with the founding of the industry in the South said that while his fellow-manufacturers would want to hang him if they thought he expressed such a belief, he saw no real reason why negroes cannot be profitably and suitably used as operatives. 220 THE RISE OF COTTON MILLS IN THE SOUTH [326 tion and the entrance of trade unionism, may lead to further tests of negro operatives. 112 112 Cf. the writer's " The End of Child Labor," in Survey, Aug. 23, 1919, pp. 749-750. Tompkins in 1895 said : " It is impossible for me to come to a conclusion as to whether the colored people would make successful mill hands or not. ... I would be willing to be one of 100 persons to subscribe $1,500 each for a mill to be operated by colored people until by losses it should be determined that the ex- periment was a failure. . . . This experiment is important to the whole South. . . . With white labor alone it will be only a few years before we reach the limit of supply. Then we will without doubt have the same laws, the same experience and the same accessories [sic] of new labor from various sources that New England has had." Foreseeing child labor laws and legislation governing hours of work in the South, he felt that " the general conditions will con- stantly approach closer and closer to ^hose that have been already brought about in old and New England." While he had no doubt of the negro's intelligence, he thought he lacked tenacity of purpose where the work was monotonous, and that in the warm rooms of a mill, doing light work, he was apt to fall asleep (Cultivation, Pick- ing, Baling and Manufacturing of Cotton, p. 11 ff. Cf. ibid., p. 15 ff.). Mr. Thompson wrote that "speaking broadly the difficulty with negro operatives is not an intellectual one," believing that the chief failings of all negro labor are moral and temperamental. Draw- backs are dislike of the negro of working alone, insufficient ambition and pride in his work ; daily association in the same employment might make the negro less respectful to the white man (Cotton Field to Cotton Mill, pp. 240-250). However, this student admitted that bettered standards of life might enable the negro to enter occupa- tions which growing scarcity of white workers must open to him (ibid., p. 266). Cf. Tompkins, Cotton Mill, Commercial Features, pp. 109-110. Relying upon disqualifications usually noted, Mr. Cope- land concluded that " There is little likelihood that the negro will become the mill operative of the future," and that " he would require more supervising than his labor would be worth." This writer is mistaken in saying that " Before the Civil War the use of slaves in the factories was occasionally suggested." As has been seen, they were in several instances actually used ; in declaring that " no com- petent business man has yet ventured to make a real test" of negro labor, he overlooks the management of the mill in Charleston (p. 47 ff.). Cf. Uttley, p. 45. Mr. Goldsmith says dogmatically that " The negroes cannot be utilized in the manufacture of cotton " (p. 10). In a discerning summary of reasons for non-employment of negroes in the mills, Murphy placed chief stress on the natural pref- erence of managers for the stronger race, it being often difficult to employ the two together (pp. 103-104). In conversation, Mr. Kohn confirmed the position taken in his writings, and emphasized the hurtful absenteeism of negro labor (interview, Columbia, Jan. 5, 1917). A deterrent to the employment of negroes at the emergence of the mill period not sufficiently dwelt upon, is found in the bitter hatred, born of political and racial fear, that followed the war and Reconstruction. For an excellent statement illustrating this point, see testimony of Lewis W. Parker in report of Hearing before 327] THE LABOR FACTOR 221 Not only do opinions differ in regard to negro labor, but facts point in contrary directions. A hosiery manufactur- ing company which began using negroes in one plant in 1904 has recently installed colored help in two more factories, in all three cases due to shortage of white operatives. Wages of these operatives run from 20 to 40 per cent lower than for white knitting mill hands ; they get no better than 80 per cent production from the machines; special care must be used to hold absenteeism in check ; difference in production of negro and white workers is not so great as difference in wages, but the number of " seconds " turned out by negroes is greater than in the case of whites. The cost of supervision is higher with negroes, but under prac- ticed management their skill in these mills, as judged by fineness of work, has more than doubled. Negroes offer themselves in sufficient numbers to allow of some selection. All are piece workers. Only superintendents and foremen are white. The judgment of the manager is that where white operatives can be secured, negro labor in textile mills should not be attempted. He could readily understand why a silk mill operated in the South with colored hands had failed, and uses negroes only on coarse work. It is pleasant, in a study such as this, in which many con- clusions as to broad social conditions must be reached by inference, to come upon a part of the subject in which there is an absolute expression of facts under consideration. In speaking of wages paid in cotton mills of the South there are, happily, some figures to form the center of the discus- sion. Although occasionally distorting the image a little, the wages, in the instance of this industry in the South, con- stitute a mirror to reflect complicated economic phenomena in a way to make them realizable and concrete. Much evi- dence which, after passage of time, is undiscoverable in itself, many factors which no one would even think to look for as bearing upon the problem, are unfalteringly assimi- House Committee of Judiciary upon proposed constitutional amend- ment giving congress power to regulate hours of employes in fac- tories, April 29, 1902, part 2, pp. 11-12. 222 THE RISE OF COTTON MILLS IN THE SOUTH [328 lated in the wage scale. Wages paid the operatives are a composite photograph not only of the state of the industry at the time of its commencement, but of the agricultural, social, commercial and educational situation of the South at and just previous to the period here treated. It will be found that wages varied very markedly from one locality to another for practically identical work; this, so far from weakening the force of what has just been ob- served, strengthens it, for it has been impressed all along that there was in the South only that standardization which proceeded from the weight of poverty; that it is impossible, as to most aspects, properly to speak of industry in the sec- tion as a whole, but only of the particular facts for separate communities. In these pages regarding wages the reader should keep in mind not only preceding discussionis in this chapter — as to condition of the poor whites before they entered the mills, the generally superabundant supply of native white people, the large proportion of women and children in the first mill populations — but larger aspects of the whole study as well. The part played by labor in forming an argument, selfish or philanthropic, for the building of factories; the earnestness with which communities cooperated to raise capital in the face of meagre resources; the faith with which projectors of enterprises reached out for support from the North; the character of plants erected and of machinery put in them; the relations with creditors and commission merchants* — all these have their bearing. If it be true that the cotton mills of the South rested to a large extent upon plentiful supply of native white labor, then wages paid to operatives afford a convenient indication of the level from which the industry took its rise. First to glance at some ante-bellum wages. It is said that in the early fifties $116 per year was the average cost of white labor; then average wage's on the basis of 300 working days amounted to thirty-nine cents. 113 113 A table given by Montgomery for 1831 includes only two South- 329] THE LABOR FACTOR 223 In 1856 Olmsted wrote that there were from 15,000 to 20,000 spindles running in Columbus, the largest manufac- turing place south of Richmond. " The operatives of the cotton-mills are said to be mainly ' Cracker-girls ' (poor whites from the country), who earn, in good times, by piece- work, from $8 to $12 per month." Workers in all Colum- bus factories of various sorts were declared to be " in such a condition that, if temporarily thrown out of employment, great numbers of them are at once reduced to a state of destitution, and are dependent upon credit or charity for their daily food." 114 At a time when negro labor was dearer than that of free whites, when slaves were better looked after than white people doing similar work, it is not surprising that there should have been no social watchfulness of the conditions of employment of the latter. 115 Slavery precluded moral and. economic alertness on the part of the public. As comes out more clearly in post- bellum days, it was a miracle if there was work for mien and women to do; everyone was far from quarrelling with the terms of engagement. 116 ern States, Maryland and Virginia. Average weekly wages of males in Maryland amounted to $3.87 and of females $1.91 ; male operatives in Virginia received $2.73 per week and females $1.58 (Practical Detail of Cotton Manufactures, p. 161). Cf. ibid., p. 133. In 1849 in the DeKalb factory in South Carolina operatives exclusive of weavers received from 13 to 36 cents per day; in the same year the average wage at Vaucluse was 37 cents, most of the hands being women and chil- dren; 300 hands in an Augusta mill averaged $3.05 per week; in an Alabama town the average was $8 per month; at Columbus the pay was from 12 to 75 cents per day for operatives, and for overseers from $1 to $1.25 (Ingle, pp. 75-76). 114 Seaboard Slave States, pp. 547-548. " Public entertainments were being held at the time of my visit, the profits to be applied to the relief of operatives in mills which had been stopped by the effects of a late flood of the river." (Cf. ibid., p. 543.) 115 Cf. Olmsted, p. 543. 116 The position of the South in i860 can be fancied when it is said of the entire country at the same date that " One or two states had passed laws regulating hours of labor; but none had thought of the cost to the race of hard toil and long hours for women and children, and most men regarded the builder of a mill as a public benefactor because he furnished employment to just this element in the population" (Dodd, Expansion and Conflict, pp. 209-210). 224 THE RISE OF COTTON MILLS IN THE SOUTH [33O An Alabama cotton manufacturer declared in 1878 it was cheaper by 42 cents per hand per day to operate a mill in his State than in Massachusetts. 117 In 1883 a New Hampshire hosiery manufacturer purposed establishing a mill at Columbia, South Carolina, and said he could run the plant there " without counting the cost of raw material (which he could procure ... at less cost than further North) at least twenty-five per cent cheaper, in the cost of labor alone, than he could in New England."' 118 Major Hammett, showing the advantage of the South over the North, especially in the manufacture of coarse goods, estimated that the difference in labor amounted to not less than one and one half cents per pound. 119 A writer in the New York Times observed that " a Fall River, Lowell, or Manchester operative would hardly be able to live on the $4 a week which will make a Georgia operative's family comfortable. There is an abundance of native white labor to be had at from 50 to 60 cents a day." 120 Coming to a typical mill of the early eighties, it was re- ported of Clifton, in South Carolina, proposing to employ four hundred native whites, that wages would amount to 5° cents to $1 per day. 121 In one of the little note books giving informal estimates for the Charleston Manufacturing Com- pany, there is a memorandum indicating that a good super- intendent would cost $4000 a year, and " Labor 25 cts to 1.50 pr day" with the additional remark: "these wages paid in cotton mills in the State — good authority for this statement." 117 It is difficult to tell how much of this saving was imputed to lower wages, and the problem is not much helped by the calculation that a 4000-spindle, 125-loom plant in Alabama had a cost for labor and mill expenses amounting to $63.44 per day (Haralson, in Burney, Handbook of Alabama, p. 271 ff.). 118 Manufacturers' Record, Baltimore, Feb. 1, 1883. 119 Quoted from Atlanta Constitution, in ibid. 120 Quoted in News and Courier, Charleston, Nov. 5, 1881. Low wages were ascribed to low cost of living. 121 News and Courier, Charleston, May 21, 1881. This plant had 17,000 spindles, 500 looms, and a capital stock of $500,000. 33 I ] THE LABOR FACTOR 225 It must be apparent that low wages paid to operatives in Southern mills were bound up with the low cost of living. Remuneration which would otherwise seem impossibly small becomes understandable when expenses of the operatives are seen to have been very little. In order to know the con- dition of the workers, it is obviously necessary to keep in mind real wages and not money wages. It must be noticed that lower wages prevailing in the South, however accom- panied by greater purchasing power and by other payment by the mills in kind, showed the less advanced economic position of the South as compared with other parts of the country. Also, the standard of life of the Southern opera- tive was lower than that of the New England operative, and however completely Southern wages allowed the for- mer to reach his standard of life, he was probably not so well off as his New England brother who saved no more money. In estimating the real income of workers in the Southern mills it must be remembered that the companies made up a considerable part of the pay in goods and services rather than in coin, and this practice of the early establishments has in large measure endured through the years, affording one of the most striking particulars in which the old eco- nomic system, born of slavery and fostering a paternalistic attitude of master toward servant, of employer toward em- ploye, has persisted into a new day. A little mill in the deep country, which got its start shortly before the indus- trial era, manufactured the coarsest yarns on 880 spindles, and had 600 acres of cultivated land and a gin. Twelve operatives, all white, received an average wage of 33^2 cents per day, and 120 persons in all were dependent upon the factory for support. The very low pay and the number of those looking to the mill for a living occasions less sur- prise when it is learned that the company furnished its operatives with houses free of rent. 122 122 Blackman, p. 11. A factory a little larger, but otherwise about similar, on a waterpower located eighteen miles from the railroad, employed 65 white operatives at an average wage of 40 cents for IS 226 THE RISE OF COTTON MILLS IN THE SOUTH [332 Low wages were partly due to the limited money econ- omy. Companies frequently could easily (bear the prime living costs of their operatives and their families, when the equivalent amount could not well have been paid in cash. The smaller the quantity of money required, the more con- venient it was for manufacturers cramped for capital. The company store, which became a widespread institution, as well as being a necessity in isolated factory Ideations, was designed to limit the amount of circulating capital required by the mill management. The company-owned village has been an extension of the company store. At Piedmont, with 300 operatives and 600 dependents in all, $50,000 was suf- ficient to cover the annual pay roll, including salaries, the average wage for spinners being 50 cents a day. Opera- tives lived in seventy-seven tenement houses furnished free of rent by the company. 123 In such " free villages "' many lesser gifts are implied in remission of rents. Often wood might be cut from the company's land and cows pastured in the company's fields, and garden patches about the houses were well-nigh universal. In these first villages of the Campaign years poor whites from the neighborhood, des- perate for a means of livelihood, thanked the mill for all their needs. Whatever other factors contributed to the low scale of wages, the primary cause, of course, was the lean condition of the South and the relatively small number of jobs as contrasted with the large number of those wanting work. Wages were really a question of what the factories could pay, rather than of what the people might ask. With economic progress, as agriculture has become more pros- perous and industrial plants thicker, wages have steadily risen. It has been noticed that wages varied greatly from place to place. At Crawf ordsville, for example, wages were 33^ cents a day; Fork Shoals, not far away and a mill almost spinners ; there were 200 people dependent upon the factory, which provided houses free (ibid., p. 13). 123 Ibid., p. 16. 333] THE LABOR FACTOR 227 precisely circumstanced, paid about 40 cents a day. This is attributable not only to want of knowledge of workpeo- ple in one neighborhood of conditions of pay and of living prevailing in another, but to the extreme provincialism of employers as well. An important factor besides was wide variableness of circumstances of employment from one locality to another. One mill might have operatives for some years already in its service, and as unwilling to leave for the farms again as they were not likely to go to another factory paying more wages ; relative proximity of a mill to a city or town, controlling to some extent the agricultural rents and value of farm produce, and the ease or difficulty with which families could move their few effects, as well as the outlook of the people upon whom the plant drew for operatives, would make a difference in wages. The pro- portion of women and children employed would have a vital bearing. These things influenced also the operating costs of the factories, and thus indirectly as well as directly helped to determine the amount of money that might Ibe paid in wages. These considerations apply more conspicu- ously to the mills of the seventies than to those of the eighties, but the distinction is one of degree and not of kind. The larger and more numerous factories became, the less, of course, such forces prevailed. The summary of the Blackman Report said there were 2,612 operatives in South Carolina in 1880, upon whom 8,143 persons were dependent for support. The amount paid out in wages monthly was $38,159, and the rate of wages for spinners ranged from 25 cents to 78 cents a day, " according to the situation and the character of the labor." The Valley Falls factory, near to Crawfordsville and Fork Shoals, mentioned above, was making profit on coarse yarns although the machinery was old and " despite . . . the great disadvantage of being situated in an almost inacces- sible region." Fifteen operatives were employed at an average wage oi 40 cents a day. 124 124 Ibid., p. 11. At Reedy River, a larger mill in the same locality, 228 THE RISE OF COTTON MILLS IN THE SOUTH [334 The only indication found of dissatisfaction in the early- years with prevailing wages paid is that contained in the following item which appeared in 1882 : " Last Monday morning four of the 'warpers' employed in the Rock Hill Cotton Factory waited upon the superintendent and de- manded an advance in wages. Their demands were not considered. They were told that their services were no longer wanted at any price. They left and their places were supplied immediately." 125 Hard times in the industry, especially in the Augusta dis- trict, where mills had never seemed to ride on the crest of the wave, were responsible for sharp wage reductions in 1884 in some of the largest factories. In May one cut officers' salaries 20 per cent and employes' wages 15 per cent, and another reduced both salaries and wages 15 per cent. The policy was inaugurated to allow continued sales at market prices, 126 but was only partly efficacious, for the the average pay of the 65 operatives was 50 cents a day (ibid., p. 13). At Glendale, where 60 per cent of the 120 operatives were women and children, wages averaged 67 cents a day (ibid., p. 10). Langley in 1879 made coarse shirtings and drills as well as yarns, and paid its operatives, two-thirds of them female, an average of 78 cents per day (ibid., p. 7). Not only might wages vary from mill to mill, but the range might be wide in a single factory. Thus at Glendale the highest was $1.50 per day and the lowest 12^2 cents (ibid.). Persistence of variations in wages between mills may be strikingly seen in facts gathered by Mr. Kohn (Cotton Mills of S. C f P- 43). A table of wages paid annually by the mills of South Carolina, and of the capital and spindleage of these factories, pub- lished in 1883, yields little that is helpful because " capital " some- times meant paid up capital and sometimes allowed capital ; a com- parison of spindleage with wages is not more helpful, because a part of the wages was undoubtedly paid for a greater or less amount of weaving. Two mills with combined capital of $600,000 and 32,368 spindles paid in wages $180,000 a year; Clifton and Piedmont each had $500,000 capital and paid annually $100,000 in wages, but the former had 19,000 spindles and the latter 23,000. These factories were probably making a larger proportion of yarns than Langley, which, with $400,000 capital and only 10,000 spindles, paid in wages $87,500 a year. Another mill, with less than half as many spindles, paid $18,000 a year for wages, or about one fifth as much as Lang- ley. The factory with the smallest capital was expending in wages only $2400 annually (Columbia correspondent of Augusta Chronicle and Constitutionalist, quoted in Manufacturers' Record, Baltimore, Jan. 18, 1883). 125 N ews and Courier, Charleston, Jan. 23. 126 Evening News, Augusta, May 28, 1884. 335] THE LABOR FACTOR 229 Enterprise mill soon shut down, and by the time it resumed in October the King and Augusta mills had reduced wages 25 per cent. 127 In the history of the industry in the South, increasing cost of living and, more importantly, growth of mills and diminishing supply of labor have given to wages an upward trend that, despite lapses and spurts, has been strong and inevitable; wages have advanced not gradually, but in jumps mainly as a consequence of accelerated mill building, though wage reductions in periods of slump have to consid- erable extent been avoided through absorption of the field from the factory, the opportunity open to operatives to re- turn to cotton raising. The low wage scale paid in the South as contrasted with other textile sections, notably New England, has often been remarked. The advantage flowing to the Southern manu- facturer from cheap labor, partially offset by the lower price received for Southern goods in many cases, and by inability of unskilled workers to get full production from expensive machinery, has proved more persistent than that resulting from other factors. 128 In assuming that, in real wages, Southern operatives have been as well off as those of New England, it must be remembered that the lower level on which Southern mills have been conducted has involved certain very definite social disutilities which do not appear in any calculation of expenses of living. Such are the results of child labor, long hours of work, poor schooling, mischievous abetting of harmful politics, a contracted economic outlook linked with difficulty in working through the mills to better employ- ments. That the mills have brought a better living, generally speaking, than most other employments open to the people, 1 27 Chronicle, Augusta, Oct. 21, 1884. 128 Cf. Kohn, Cotton Mills of S. C, p. 33 ff. ; Uttley, p. 56 ; Thomp- son, pp. 152-153; Tompkins, Storing and Marketing of Cotton, in transactions of New England Cotton Manufacturers' Assn., vol. 77, pp. IO-II. 23O THE RISE OF COTTON MILLS IN THE SOUTH £336 must 'be patent from the comparative ease with which fac- tories have obtained labor. 129 Apologists for Southern mills regularly, and others less disingenuous frequently, have laid stress on the favorable " family wage " received by opera- tives. 130 The error here would seem to be too evident to require correction. To the extent that the head of the fam- ily, certainly if assisted by one or two other adult members, cannot by his wages provide for those dependent upon him, the employing industry is socially parasitic. With the en- tire, or nearly the entire family in the mill, children com- peting fatally with their elders, money income, augmented by many payments in kind, has been only about sufficient to support tolerable existence. It has been observed that Southern mill wages have ad- vanced in jerks. Such a sharp rise came in the years lead- ing to the panic of 1907, when the increase was 25 per cent, more or less. 131 Besides paying higher wages, many mills introduced bonus plans. 132 In some cases these were not bonuses at all, but simply wages, operatives working through the noon hour or otherwise speeding up to earn them, but the bonus and less obvious devices took on stronger significance in the hands of employers in the period of sharper competition for operatives during the European War. In February of 1917 it was said: "Few industries can boast of an enterprise which would actually lose money in certain lines in order to benefit its workers, and yet that is just what the Loray Mills are doing. For example, they sell wood and coal to their workers at actual cost always, 129 £f Thompson, p. 159. Comparison of mill wages with those in argriculture, here given, lose force from the consideration that food might be given to farm workers and, besides, account must be taken of the depressing effect of negroes working in the country. Cf. Kohn, Cotton Mills of S. C, p. 50. 130 Cf. Kohn, Cotton Mills of S. C, for many references. 131 Cf. Thompson, pp. 159 ff. ; Goldsmith, p. 10; Kohn, Cotton Mills of S. C, pp. 33, 36 and 39. Statements of increased incomes to specific families in the years 1902 to 1907 are without point, be- cause of addition of new members to the working group and in- crease in ages and skill of operatives during the period. Cf. Thomp- son, ibid., pp. 148-149. is2 Cf. Kohn, Cotton Mills of S. C, p. 49. 337"] THE LABOR FACTOR 23 1 16 1 ,. "• "?f but during the past two months they have been furnishing their help coal at a price fifty cents per ton cheaper than the mills actually pay for the coal delivered to them." 133 Just as mills have had to compete in welfare work to hold their operatives, it is clear that these extra concessions were simply a convenient avoidance of cash payments. It might have been foretold that increases in money wages would not be allowed to stand frankly as such. The event proved that these were usually painted as " voluntary " on the part of the companies. 134 These things are not worth mentioning either as naive subterfuges of employers or more naive efforts of writers to create an impression favor- able to hard-set manufacturers, but are interesting as show- ing now a canny continuance of practices which were sincere and acceptable in the earlier history of the Southern indus- try. Wage advances were general and rapid in succession, and without doubt many employers would have given these out of pure goodwill to their operatives and to let them share in enormously mounting profits. But it did not need the negro migration that left farm vacancies, unprecedent- edly high prices for cotton, army drafts that took their thousands and war construction that held out unbelievable wage opportunities, joining with the first concerted union attention to the South as an unorganized field with resulting strikes at Anderson, Graniteville, and Columbus, to make it plain that the great stimulus back of increases in pay was the necessity of the mill managements. 135 133 Charlotte News, Textile Ed. Throughout this survey and a like one of the Columbia Record in 1916 are any number of similar references. In estimating the gratuity of the mills, it is well to note an observation of a North Carolina manufacturer : " Wages, though mill men may not recognize the fact, tend to be determined by the cost of living in the particular mill village: At High Shoals, where wood is $1.50 a cord, wages are less than at Charlotte, where wood is $4 a cord" (Sterling Graydon, int., Charlotte, Sept. 4, 1916). 134 Cf. Textile Editions of Columbia Record, 1916, and Charlotte News, 1917, the writer's "End of Child Labor," in Survey, Aug. 23, 1919, and Mill News, Charlotte, quoted in Literary Digest, Dec. 9, 1916. 135 Before these phenomena appeared, Tompkins said : " The pay of operatives rarely varies in the South with the price of goods . . ." (Cotton Mill, Commercial Features, p. 55). CHAPTER IV The Role of Capital The achievement represented in the rise of cotton mills in the South is not more clearly apparent than in the story of how capital was gathered and how financial operations of the factories were conducted. Here was an agricultural community made poor by war, economically disorganized by Emancipation and estranged from the capable North through Reconstruction, face to face with an unaccustomed task needing wealth, concert at home and cooperation from without. No new industrial movement has been a shorter time an the talking stage; the South met the acid test of purpose by plunging instantly into actual performance. Investments in Southern cotton mills increased about $2,- 000,000 each decade after 1840 until that of 1870-1880, when the advance was roughly $6,000,000 — from $11,- 088,315 to $17,375,897. The figures for the decade 1880- 1890 reflect the suddenness and rapidity of the growth once the undertaking was entered upon ; capital trebled to $53,- 821,303 and by 1900 had reached $124,596,879. 1 The for- ward leap was marked right from 1880. In the fall of 1882 it was estimated that the new paid up capital, not including increases from earnings, had amounted in the two years previous to between $15,000,000 and $i8,ooo,ooo. 2 It was of the genius of the movement that Southern capital should be drawn upon to the limit. It will presently be seen how valuably this was augmented from the North, 1 U. S. Census of Manufactures, 1900. In such an aggregate, these figures may be taken as sufficiently accurate; a local estimate for South Carolina in 1880 was a little under the census return. Cf. Blackman, p. 3. 2 Baltimore Journal of Commerce and Manufacturers' Record, Sept. 2, 1882. Cf. Manufacturers' Record, Baltimore, March 8, 1883. In Augusta, between 1880 and 1883, two mills were built and five enlarged, making 63,000 new spindles, 2200 new looms, and repre- senting an added investment of some $3,000,000 (ibid., Feb. 15, 1883). 232 339] THE ROLE OF CAPITAL 233 but as to both amount and importance it is right to say that " the chief sources of capital employed in starting the mills were local." 3 While recognizing the extent of outside as- sistance, an acquainted observer accurately said: "The great majority of cotton mills in the South represent the sacrifices and great efforts of the communities in which they are situated. In the East the cotton mill is built from the capital of the rich ; in the South it is built from the com- bined capital of many of little means." 4 It has already been remarked that a symptom of the Cotton. Mill Cam- paign was the location of factories in towns rather than on isolated water powers, and this was because community enterprise was coming forward. 5 "A most gratifying fea- ture connected with the establishment of cotton mills in the South," it was declared in 1881, "is that the great bulk of the capital employed in their operation has been furnished by Southern people. . . . More than three-fourths of the capital invested . . . has been subscribed by our own peo- ple." 6 Southern savings were almost under compulsion to 3 S. S. Broadus, Decatur, letter, Jan. 27, 1915. Cf. Edmonds, Facts about South, p. 32. Many interviews supported this point. 4 Testimony of Lewis W. Parker, Hearing before House Commit- tee of Judiciary, 1902, part 2, p. 12. 5 J. A. Chapman, int., Spartanburg, Sept. 5, 1916. In explaining how a place without wealth could establish the industry, Tompkins overstated the fact in saying that " every one of the towns and cities of the southeast that are now manufacturing places built their first factory out of native resources and without outside help" (A Plan to Raise Capital, p. 25). 6 News and Courier, Charleston, Sept. 1. Cf. quotation from C. C. Baldwin in ibid., July 11, 1881. "The industry is distinctly a home enterprise, founded and fostered by natives of the State," says Mr. Thompson, who agrees that 90 per cent of the mill capital of North Carolina was native (Cotton Field to Cotton Mill, p. 81). Cf. ibid., p. 59 ff. ; Augusta Trade Review, Oct., 1884. There is abundant support for the assertion that " the industry taken as a whole is almost strictly a North Carolina achievement" (Charlotte News, Textile Ed., 1917). Cf. ibid, as to Rhyne's mills and other instances. Twenty years after the commencement of the develop- ment, it was estimated that 80 per cent of South Carolina mill stock was owned in the State; as will be seen later, some of this had gravi- tated South from Northern hands, but against it might be set off shares held elsewhere in the South (Kohn, Cotton Mills of S. C, ed. of 1903, p. 32 ff.). One of the founders said that early in the period over 65 per cent of the capital invested in South Carolina was native (E. A. Smyth, int., Greenville). 234 THE RISE OF COTTON MILLS IN THE SOUTH [34O become cotton mill shares: "We may look in vain for the dawn of an era of enterprise ... so long as thousands and millions of money are deposited in our banks on four per cent interest, when its judicious investment in manufac- tures would more than quadruple that rate. . . ." 7 Many specific instances show the embodiment of this spirit. The considerable cotton industry of Columbus was wiped out by the capturing Federal army in 1865, and within fifteen years had been rebuilt by local capital to the point where the mills took nearly 17,000 bales. 8 Right after the War, a Northern observer believed that Charleston did not possess recuperative power for its rebuilding, that unless New England sent energy and capital the city would remain a wreck. 9 Yet by 1881 Charleston was leading the indus- trial advance in the State and furnishing a model of home enterprise in the Charleston Manufacturing Company. 10 It was declared in 1865 that all the mercantile stocks in Co- lumbia, in the heart of the devastated area, could be bought for $20,ooo. 11 In a decade and a half the citizens were buying out New England interests which had failed to de- velop the water power and plant cotton manufactures, sub- scriptions of $55,000 being received in one hour and reach- ing $117,000 in two weeks. 12 7 News and Observer, Raleigh, Nov. 9, 1880. Cf . ibid., Dec. 24, 1880, as to Edward Richardson. " I am tired of hearing the cry of ' We want Yankee brains and enterprise.' We don't want any such thing; we want Southern brains and enterprise. What the South wants is common sense and action" (C. M. Clay, quoted in Gannon, p. 18). Mills before the war, being usually neighborhood affairs, regularly had local capital, ordinarily from a few investors (cf. Thompson, p. 51). Perhaps the most ambitious suggestion for home subscriptions contemplated the building of mills by groups of fifteen planters who should take $4000 in stock each ; nothing came of the plan (cf. Barbee, Cotton Question, p. 138 ff.). 8 Observer, Raleigh, Sept. 10, 1880. 9 Andrews, p. 3. 10 It was said that more than three-fifths of the capital was con- tributed in Charleston (News and Courier, Charleston, July 6, 1881. George W. Williams, int., Charleston). 11 Andrews, p. 34. 12 N ews and Courier, Charleston, March 17, 31, 1881. Cf. Char- lotte News, Textile Ed., 1917, regarding Hope Mills. An enter- priser in a small South Carolina town announced in the local paper : 34 1] THE ROLE OF CAPITAL 235 Not only was a large proportion of the stock held locally, but it was chiefly native investors that actually paid in cash ; it will be seen soon that machinery manufacturers received stock in return for equipment. 13 It frequently happened that after local capital, largely from community spirit, had been adventured generally in a first enterprise, succeeding mills would be erected by a small number of investors as private establishments. 14 Also, even with initial mills, stock after a few years tended to come into the hands of the larger investors who had been central in the subscription. 15 It has been observed that Gregg, before the War, plead that the dormant wealth of Charleston might be directed into the industrial development of South Carolina. 16 In " I am now engaged in getting up a mill of 2,500 spindles to manu- facture yarn at this place. I do not expect to seek a dollar of for- eign subscription, but I want our own citizens throughout the county to be interested in it and to help me build and operate it" (D. J. Winn, in Sumpter Southron, quoted in News and Courier, Charles- ton, March 31, 1881). Cf. ibid., Jan. 21, 25, 27, 1881. "The project of establishing a manufactory for cotton near Walhalla is being mooted. An informal meeting of some of the citizens of that place was held . . . and stock to the amount of nearly $10,000 was sub- scribed by the few present. It is believed strongly that as much as $25,000 will be subscribed in that neighborhood, and if the people of the county will join in the enterprise as much as $50,000 might be made available" (ibid., Feb. 26, 1881). The instances of the first mills at Salisbury and Laurens, applicable here, have been recited. Mills at Rockingham were built principally with money from " home people" of that and adjoining counties (William Entwistle, int., Rockingham). Cf. Columbia Record, Textile Ed., 1916, as to mills at Greenwood, Ninety-Six and Lancaster ; Grady, pp. 197-108 ; Tomp- kins, History of Mecklenburg, vol. ii, p. 198; News and Courier, Charleston, March 17, 1881 ; Baltimore Journal of Commerce and Manufacturers' Record, June 24, 1882, as to Southern subscriptions to Atlanta Exposition. 13 Benjamin Gossett, int., Anderson. 14 Cf . Tompkins, Cotton Mill, Commercial Features, pp. 29-30. When the Charleston mill, which had been peculiarly public in its inception, was to be sold in 1809, the little group of intending pur- chasers prepared for the transaction with secrecy (Bird Memo- randa). Some who had withheld subscriptions from the first Salis- bury mill went in on the second, and the third was distinctly private in character (Theodore F. Klutz and O. D. Davis, interviews, Salisbury). 15 Hudson Millar, int., Charlotte, N. C, Sept. 4, 1916. 16 Speech on Blue Ridge Railroad, pp. 6-7, 29. 236 THE RISE OF COTTON MILLS IN THE SOUTH [j42 the eighties his wish was amply, if tardily, satisfied. It was principally Charleston capital which developed such up- country mills as Piedmont, Pacolet, Clifton and Pelzer, 17 and many smaller factories drew partially on that city. 18 Many Charleston men, besides, went out as mill builders. Gregg succeeded in persuading South Carolina to grant limited liability to incorporators of industrial enterprises, 19 and here again after events showed his wisdom. When, during the Cotton Mill Campaign, poor communities felt a stirring which wealthy individuals had not experienced, the companies were regularly incorporated; also, established factories wishing to enlarge sought this facility. 20 Noticing that many towns which despaired of being able to project a cotton mill yet had building and loan associa- tions with accumulated cash in excess of the amount neces- sary for a factory, Tompkins took the lead in applying the building and loan principle to manufacturing enterprise. Under his guidance a score of plants, mostly in the Caro- linas, were successfully set going; the instalment payment was usually 50 cents a week, though sometimes $1.00 or as little as 25 cents. The mill might be erected as capital came in, or might be completed sooner with money borrowed on endorsement of directors or notes of subscribers used as collateral. 21 The instalment plan did not come into use 17 E. A. Smyth, int., Greenville. 18 Mills at Sumter and Anderson, after exhausting local resources, appealed successfully to Charleston investors (News and Courier, Charleston, Dec. 17, 1881; Marshall Orr, int., Anderson). 19 Cf. Propriety of Granting Charters of Incorporation, p. 4ff. ; Domestic Industry, p. 16. 20 The South Carolina legislature in the 1882 session granted char- ters to nine mills with an aggregate capital of $1,725,500 (Manufac- turers' Record, Baltimore, Jan. 11, 1883). By 1910 the South showed 20 mills owned by individuals, 13 by firms and 620 by corporations (U. S. Census of Manufactures, 1910, Cotton Manufactures, p. 44). The old subscription list of the Bivingsville mill, 1856, in the posses- sion of Mr. J. B. Cleveland, of Spartanburg, contrasts with the model' charters explained in Tompkins' writings (cf. Tompkins, Water Power on the Catawba River, p. 20 ff.). 21 Cf. Tompkins, Plan to Raise Capital. With a mill operated be- fore all capital was paid in, earnings would balance interest; some- times profits were so large that a plant under these circumstances even paid dividends (cf. Thompson, p. 82 ff.). In one instance half 343] THE R ° LE 0F CAPITAL 237 right at first, and was generally employed in modest enter- prises. 22 Turning now to financial participation from without the section, it is to 'be remembered that by 1880 the Southern attitude toward Northern assistance was warmly, even ar- dently cordial. Willingness to welcome help of Northern money in Southern mills was a test of earnestness in the new program, the characteristic mark of conquest over hurtful pride and estranging rancor. The wish for nation- alism and for industrialism on the part of the South was necessarily one. Immediately after the War only the wisest men- championed the entrance of Northern enterprise, and found the up-country far more favorable to this view than the low-country. 28 But fifteen years afterwards, Southern sentiment responded outspokenly to even the imperfect sym- pathies of Edward Atkinson. 24 In connection with the At- lanta Exposition it was said : " We have in the midst of us the raw material ... of a magnificent prosperity. We lack knowledge, population and capital. These may be slowly accumulated in the course of years, or they may be rapidly by well directed efforts to obtain them from beyond our own borders. We advocate the latter plan." 25 A com- petent interpreter of South to North asserted: "I say on the strength of recent and extended observation that what- ever of antagonism to Northern capital may have existed in the South has disappeared. I never met it, at any time. . . ."" 26 Grady was representative in regretting " that our the subscription was paid down and the balance piecemeal (J. L. Hartsell, int., Concord). 22 The mill built by negroes at Concord was virtually shut up to this method. 23 Cf . Andrews, pp. 79-80, 176, 320, 378. 24 Cf. Atkinson, Address at Atlanta, preface, p. 4; p. 8. 25 News and Courier, Charleston, March 14, 1881. A country cor- respondent declared that the South could not afford to remain solid ; that the party that could guarantee the safety of incoming capital was the party for South Carolina (ibid., May 25, 1881). 26 C. C. Baldwin, quoted in News and Courier, Charleston, July 11, 1881. The slave States, themselves emancipated, stood "with a warm and generous recognition of the right of all men of every section of their own country, and of every foreign land, to come into their territory, whether with muscle or money, and share with their 238 THE RISE OF COTTON MILLS IN THE SOUTH [344 brothers from the North have not taken larger part with us in this work " of building up the South. 27 Before speaking of solicitation of capital from particular sources, some general bids for outside assistance may be mentioned. Exemption of new cotton factories from taxa- tion, if losing to a State a few thousands, might be expected to induce Northern capitalists to invest in the industry. " Once here, they will be so pleased with our advantages that they will never think of leaving us." 28 The wide- spread rebuttal which met the statement of Edward Atkin^ son that he " could not conscientiously recommend invest- ments in Southern cottgn mills" showed how keenly the South desired Northern capital. 29 " Unfriendly comments " drove him to conciliation. Superiorities of South over North were set forth in a business-like way, much in the manner of a prospectus, often concluding with a suggestion own people in developing its riches " (Richmond Industrial South, quoted in Baltimore Journal of Commerce and Manufacturers' Rec- ord, June 17, 1882). "... Southern investment encourages North- ern capital to come into the same field, and the rate of progress is far more rapid than if it depended on either Southern savings or Northern capital alone " (News and Courier, Charleston, Sept. 1, 1881). Cf. Daily Dispatch, Richmond, March 5, 1880, as to North- ern money in a railroad project. 27 Expressions of suspicion were rare; a small paper, out of the current, said : " Well enough is it to talk about repelling Northern capital by discriminating legislation, but far better have no Northern capital than have it holding native noses down to the grindstone. The half-starved mountain wolf refused to change places with the sleek mastiff that wore a master's collar " ( Winnsboro News, quoted in News and Courier, Charleston, July 7, 1881). 28 " One mill owner, himself a Northern man, stated that if their advantages were fully understood at the North, a great many North- ern capitalists would make investments in factories at the South " (Observer, Raleigh, Feb. 13, 1880). An Augusta correspondent of a Cincinnati paper, reciting the success of mills in the Southern city, gave this data " for the information of the loose capital which is floating around Cincinnati, seeking five or six per cent investments . . ." (quoted in Manufacturers' Record, Baltimore, Jan. 4, 1883). Cf. Baltimore Journal of Commerce and Manufacturers' Record, June 3, 1882. 29 Atkinson, Address at Atlanta, p. 27 ; cf . p. 14. It turned out that above other benefits of the Atlanta Exposition was "the confi- dence begotten in Northern capitalists by the astonishing display of material wealth and the opportunities offered them of making perma- nent investments . . ." (Baltimore Journal of Commerce and Manu- facturers' Record, Sept. 28, 1882). 345] THE ROLE OF CAPITAL 239 that "to the anxious capitalists tired of a petty 4 per cent . . . such facts are not without interest. They go to sup- port the claim that the Southern mill has an advantage of from 10 to 20 per cent over its New England competitor." 30 Not all appeals for outside help gave promise of realiza- tion. Such was the " Cotton Syndicate " proposed to link Southern plantations with Manchester weaving mills. The abortive project is interesting, though, as indicating how even farmers could look to cotton manufacturing for sal- vation. 31 Some Southern men made active hunters after Northern investors : " Mr. D. L. Love, the pioneer of cotton factories in Huntsville, lefit for New England . . . for 80 Atlanta correspondence of New York Times, quoted in News and Courier, Charleston, Nov. 5, 1881. Southern papers eagerly presented news of successful enterprises in order to attract Northern attention (cf. Manufacturers' Record, Baltimore, Nov. 23, 1882). " We are persuaded that once the folks in New England, who have surplus money awaiting employment, thoroughly investigate the points Richmond presents for a safe lodgment of that capital in manufacturing, the flow will start this way" (Dispatch, quoted in News and Courier, Charleston, March 25, 1881). 31 Cotton lands lacking value, planters requiring capital and profits being diminished in charges of middle men, the National Cotton Planters' Association of America in 1882 sponsored a scheme by which Southern farmers should erect spinning mills on the railroads, to be equipped with machinery supplied by Manchester manufac- turers and operated for three years by English workers; farmers would supply food for these operatives and pay four cents a pound for the spinning of their staple. The Manchester men would be guaranteed 10 per cent profit on their stock by mortgage on planta- tions. Thus Manchester would be certainly supplied with yarn, the Southern cotton growers could borrow on their lands from the Bank of England, all charges between field and mill would be saved, as well as interest on capital for buying cotton and all expenses for sale of goods ! (cf. Baltimore Journal of Commerce and Manufacturers' Record, July 15, 22, Sept. 2, 1882). The hardly more likely plan of the Region of the Savannah Colonization Association has been noticed already. The frequent reports in these years of English manufacturers or capitalists about to acquire extensive Southern cotton interests were without foundation. Lumber and minerals, too, were said to be constantly on the point of English exploitation. Cf. News and Courier, Charleston, Nov. 10, 1881 ; Baltimore Journal of Commerce and Manufacturers' Record, Aug. 26, 1882; Manufac- turer's Record, Baltimore, Nov. 23, Dec. 21, 1882; Feb. 8, March 8, 1883. A Southern cotton manufacturer, English by birth and early experience, who has seen the whole development, said that no Eng- lish capital came to mills of the South in any quantity (William Entwistle, int., Rockingham). This was confirmed by Tracy I. Hickman, int., Augusta. 240 THE RISE OF COTTON MILLS IN THE SOUTH [346 continuous exertion for the establishment of factories in the South." Projects for mills in Mississippi, Alabama and Tennessee were declared promising ; old enterprises were to be set running again and a Connecticut manufacturer wish- ing to relocate was to be brought to Huntsville. 32 With passing years welcome to Northern capital became, of course, more wide-spread, but it could not gain in sincerity. 33 Southern overtures to Northern capital were matched by Northern liberality toward Southern opportunities. It will be seen later that most Northern support came not from in- vestors as such, but from commission men, machinery makers and from manufacturers establishing plants in the cotton fields. The impression of the president of the Na- tional Cotton Planters' Association was well-nigh universal : " I have been for some weeks in New York and Boston, and I find capitalists entirely willing to back any scheme which is founded on any right basis. Cotton mills are espe- cially attractive. . . ," 34 Ready subscriptions from the North to the Atlanta Exposition had significance for the future. 35 Occasionally, even in the first years, the mountain moved to Mahomet, as the Southerners must have viewed it; Northern mills or cotton firms sought to manufacture in the South. There was even an instance, that of Athens, where a town by embarrassing delays lost the placement of a fac- 32 Huntsville Democrat, quoted in News and Courier, Charleston, July 30, 1881. An Atlanta man conducted New York capitalists in an inspection of a Georgia water power; it was bought to propel a cotton mill (Manufacturers' Record, Baltimore, March 1, 1883). " Outside capital ... is beginning to seek this Southern field to aid in a more rapid and thorough work of restoration. . . . This move- ment needs a wise encouragment by public and private approval " (News and Courier, Charleston, June 28, 1881). 33 Cf. Charlotte Daily Observer, Nov. 4, 1897 ; Tompkins, Storing and Marketing of Cotton, pp. 11-12. 34 Baltimore Journal of Commerce and Manufacturers' Record, July 1, 1882. 35 News and Courier, Charleston, April 1, 1881. Cf. ibid., May 21, 1881 ; Baltimore Journal of Commerce and Manufacturers' Record, Sept. 2, 1882; Daily Dispatch, Richmond, March 25, 1880; Manufac- turers' Record, Baltimore, Dec. 14, 1882, March 1, 1883. 347] THE ROLE OF CAPITAL 24 1 tory by a New York firm that was otherwise certain. 36 An interesting, though incorrect view, applicable at most to the South Carolina up-country, has it that " the first movement was from the North to the South. Northern capital was looking for investment ; due to the proximity to raw cotton and labor it came to the South. Camperdown and Bates- ville had only Northern money in them. The idea began to smoulder; other mills came into being, Southern enterprise appealing to Northern capital." 37 Unsolicited transfer of capital from North to South, especially where plants were founded outright by Northern interests, was not prominent in the opening decade of the period. 38 Given a Southern community of restricted means intent upon establishing a cotton mill and yet unable to appeal ef- fectively to general capital sources because venturing upon an untried experiment, and the natural thing happened : the local projector would exhaust home resources, likely secur- ing enough money to erect the building, and then would ask makers of cotton manufacturing machinery to take part payment in stock, and apply to commission firms handling goods to subscribe, usually in return for the agency for the product. Thus special inducements were offered and there was redoubled interest in putting the plant in operation promptly. " A promoter had to have his home money first. He would get, say, $50,000; he would go to the machinery men and explain that he had so much subscribed, and would 36 Baltimore Journal of Commerce and Manufacturers' Record, Oct. 14, 1882. "Mr. Boyd, a capitalist of Providence ... is in Georgia in behalf of several New England capitalists, and is pros- pecting for the best place in the State to erect a large cotton fac- tory" (News and Courier, Charleston, April 9, 1881). Cf. Clark, in South in Building of Nation, vol. vi, pp. 266-267; Dry Goods Economist, Jubilee Number, 1896, p. 79; Murphy, Present South, appendix, p. 317. 37 Mrs. M. P. Gridley, int., Greenville. 38 A South Carolina town refused to cooperate with a Philadelphia firm which wished to build a mill to use machinery from a Pennsyl- vania mill that had failed, but the community erected a factory on its own account, stimulated by neighboring Southern enterprise (J. A. Brock, int., Anderson). Cf. Tompkins, Cotton Mill, Commercial Features, p. 39. 16 242 THE RISE OF COTTON MILLS IN THE SOUTH [348 they sell him the equipment and how much would they take in stock. Commission and machinery firms would give him 40 to 50 per cent of his total capital. If a man had no pre- vious mill connections, his local subscriptions would be his sole backing.'' 39 It seems the best opinion that machinery manufacturers took more stock than commission houses — anywhere from a fourth to a half the price of the equip- ment, depending partly upon demand for their product at the time. 40 However, as will be seen, commission firms often supplied ready money for working capital. Even as to these predisposed helpers there was weight in the re- flection that " nothing so attracts investors in other States as the knowledge that people on the ground have proved their faith in an undertaking by putting money in it." 41 39 W. J. Thackston, int., Greenville. " In most places where a new mill is proposed, an idea is prevalent that if half the money is raised at home, then somebody 'from somewhere will furnish the other half" (Tompkins, Cotton Mill, Commercial Features, p. 39). This statement needs the reminder that local ponds were regularly dragged and redragged. > 40 W. J. Thackston, int., Greenville; W. W. Ball, int., Columbia, Jan. 3, 1917. 41 News and Courier, Charleston, March 8, 1881. " Books have been opened in Newton, Alabama, for subscriptions to a cotton fac- tory at that place, and Northern capitalists have pledged $100,000 as soon as Newton raises $50,000" (Athens Banner, quoted in Balti- more Journal of Commerce and Manufacturers' Record, June 24, 1882; cf. Manufacturers' Record, Baltimore, March 8, 1883). When a movement was started for a factory at Vicksburg, nearly $200,000 was subscribed in the city, and it was expected that as much more would come from the " East," and that these latter stockholders would manage and equip the plant (News and Courier, Charleston, Aug. 12, 1881). Commission houses participated in the Charleston Manufacturing Company: "... the books of subscription to the stock of this company were closed yesterday. . . . Our citizens re- sponded well to the call made upon them, and the full amount of stock desired in Charleston for the immediate organization of the company was subscribed . . ." (ibid., March 16, 1881 ; cf. ibid., March 15). Of $500,000 wanted for a factory at Gaffney, it was felt that $200,000 could be raised in the county. " The other $300,000 will be obtained at the North" (ibid., Oct. 24, 1881). A typical distribution of stock was that of the Clifton mill, with half of its $500,000 capitalization paid up ; $50,000 was held in Boston, $150,000 in Charleston, and $200,000 in Spartanburg, the latter being the local community (ibid., May 21, 1881). Cf. Blackman, p 17, as to Pied- mont. Half the stock of Langley was held in New York, the other half equally divided between Augusta and Charleston (ibid., p. 7). 349] THE ROLE OF CAPITAL 243 Sometimes it was proposed to inaugurate a small mill completely and then seek outside aid in extension of plant : " It is said that fifteen thousand dollars will set in motion over five hundred spindles, and continual additions can be made. . . . We believe there is money enough in the county, here and there, to make at least a modest beginning so as to attract outside capital." 42 " If a Southern promoter had no business connections at the North, he went immediately to the machinery and com- mission men as those most interested." 43 It will be inter- esting to notice some details of this soliciting method. Many Southern mill projectors entering uncertainly an un- familiar field would stop in Baltimore to call on the com- mission house of Woodward, Baldwin & Co., and there were given a kindly reception that bolstered up self-confi- dence. "Many times we would, not know promoters that came, but going about the South we would hear of their en- terprises. They would bring letters of introduction, and be in town several days. A party of gentlemen might put up money to erect the building and buy machinery, coming out of the arrangement with more or less indebtedness to the machinery people and lacking working capital. The propo- sition would be broached to us to take the account of the mill and put up sufficient money to operate the plant. In other cases they would set out to raise $100,000, get half this amount, and come to us. We would subscribe to some 42 Winnsboro News, quoted in News and Courier, Charleston, Feb. 8, 1881. Cf. item from Chester Bulletin in ibid. The first mill at Gaffney (the initial enterprise projected, referred to in the previous note, did not eventuate) had $50,000 capital and 5000 spindles and drew nearly entirely upon local resources. When a second plant was built, costing $800,000, machinery firms took much of the stock (H. D. Wheat, int., Gaffney, S. C, Sept. 13, 1916). A small Texas mill had room for more machinery. " Up to this time home capital alone has been put in it. An invitation is extended to foreign capi- talists for more capital. At least $50,000 more can be used to great advantage" (Baltimore Journal of Commerce and Manufacturers' Record, Aug. 12, 1882). Cf. as to Rock Hill Factory, News and Courier, Charleston, Jan. 12, 14, 1882. 43 Tracy I. Hickman, int., Augusta. Cf . Copeland, pp. 49^50. 244 THE R J SE OF COTTON MILLS IN THE SOUTH [35O of the stock and then see friends who might be interested in the project and secure additional subscriptions." 44 Sometimes it was felt a commission house was tying too many strings to its offered subscription, and its assistance was refused ; on the other hand, a firm subscribing uncon- ditionally might create an impression resulting in its receiv- ing the agency. 45 If the Cotton Mill Campaign in the South did not evoke great outward misgiving on the part of the New England industry, it was reflected in the sharply increased business 44 The commission firm might thus lend influence that was as valuable as direct participation. " We generally required that they should have local subscriptions, local officers and local board. They usually had this arranged for. Then they would perhaps put in a member of the firm of subscribing agents who would attend the meetings. This kept the commission house in touch with the mill, but the business of the firm was to sell and not to manufacture the goods. I do not know that we exacted this last as a requirement. It was recognized as a proper thing to do ; the mills wanted it " (Summerfield Baldwin, Sr., int., Baltimore). This firm, especially through William H. Baldwin, had much to do with the establishment of the industry in the South, and everywhere in the mill districts one hears it cordially spoken of. Participation of this and a Boston house in the Charleston Manufacturing Company was representative (cf. News and Courier, Charleston, March 29, May 17, 1881). Much stock was secured in Boston through friendly offices of the commis- sion merchants there (William M. Bird, int., Charleston). Lock- wood, of Providence, engineer for the mill, took stock and influenced his friends to do so (A. B. Murray, int., Charleston). Southerners who had before hardly more than seen a cotton mill, were given rapid acquaintance with the industry by visits to plants with their Northern allies. 45 A. N. Wood, int., Gaffney, S. C, Sept. 13, 1916. "A Boston man told me he would take stock if the King Mill would make col- ored goods and give him the selling agency. I told him that if he took only a little stock he would have little to say about how the goods were manipulated, and if he didn't take any, he wouldn't have anything to say" (Charles Estes, int., Augusta). As the market for machinery in the South developed, equipment manufacturers became readier participants. An active solicitor who placed little stock of his first mill, in the eighties, with them, had their willing help for a second plant, erected ten years afterwards. In the first instance the enterpriser utilized any connection he had with men of wealth, however slight. He approached many persons he did not know at all. Often the commission house for another Southern mill would be appealed to. A Southern company proposing to buy ma- chinery outright, likely going in debt for part of it. might find the manufacturers willing to take stock, and so would increase the capi- talization of the mill. Cf. Manufacturers' Record, Baltimore, March 22, 1883; News and Courier, Charleston, April 6, 1881. 35 THE ROLE OF CAPITAL 245 of Northern machinery manufacturers. For obvious rea- sons, most of the machinery was of American rather than English make. Unacquainted Southern spinners could not buy from a distance even had they not required the close credit relations which domestic machinery men were ready to give them. Just at first some second-hand equipment was installed, less from desire of New England mills to put this off on their new rivals, than from innocence and neces- sity of Southern beginners. The number of Southern spin- dles, which had increased during the seventies from 327,871 only to 548,048, from 1880 to 1890 increased to 1,554,000 and by 1900 stood at 4,299,988; Southern looms' in 1870 numbered 6,256, advanced by 1880 to 11,898, shot up in the next decade to 36,266, and in 1900 were 1 10,01 5. 46 Yarns that had been selling at 14 and 15 cents went in 1880 to 28 cents; established mills joined with new plants in rushing, for machinery, for more money was to be made now in six months than before in two years. Prosperity resulted to makers of all equipment. 47 In June of 1882 a single South- ern railroad transported twenty-two car loads of machinery from Boston shops, and the same manufacturers notified carriers that they were working on three hundred car loads to be delivered by early fall. 48 Shops which made for the Southern trade enlarged their capacity. 49 It was reported 46 U. S. Census of Manufactures, 1900, Cotton Manufactures, pp. 56-57. 47 William Entwistle, int., Rockingham. Southern mill men did not balk at the high price caused by a 37H per cent tariff on ma- chinery, though it was occasionally complained of. The whole in- crease of business of machinery makers was not due, of course, to the South; the entire industry was reviving from the depression since 1873 ; probably, however, a larger proportion cf Northern mills installed English equipment. 48 Baltimore Journal of Commerce and Manufacturers' Record, July 22, 1882. Twenty-four car loads came to Augusta in one week for the King Mill (Manufacturers' Record, Baltimore, Jan. 18, 1883). Cf. quotation from Detroit Free Press in Observer, Raleigh, Aug. 31, 1880. 49 Cf . Baltimore Journal of Commerce and Manufacturers' Record, July 29, Sept. 2, 1882; Manufacturers' Record, Baltimore, March 8, 1883. 246 THE RISE OF COTTON MILLS IN THE SOUTH [352 that in two and a half years after 1880 the South had in- vested $12,000,000 in cotton machinery. 50 Southern mills with new machinery throughout (it was quickly learned that old equipment was a bad (bargain at any price) had an advantage over Northern mills that con- tributed to profits. 51 This spurt of the machine shops was caused both by anx- iety of Southern mills to buy and solicitude of Northern makers to sell. They were kept closely in touch with one another by trade papers. 52 Agents were sent to at least one Southern State to encourage the building of mills. 53 Mill engineers buying machinery for new plants sometimes col- lected commissions from the makers, and superintendents were accused of the same practice. 54 50 Baltimore Journal of Commerce and Manufacturers' Record, July 29, 1882; cf. quotation from American Machinist in ibid., Aug. 19, 1882. New orders were being chronicled continuously; cf. Manu- facturers' Record, Baltimore, Nov. 30, 1880, March 8, 1883; Balti- more Journal of Commerce and Manufacturers' Record, June 17, Sept. 9, Oct. 7, Nov. 18, 1882; News and Courier, Charleston, Feb. 26, March 4, 25, 1881. The demand for machinery is indi- cated in the fact that intending users met the makers much more than half way to investigate the Clement Attachment (cf. Blackman, pp. 18-19). Also, worn machinery was in some cases transferred from one Southern mill to another, though rapidly depreciating in effectiveness (John W. Fries, int., Winston-Salem; Manufacturers' Record, Baltimore, Dec. 28, 1882). In this connection it is interest- ing that some New England mills in the active period following the Armistice, despite full knowledge of its drawbacks, have been driven to install used equipment. By 1884 the shops were over the first rush ; times were disturbed, and this idleness, as will be seen, made them anxious to stimulate Southern business again. Cf. Chronicle, Augusta, Aug. 24, 1884. 51 Cf . Southern Cotton Spinners' Assn., proceed. 7th Annual Con- vention, p. 67. Probably no Southern plant installed altogether old machinery (H. D. Wheat, int., Gaffney). 52 Cf. Baltimore Journal of Commerce and Manufacturers' Rec- ord, June 10, July 15, Aug. 5, Sept. 30, 1882; Manufacturers' Rec- ord, Baltimore, March 1, 29, 1883. 53 Thompson, pp. 65-66. 54 Charles Estes, int., Augusta. Equipment manufacturers might supply building plans to companies which did not engage an engi- neer (William Entwistle, int., Rockingham). Besides, it was said in 1882 that a dozen young Southerners were gaining experience in Lowell machine shops, to be able to operate mills at home afterwards (Baltimore Journal of Commerce and Manufacturers' Record, Aug. 26, 1882). 353] THE ROLE OF CAPITAL 247 The principal difference between the participation! of commission houses and machinery makers lies in the fact that the former regularly retained their stock in Southern mills, whereas the latter realized upon their holdings as soon as possible. The one group, therefore, so far as con- cerns investment, had a continuing and the other a passing connection with the industry. Machinery men accepted stock simply in order to sell their product ; having no rela- tion to the output of the factories, they had no wish for a voice in their conduct. 55 It is generally declared that Southern mills have no com- plaint against their treatment 'by machinery firms, these having been liberal and cooperative, except that the industry as a whole and individual plants were encouraged to ex- pand beyond wise limits to create a market for equipment. 56 55 Confirmation of this point is universal. Sometimes they would sell immediately at a discount, having loaded the price of the ma- chinery to compensate for this loss ; sometimes they waited longer, but rarely held on for dividends (interviews with Summerfield Bald- win, Jr., Baltimore; W. W. Ball, Columbia, Jan. 3, 1917; Sterling Graydon, Charlotte; J. W. Norwood, Greenville, S. C, Sept. 9, 1916; Joseph H. Separk, Gastonia ; F. Q. O'Neill, Charleston, Dec. 27, 1916). " The machinery men sold out at 94 or 95. I told them to retain their stock, that it would be profitable. But they replied that to sell was their practice" (William Entwistle. int., Rockingham). Most of the shares thus thrown on the market came into local ownership. " At a South Carolina print-cloth mill I was told that 62 per cent of the stock was originally held by Northern machinists, but that by now it had all come South again and was held to a large extent locally" (see T. W. Uttley, Cotton Spinning and Manufacturing in U. S., pp. 46-47). Sometimes, certainly in South Carolina, when a mill was to be built, dealers in tin, lumber, brick and paint would be asked how much stock they would take if awarded contracts (J. B. Cleveland, int., Spartanburg, Sept. 8, 1916). It is likely that such shares were quickly resold. 56 David Clark, int., Charlotte ; J. W. Norwood, int., Greenville. South Carolina mills, with capital in larger units, may have suffered more in this regard than North Carolina plants. It is said that re- bates and other benefits were given to purchasers of machinery for cash, while buyers on time were assured the latter was as advan- tageous a plan. Tompkins was the Southern representative of many machinery firms ; besides plants that he built outright, he equipped perhaps 150 mills. Both as commercial agent and as publicist he encouraged erection of factories, but never with any hint of chi- canery attaching to his activities, though he was obliged to have some factories sold for debt (Sterling Graydon, int., Charlotte). Not all agents have had this record ; a North Carolina mill had to 248 THE RISE OF COTTON MILLS IN THE SOUTH [[354 A common failing of the mills, in the train of which came many embarrassments and drawbacks, was 1 lack of work- ing capital. This was due partly to insufficiency of ready money in the South, partly to need of large sums for pur- chase of quantities of cotton for coarse spinning, and was not unconnected, perhaps, with willingness of a community unacquainted with industry to rest satisfied when visible in- vestment had been cared for. Also, original provision for enlargement of plant, while ultimately economical, was im- mediately expensive, rendering some capital unproductive. 57 Tompkins was giving advice for the typical 1 enterprise when he counselled that $75,000 was the least that should be subscribed for a mill in a new section, and that while this did not allow of 10 to 20 per cent of capital stock being set aside for working capital, as was wisest, still most com- panies started without this facility, either borrowing from a home bank or consigning product to a commission house and drawing against it for 75 to 90 per cent of its value. So far from possessing running capital, mills often began operation with indebtedness on the plant. 58 be reorganized before it commenced operation, and while local in- vestors were scared off by the project to double its spindleage, machinery manufacturers encouraged the enlargement and took pre- ferred stock, of which there came to be an actual majority. A ma- chinery representative is president. The mill is less successful than those about it. 57 Southern cotton factories from the beginning have developed water power, installed boilers and erected buildings with a view to extension ; this showed as much faith in the future of the industry on the part of projectors as it did solicitude for future work on the part of engineers and machinery manufacturers supplying plans (cf. Uttley, p. 47). Sixty-four per cent of new spindles in 1903 were credited to established mills (ibid., p. 44). A 25,000-spindle, 600- loom mill that cost, with provision for 15,000 more spindles and their complement of looms, $28.65 per spindle, was to cost, when fully equipped, $16 per spindle (ibid., p. 48). Here was a wider difference than Tompkins calculated when he said that building with a view to doubling capacity would cost about 7 per cent more at the outset (Cotton Mill, Commercial Features, p. 51 ff.). For typical instances at the opening of the period, cf. Baltimore Journal of Commerce and Manufacturers' Record, July 22, 1882, as to Rome Cotton Factory, and ibid., Oct. 28, 1882, as to Huguenot Mill. 58 "The capital proposed for the Darlington mill is $280,000. Of this $140,000 has been subscribed and the work of construction be- gun" (News and Courier, Charleston, "South Carolina in 1884"). 35 5D THE ROLE OF CAPITAL 249 Hurtful refusal of Southern banks to cash drafts for goods sold directly to Southern merchants, remarked by a Virginia governor in the fifties, 59 became a no less compro- mising inability to finance the industry in the eighties. Nor could Southern mills look to Northern banks — the industry was too untried, the banks too far away. The factories needed friends at court, agencies involved in the future of the new enterprises. Commission firms were the natural resort. Said a manufacturer not liking this recourse : " The Yankees asserted first we could not run because of climate. We did it. Then they said we could make coarse goods, but not fine goods. Well, we are doing it. The North made two guesses out of three, and was mistaken in them. It did not make the guess that we could not run without money. But you find Southern mills failing and going to the Eastern men to whom they were in debt." 60 Believing need for working capital had led mills into damaging connections with agents for goods, it was declared that companies would have done better to depart from practice and issue bonds at the outset to provide themselves " One of the special disadvantages under which Southern mills have to work is that often they have very little working capital and at the beginning think that all they have to do is to pay for the mill, if, indeed, they do that" (S. S. Broadus, Decatur, letter). "Most Southern mills when built had to borrow their total working capital, and this is still done. Many borrowed to build the plant. It is not wrong for a mill to borrow a little fixed capital, say $3 or $4 on the spindle" (Summerfield Baldwin, Jr., int., Baltimore). 59 Cf. Clark, in South in Building of Nation, vol. v, p. 324. 60 J. H. M. Beatty, int., Columbia. Mr. Copelandhas recognized that " It is available cash rather than geographical location, which determines who will be able to buy cotton when the price falls " (see PP- 36-37)- "The unsuccessful mills are often so because of slavery to the commission houses through which they sell their product. Too many Southern mills have been built with insufficient working capital or with none at all. . . . The commission houses, many of which have banking connections, gladly advance 75 to 90 per cent of the market value_ of unsold goods, charging the mill double the rate of interest which they themselves must pay for the money. Thus interest charges often eat up profits" (Thompson, pp. 89-90). Mr. Law referred to commercial paper placed with banks, and not to more entangling alliances, when he declared latterly that a cotton mill is properly a seasonal borrower for the purchase of raw mate- rial (John A. Law, Cotton Mill Credits, in proceed. Robt. Morris Club, National Association of Credit Men, 1916, pp. 24-25). 250 THE RISE OF COTTON MILLS IN THE SOUTH [356 with ready money. " A mill could issue $500,000 in bonds at 6 per cent, carrying this $30,000 interest and pay a man in New York $15,000 to sell its product. For borrowing the same sum from the commission house, the mill must pay $50,000, and for this privilege must give, besides, a terrific bonus of 4 per cent on sales. The difficulty could have been avoided had the mills been capitalized at enough in the beginning." 61 With the exception of the labor factor, scarcely any rela- tions of the mills have been so continuous and uniform through the history of the industry as those with commis- sion houses selling the product. It has been explained how, through taking stock and lending working capital, these firms became a characteristic part of the enterprise. It is very difficult to generalize as to whether their influence has 61 Tracy I. Hickman, int., Augusta. " Mills with no working capi- tal began life with their credit strained; so they were in the grip of commission_fjims which lent them money" (W. W. Ball, int., Co- lumbia, Jan. 3, 1917). Lack of capital sometimes entailed equipment \ with second-hand machinery. This was probably the cause with Montgomery's first mill at Spartanburg, built by instalment pay- ments (L. G. Potter, int., Gaffney, S. C, Sept. 13, 1916). Later, a mill at Bessemer City, N. C, installed old machinery and put a mort- gage upon it (G. W. Ragan, int., Gastonia). For a typical advertise- 1 ment of old machinery for sale by a New England mill, cf. News/ and Courier, Charleston, April 19, 1881. The establishment of Pied- mont fell in hard times ; for three months while the machinery was being installed, the only pay of the workmen was credit for groceries at a store in Greenville, the mill giving a note in guarantee (W. J. Thackston, int., Greenville). A commission firm was of great assistance. Hammett had to mortgage some of his private property (James D. Hammett, int., Anderson, Sept. 11, 1916). The Charles- ton Manufacturing Company had contemplated setting aside $75,000 of the $500,000 capital for preliminary and running expenses, but quicksand was struck in making the foundations and this money had to be spent in piling (William M. Bird, int., Charleston). Lack of running capital precluded success. Bonds to the extent of $250,000 were issued soon after operation commenced (A. B. Murray, int., Charleston). It is asserted that the King Mill, at Augusta, is the only one built and run within its original capital stock; a fourth of its million-dollar capitalization was reserved for running expenses (Charles Estes, int., Augusta; a contemporary newspaper article gives a maximum of $181,979.57 for this purpose; cf. Evening News, Au- gusta, Jan. 23, 1884). In the depression in 1884 mills without surplus capital were seriously embarrassed (Chronicle, Augusta, Sept. 11, 1884). It is said that bad management was responsible for fewer failures than insufficient working capital (Tracy I. Hickman, int., Augusta). 3573 THE ROLE OF CAPITAL 25 I been good or bad. It is best to understand that, for a good many years at the outset, mills could not have come into being without assistance from commission men, and then, in the spirit of Tompkins' position that well-disposed houses have been advantageous allies but that many abuses need to be eliminated, to state the usual complaints and defenses. It was early felt that if a mill was to secure full prices and prompt sales, the executive officer must be a good mer- chant, correcting his reliance on the commission firm by personal knowledge of markets. 62 After some years of ex- perience, reasons for this circumspection could be enumer- ated. Pointing out that selling agents found subtler means of exploitation than the commission rate, a standard 5 per cent, a manufacturer said that an indictment of firms guilty of malpractice would include the following counts: (1) Beating other commission men to the market was preferred to shuffling in the competition and trying to make a sale that would in the end, if successful, bring a higher commission — so they would offer the mill's product at a figure below the current price. (2) Agents would sell a mill out further ahead than was wise from the manufacturer's standpoint. (3) Suppose a mill has been holding its product in hope of a rise, but the commission house, advising that no- advance will come soon, dumps the goods on the market. The market falls further. Then, to prevent product from piling up in stock, the agents apply to sell it. It has been sus- pected that commission firms sometimes in such cases sold goods to themselves for purposes of speculation, they being in a position to know that the market would recover later. Here they reaped their percentage on bogus sales and a profit on the speculation. Where a house acts as exclusive agent for the output of &■ factory, guaranteeing accounts, 62 Cf. comment on a statement of Hammett's, in Manufacturers' Record, Baltimore, Feb. 1, 1883. There were many later expressions; cf. Tompkins, Cotton Mill, Commercial Features, p. 128 ff. Precau- tions were to be taken at the very start; in deciding on the goods a new mill should make, it should be remembered that the commission firm was likely to advise that particular product in which it special- ized, and this would not necessarily be profitable in the long run (ibid., p. 56 ff.). 252 THE RISE OF COTTON MILLS IN THE SOUTH £358 it is best considered that the goods are sold to the house out- right, to do with as it will. It is the duty of the mill man- agement under such an arangement, of course, to keep up with the market and insure fair play. (4) A commission firm might tell a mill its product was bringing less than was the case, the firm keeping the difference. This was theft. (5) Where the yarn of a mill was made of cotton superior to that used by its competitors, the agents would not al- ways take pains to explain this and so get a better price than the average ; demand for yarn of that count might be supplied from lower grade goods, the difference in commis- sions not making it worth while to push the finer product. And yet the commission men were supposed to represent the interests of their clients. (6) A firm might even seek to accomplish the buyer's interest exclusively. Suppose it did not handle all the product of a mill, but sometimes placed it in specific orders. Seeing a chance to sell goods, the commission house, knowing the factories of the district, would shop around and get the bottom price from the most necessitous maker, and offer this to its client. 63 63 Sterling Graydon, int., Charlotte. Several of these and addi- tional points are brought out by Tompkins (Cotton Mill, Commer- cial Features, p. 128 ff.) ; agents have thrown cancelled orders back upon the mills without much scrutiny into claims of the purchaser; on the other hand, they have been too accommodating about holding goods and advancing money on these, so that interest account ab- sorbed the mill's assets. It has been a custom for commission firms not to reveal to mill managements names of customers. Out of this has grown much suspicion and perhaps some misdealing. Said an official : " Some years ago a mill in which I was a stockholder had a lot of goods piled up while prices were at rock bottom. The com- mission house wanted to sell the goods. The mill begged the house not to sell; we were confident the market would recover. But the firm wrote back that the mill owed it $200,000, and that unless this was paid the stock would be sold. The commission house knew the mill could not pay the amount then, and the goods were disposed of at a great loss. I have always thought the selling agent bought on his own account." In another instance a mill received through a commission firm an order for 10,000 pounds of yarn to be delivered each month for ten months. The mill shipped the first batch, but when time came to deliver the second, was advised by the house that the market had gone off some points, and not to deliver. Months passed with no more deliveries until the tenth month arrived. By now the price was above that named in the contract, and the mill was instructed to ship goods covering the entire order. Expostula- 359] THE ROLE OF CAPITAL 253 Observing that a mill partly owned by a commission house frequently must see its goods sold under the market and that factories that would not have come into being ex- cept for assistance of selling agents were built as feeders for commission firms and not to make money for local stockholders, a progressive manufacturer advised that in- terference by agents be eliminated at any cost ; on the whole, their influence in the Southern textile industry has been bad. 64 Stock in mills about Greenville, once largely held at the North, is coming to the locality of the factories ; com- mission men were not sorry to ruin a mill if they could be indemnified in charges first. 65 The story is not one entirely of condemnation. Some commission firms have had close and thoroughly helpful relations with Southern mills, lending not only financial support but valuable business judgment. Rates of commis- sion have declined one-third since the eighties. 66 Many bad tion was unavailing, the commission house replying that the customer would be lost if pressed upon the fault, and that unless the yarn was delivered the house would sever relations with the mill. The director who wished to sever connections was bought out. It was his con- viction that the selling agent was his own touchy customer. This sort of experience has led many to the conclusion that the only way is to sell to the house direct, demanding payment from it and not knowing or caring where the product goes (Charles E. Johnson, int., Raleigh). There was a famous case in South Carolina where a mill owing a commission house an amount equal to a fourth of its capi- talization and feeling itself mistreated, changed agents, borrowing from the new firm money to pay the old debt. But the offended agents secured enough stock to control the mill, and finally sapped it. The same commission house, it is asserted, tried in every way to show that colored labor was unprofitable in a mill in which it had been installed in an extremity. Commission men, especially where they had banking interests, found it profitable to borrow money in the North at 3 or 3^ per cent and lend it to Southern mills at 6 and 7 per cent. This caused much dissatisfaction, particularly when the sums going to the agents in commission were considered (Tompkins, ibid.; William M. Bird, int., .Charleston). Commissions have in special cases, where a firm was eager to secure agency of a mill, been as low as 2 or 3 per cent (F. Q. O'Neill, int., Charleston. W. W. Ball, int., Columbia, Jan. 3, 1917). 64 Joseph H. Separk, int., Gastonia. 65 Clement F. Haynsworth, int., Greenville. Lancashire half a century after the South, developed the same devices for financing new mills as have here been described, with the same drawbacks attending. Cf. Copeland, pp. 317-318. 66 Summerfield Baldwin, Jr., int., Baltimore. 2 54 THE RISE OF COTTON MILLS IN THE SOUTH [360 practices have disappeared. It is probably true that along with regret for early faults must go the recognition that, whether participation of commission men was damaging or beneficial, it was necessary and that the industry owes its establishment as much to them as to any other factor. 67 It is to be borne in mind, too, that the mills were started by men new to that industry and, in many cases, to all indus- try; these might be too quick to charge exploitation by powerful agents a thousand miles away in the North. Also, selling houses had the money motive, not the patriotic one present with local projectors. 68 There is an evident movement in the South away from any reliance upon outside commission houses. This mani- fests itself in action by sales through brokers or directly to jobbers and mills, by patronage of Southern selling agents, and by establishment by groups of mills of sales offices in the North. Moreover, that the South is thinking of a still better solution is apparent in frequent mention of the pos- sibility of building up a distributing point within the sec- 67 J. A. Chapman, int., Spartanburg; C. S. Morris, int., Salisbury. 68 As early as 1903 a spirit of conciliation toward commission men, of letting bygones be bygones, was shown in the Southern Cotton Spinners' Association (proceed. 7th Annual Convention, p. 162). It should be clear to the reader why selling houses, in contrast to machinery manufacturers, retained their mill shares. Critics at the South have charged this degree of control permitted peremptory methods in disposal of product that proved harmful; on the other hand, this may be interpreted as a proper watchfulness over enter- prises being credited with large sums at a time when money could not be gotten by them otherwise (cf. Copeland, p. 197). "This Southern development would have been delayed twenty years if the commission men had not taken hold. Southern promoters absolutely did not have enough money at home to accomplish it. Besides, it was credit extended to projectors of mills by selling houses that gave them a measure of credit with machinery people." Commission firms have had to keep large capitals ready against calls of their Southern clients; over advances have usually been made just on the credit of the mill (Summerfield Baldwin, Sr., int., Baltimore). Frequently money was lent to mills at less than local rates, even had banks been willing to furnish funds (Washington Clark, int., Co- lumbia. Cf., as to slightness of available banking facilities, Ham- mond, pp. 160-161). Commission percentages were not so high where mills were not heavy borrowers (J. A. Chapman, int., Spartan- burg). The fact that selling houses placed little real money at dis- posal of the mills, except in advance on goods, does not modify the importance of their assistance. 36l] THE ROLE OF CAPITAL 255 tion that will make dependence upon Philadelphia and New York no longer necessary. It is generally recognized that this last cannot be accomplished until finishing plants have become numerous in the South, until financial support of outside selling firms is not required, and until diversity of prices for identical product, largely resulting from par- ticipation of commission firms, disappears. The Panama Canal, development of banking resources, competition itself through extension of the industry, and, all in all, total maturing of Southern economic life, will hasten realization of this ideal. 69 There are enough who believe that the strong tradition, mainly in favor of New York, cannot be broken down, and that other obstacles preclude complete conduct of the industry, in manufacture and commerce, within the South. But others accept and challenge these difficulties in much the same spirit that characterized the founders of mills forty years ago. These cannot see the logic of sending goods to the North to be bleached and finished when the South has every facility for these proc- esses if only they are taken advantage of ; if it was once raw cotton that was to be manufactured near the source, it is now goods in the grey that are not to be wastefully shipped away and shipped back. Before the Southern industry as such took its rise, before Northern selling firms came into the field, the majority of mills relied mainly upon local, or certainly Southern, de- mand for their product. Exchange of yarn for butter and beeswax by the smallest factories was matched in some- what more extended barter of larger mills. Such practices survived the war, and were occasionally present even after a commission house took charge of part of the output of a plant. 70 Even in the eighties some mills had traveling sales- men covering the South, and others were terminating brief connections with commission houses and selling direct or 69 George W. Williams, Charleston ; Thomas Purse, Savannah, Ga., Dec. 26, 1916, interviews. 70 Cf. Thompson, p. 51; Daily Dispatch, Richmond, Jan. 2, 1880; William Banks, Columbia, and W. J. Thackston, Greenville, gave instances in interviews. 256 THE RISE OF COTTON MILLS IN THE SOUTH £362 through brokers (the latter giving the mill right of accept- ance or rejection of orders, costing far less and involving risk that was small beside commission charges) ; but for the most part this was the day of leaning upon selling agents. 71 From this dependence the South is emerging and returning to the old self-reliance, but a self-sufficiency accompanied by improved facilities and wider outlook. Distributing companies formed by chains of Southern mills, especially where under united ownership, are becoming fac- tors in Northern markets; with more commercial knowl- edge direct selling by individual mills loses its dangers. 72 True, international representation has not yet come, but this may lie just over the horizon. 73 It has been said that merchants were often mill builders. These sometimes got outside capital from other sources than machinery and commission men, namely, from those with whom they had commercial dealings, but even in this instance there were special reasons prompting investment. Concord presents typical cases. The largest manufacturer of the place went to the town as clerk in a general store and later went in business for himself. He determined to build a factory, securing some $60,000 in local subscrip- tions. Then he went to the firms from which he bought 71 Copeland, p. 216; Charles Estes, Augusta; W. R. Odell, Con- cord, interviews. 72 Copeland, pp. 172, 209; C. B. Armstrong, int., Gastonia; for ref- erences to new selling methods, see Charlotte News, Textile Ed., 1917. 73 Cf. Law, in proceed. Morris Club, National Assn. Credit Men, 1916, p. 23 ; Tompkins, The South's Position in American Affairs, p. 5 ff . An interesting approach to later developments was the forma- tion of the Cone Export and Commission Co., sometimes known as the " Plaid Trust," in 1891. The firm sought to represent all the Southern plaid mills, these buying stock and securing a measure of control in the enterprise. It came too early to succeed, unity proving impossible; it can hardly be called an effort by a commission house to anticipate autonomous action on the part of the factories. (In- formation as to this project is contained in an announcement by the Company issued in May, 1891, giving quotations from the Daily Commercial Bulletin, N. Y., May 14, and Dry Goods Chronicle, May 23, 1891 ; cf. Copeland, p. 206, and as to a somewhat similar organi- zation, p. 161 ff.; Mrs. Moses Cone, Baltimore, Md., Nov. 14, 1916; Bernard . Cone, Greensboro; R. G. Vaughn, Greensboro, Aug. 30, 1916, interviews.) 363] THE ROLE OF CAPITAL 257 "brogans," and cloth and to which he shipped raw cotton, explaining to them his plans and showing that a mill would enable the town to grow and permit him to do a larger mer- chandise business with these wholesalers. It was almost worth the subscription to keep his business, so each firm bought $5000 of stock. 74 The only shares of another Con- cord company owned outside of North Carolina are held by Baltimore men who had business relations with the mer- chant who built the mill. 75 Sometimes mercantile connec- tions of many years before were recalled to serve a purpose in the eighties. 76 Much of the early mill building consisted of extension of plant by means of earnings. Tompkins indicated that profits to the amount of 5 per cent of capital were ordinarily de- voted to this extension, but in frequent individual cases much more than this was found available. 77 Often by the time a mill was put in operation a company had exhausted credit facilities and local capital resources; the large earn- ings brought a new increment of cash and additional com- mand over credit, which were employed in augmentation of plant. 78 Vaucluse was built from the profits of Graniteville and a third plant, the Hickman, was created with money borrowed. 79 While many mills with extended plant have little more 74 J. L. Hartsell, int., Concord. Years later, when merchandising is no longer thought of, this manufacturer can readily get subscrip- tions, it is said, from retired wholesale dealers of New York, Phila- delphia and Boston (Charles McDonald, int., Charlotte). 75 Ibid. 76 Charles Estes, int., Augusta. 77 Cotton Mill, Commercial Features, p. 172. Besides, he confused new construction with upkeep. 78 Benjamin Gossett, int., Anderson. There might be combination of methods. The 5000-spindle Williamston Mill issued extra stock to $300,000, increasing spindleage to 15,000; afterwards the plant grew to have 32,000 spindles, all on earnings and credit. 79 Tracy I. Hickman, int., Augusta. Cf. Blackman, p. 5. The Gaffney mill erected a two-story addition from the first three years' earnings (L. Baker, int., Gaffney). The Arlington Mill, Gastonia, organized with $130,000 capital and 3000 spindles, after three years issued a stock dividend of $45,000 and increased its spindles to 9500, and by later earnings enlarged to , 12,000. Cf . Charlotte News, Tex- tile Ed., 1917, as to Cliffside and McAden mills. 258 THE RISE OF COTTON MILLS IN THE SOUTH ^64 than their original capitalization, that of others' has been increased 1 by additional issues of stock, frequently to sub- scribers at a reduction. Though thus put out at 75 or 80, the industry was profitable enough to keep shares at the par of ioo. 80 "The stockholders of the Matthews Cotton Factory, at Selman," it was reported, "have resolved to in- crease the capital stock from $100,000 to $300,000. Ex- tensive plans for enlargement have been determined on, and they will be commensurate with the amount of increased stock taken." 81 The Anderson mill, capitalization of which was raised from $100,000 to $250,000 and by three more increases to $800,000, had a debt on the plant in the begin- ning; earnings going to take care of this, and further credit probably being difficult, stock issues were resorted to for enlargements. Machinery and commission men participated heavily. 82 Local brokers negotiated for the entire addi- tional stock of the Enterprise Factory, Augusta; it was un- derstood that one man and his friends would take $140,000 of the $350,000 issue. 83 It was characteristic of the establishing of an industry among people not very familiar with financial devices, with few investors beside those interested in cotton rnanufactur- 80 Increasing capital to $1,000,000, stock of the Sibley mill was offered to original subscribers in pro rata amounts, any not so taken to be sold in the general market at not less than par. The directors were empowered, too, to issue $100,000 in bonds (circular letter of William C. Sibley, Augusta, April 26, 1882, in Raworth scrapbook). It is said now that $500,000 in stock should have been issued to build the Hickman mill, rather than borrowing for this purpose (Tracy I. Hickman, int., Augusta). 81 Manufacturers' Record, Baltimore, Dec. 21, 1882. At reorgani- zation of the Charleston Manufacturing Co., $300,000 fresh capital was subscribed and equipment of the plant was completed (A. B. Murray, int., Charleston). 82 J. A. Brock, int., Anderson. Most of the original shareholders increased their subscriptions to the Cannon Mill before the plant was completed (J. L. Hartsell, int., Concord). Enlargement of capital and plant often was undertaken in this early stage. In a period of similar activity thirty-five years later, companies at Gas- tonia had hardly received their charters before deciding to increase capitalization. Clifton and Trough Shoals mills intended to double capacity and capital when a successful beginning had been made (Blackman, pp. 10-11; News and Courier, Charleston, Feb. 10, 1882). 83 News and Courier, Charleston, Feb. 24, 1881. 365] THE ROLE OF CAPITAL 259 ing as such, and with commission houses and machinery makers to assist, that bonds and preferred stock should have been employed rarely. " The people just put in their money and made it go as far as it would, without thought of preferred stock and bonds. Mills were generally small because the money did not go far." 84 Also, if fixed assets in land, buildings and machinery were mortgaged by is- suance of bonds, there was only material in course of 'manu- facture and finished product on which to base commercial credit. It is said that in most instances where bonds were sold, the practice was found to be bad. 85 No case has been discovered where preferred stock or bonds were issued at the outset ; it was when a mill got into trouble, needed addi- tional capital or had to be reorganized that such helps were turned to. 86 A mill at Bessemer City was under-capitalized, the projector could not persuade stockholders to increase their subscriptions, and so machinery could not be installed. The promoter built a second mill and a third which were 84 T. C. Leak, int., Rockingham. 85 W. J. Thackston, letter, Greenville, S. C, Nov. 25, 1916. 86 Tracy I. Hickman, Augusta ; F. Q. O'Neill, Charleston, inter- views. Cf. U. S. Census of Manufactures, igoo, " Cotton Manufac- tures," p. 31. The following from the report of the president of the Enterprise factory to the stockholders in 1885 is sufficiently revealing to bear quotation : " Four months ago, when the Board of Directors took charge of your property, they found it burdened with a floating debt of $200,550.25, largely the result of embezzlement on the part of its former President. . . . The company had also issued second mortgage bonds on property, to the amount of $150,000.00, with which to pay off such debts as were most urgent, but, not finding a ready sale for these bonds in their then crippled condition, the Presi- dent had hypothecated them to the extent of their issue, as collateral to secure creditors. ... It was determined to issue preferred stock to the amount of $250,000.00, and with the proceeds pay off the debts of the company, take up the second mortgage bonds, and operate the mill. Since then there has been $148,200.00 subscribed to the preferred stock, of which $85,750.00 has been collected. We have already taken up and destroyed $100,000.00 of the second mort- gage bonds, and paid $95,121.34 of the outstanding indebtedness. There remains $101,800.00 of the preferred stock not yet taken, but your Board believe, that in the improved condition of the company, it will not be necessary to dispose of more than one-half of it" to care for all indebtedness except $42,000 to machinery makers, to be paid by company's notes running for ten years. It was hoped soon to redeem the preferred stock (Raworth Scrapbook). Cf. Augusta Trade Review, Oct., 1884, as to Augusta Factory. 260 THE RISE OF COTTON MILLS IN THE SOUTH [[366 bonded. Neither venture succeeded. 87 Earnings having been slight because of a bad market for goods, it was de- cided in 1884 to cut down the proportion of overhead ex- pense of the Sibley mill by completing the equipment of the plant; this was thought advantageous, too, because makers were not busy and machinery could be purchased cheaply. Not all of the previously authorized bonds had been sold; the directors recommended an additional issue and urged that shareholders follow the example of the directors in subscribing as heavily as possible. This course would pre- serve the ownership of the property, make 6 per cent earn- ings possible and arrest the decline in value of stock. 88 In some instances, as in Columbia and Charleston, local banks were of substantial assistance in furnishing working 87 S. N. Boyce and J. Lee Robinson, int., Gastonia. 88 See annual report of President Sibley in Chronicle and Con- stitutionalist, Augusta, May 1, 1884. In two weeks nearly two-thirds of the bonds had been placed and the president, " notwithstanding the disappointments of the past," had "more faith now than he ever had in the final success of the company" (ibid., May 14, 1884). Striking difficulties with commencement of operation, the Charleston mill issued bonds to half the value of its property (A. B. Murray, int., Charleston). Later, an instructive operation took place be- tween an involved South Carolina mill and its commission house. The plant cost $21 per spindle — $1,748,000 — but there was a debt of $9.64 to the spindle, the company being capitalized at $976,700, all common stock. The net indebtedness had run as high as $830,000, but at the time of this episode it stood at $510,000. Trouble of a serious nature being discovered in the books, the mill would have gone into the hands of receivers unless relief had come. The com- mission house was so heavily interested that it had to act, and took $500,000 of preferred stock at par, though a banking house would not have given above 80. Other creditors insisted upon payment of their accounts, and the selling firm had to put up $560,000 to care for these items, making its total interest in the company well over a million dollars. Though the original debt of mill to commission house had been liquidated through conversion into preferred stock, by the whole operation this obligation was greater than before, and there was an increase in stock from $980,000 to $1,500,000. Condi- tions speedily improved from this date. In a case such as this the preferred stock would ordinarily be offered to the shareholders who, however, were not usually able to take it. Though there are state- ments to the contrary, one of wide observation did not know of an instance where a commission house bought into a healthy mill to gain control of it. Probably a similar happening was the taking of all the bonds of a mill by a firm of selling agents thirty years ago (A. A. Thompson, int., Raleigh). Cf. Columbia Record, Textile Ed., 1916, as to Lydia mills. 367] THE ROLE OF CAPITAL 26 1 funds, but such assistance was far from general. Reasons for this have been noticed. Banks were few and had slender resources. 89 Industry and banking, developing to- gether, were mutually helpful, but neither could greatly take the initiative. Interest rates were high. Even latterly, mills borrowed at 7 and 8 per cent, and besides were com- pelled to keep a balance of 20 per cent of the loans, on de- posit without interest. On the other hand, commission firms might lend at less than these rates and require no balance. Northern mills could get money from banks at half the interest paid in the South. 90 Southern banks as- sisted the mills indirectly to some extent by loans to stock- holders on their shares. 91 As early as 1884 it was suggested in the South that those contemplating the founding of mills should consider the plan of securing subscriptions from operatives, papular in England. 02 In the nature of things, the scheme could not have been employed in that stage of the development. Prob- ably it was not relied upon to any extent until, recently, half the stock in a Gastonia mill was taken by operatives. 93 Leaving now the means of acquiring capital, the subject of profits and dividends is to be examined. In 1890 it was estimated that for the country at large profits in cotton 89 Immediately after the war, the largest bank in Charlotte had a capital of only $20,000 (Hudson Millar, int., Charlotte). The rate of growth of Southern banking is eloquent of former leanness (cf. John Skelton Williams, The Billion Arrives, p. 7). 90 Summerfield Baldwin, Jr., Baltimore; Benjamin Gossett, Ander- son ; interviews ; S. S. Broadus, Decatur, letter. Generally, condi- tions have improved. A good mill can sell paper through brokers at 4 per cent and commission. (Summerfield Baldwin, ibid.; C. B. Armstrong, int., Gastonia). Richmond banks have come to bear important part in the Southern industry. One group of mills at some seasons has owed Richmond as much as $4,000,000. 91 August Kohn, int., Columbia. Barely worthy of mention is the fact that mills sometimes, with more or less surrender of strictly local control, got assistance from established manufacturers in the South. Or a factory might merge entirely with a group of mills skilfully managed. 92 An Augusta paper of Nov. 6, referring to earlier publication in New Orleans Times-Democrat, in Raworth Scrapbook. 93 Charlotte News, Textile Ed., 1917. In limited adoption, the plan has frequently been used as an employer's device. Cf. ibid., as to Saxony Spinning Mill, Lincolnton. 262 THE RISE OF COTTON MILLS IN THE SOUTH [368 manufactures, allowing only for ordinary repairs, were 7.59 per cent, or, deducting 3 per cent of the value of plant to care for depreciation, 5.83 per cent. 94 The estimated average rate of dividends paid by New England mills from 1889 to 1908 was y.y per cent. 95 Such calculations are not fairly comparable, and yet 'some statements as to earnings of Southern factories will help to give a notion of the rela- tive position of the industry in that section. Experience in the Carolinas showed that mills on all classes of goods made there could have a profit of from 10 to 30 per cent. 96 The same competent observer thought that the average annual net profit of the best mills for the first twenty years of the period was 15 per cent. 97 But assertions vary widely, even considering differences in periods embraced. One with knowledge of the industry believed average profits from 1880 to 1914 were not as high as 10 per cent, 98 while a writer on the subject said that South Carolina investors would have been better off financially had they put their money in real estate mortgages at 7 per cent. 99 Besides the confusion between profits and dividends, statements as to earnings are difficult of comparison be- cause uniformity of calculation was lacking. A common error was to quote dividends paid on capitalization rather than earnings on total investment. Plant cost of some milb was as much as four times their capital. Dividends might be paid on shares, neglecting large liabilities of stock- holders ; sometimes, also, gains seemed great because stated in terms of paid up capital only. 100 Even late in the period 94 U. S. Census of Manufactures, Cotton Manufacture, p. 167. 95 Copeland, p. 263. 96 Tompkins, Cotton Mill, Commercial Features, p. 51. That this was true for the smallest as well as the largest plants is verified by the case of the 1000-spindle Fingerville Factory (cf. Blackman, p. 11). 97 Tompkins, Cotton Mill, Commercial Features, p. 172; cf. Thomp- son, p. 88. 98 John W. Fries, int., Winston-Salem. 99 August Kohn, int., Columbia. 100 Thirty-five years after the mills around Greenville were founded, plants costing $21.08 per spindle were capitalized at $12.72 per spindle. Dividends on actual plant cost have not been over 12 per cent (W. J. Thackston, int., Greenville; cf. Goldsmith, p. 6). 369] THE ROLE OF CAPITAL 263 there was occasion for the advice of a trained manufac- turer that all profits not required for dividends be not passed to surplus without providing for depreciation, work- ing capital, and reserve for paying plant debt. 101 A friend of the Southern industry right at the outset sought to dis- credit the overstatement of promoters of the Clement At- tachment that 28% per cent could be made on a capital of $6000, showing that they had included no charges for super- intendence, commission and freight, insurance, taxes and wear and tear, and allowed too little for incidental ex- penses. 102 Mills in the beginning, from reasons that have been noted — general prosperity of the time, newness of equip- ment, nearness to raw cotton, cheapness of power and labor, length of working hours, unexploited home market — were extraordinarily profitable. "A cotton factory which is not making money now had better close up at once," it was said in 1880. 103 Two years later it was recited that the Augusta Factory in the previous seventeen years had paid an aver- age of 1^/2 per cent dividends besides laying aside a sur- plus of $350,000; Langley for several years had been pay- " In the old days it took four or five years to pay for the plant if they did not make the mistake of trying to pay normal dividends instead of liquidating this debt" (Benjamin Gossett, int., Anderson). It has even been said that all mills were undercapitalized and started out in debt (Marshall Orr, Anderson; A. N. Wood, Gaffney, inter- views). Machinery makers and commission firms, anxious to in- crease their business, were severely blamed for inducing too great extension without sufficient capitalization, but this writer found the conditions bettering latterly (Law, pp. 19-20). 101 Tompkins, Cotton Mill, Commercial Features, pp. 83-84. He was appreciative, however, of the wisdom of a surplus to guarantee equality of dividends. It will be noticed later that failure to allow for depreciation was all too common; cf. ibid., pp. 82, 172; Thomp- son, pp. 88, 67-68. 102 Daily Constitution, Atlanta, Feb. 20, 1880. For a few of many instances illustrating neglect to offset depreciation, see Blackman, pp. 7, 10, 13. " In no case have we heard of any mill declaring less than 10 per cent annual dividends, and in every case in which only this per cent was declared a large amount was taken from the earn- ings and used for repairs, additions to machinery and increasing the . . . capacity of the mills" (Baltimore Journal of Commerce and Manufacturers' Record, June 3, 1882). Here, evidently, upkeep and extension were looked to rather than depreciation. 103 Blackman, p. 12. This was probably Hammett. 264 THE RISE OF COTTON MILLS IN THE SOUTH \_Z70 ing from 15 to 20 per cent and had an accumulated surplus of $200,000. The Wesson mills in Mississippi had just paid a dividend of 26 per cent, the Troup factory in the same State had paid 24 per cent, and a mill in Tennessee had touched 50 per cent. Such facts were often cited to show that the Southern industry, though young and ham- pered by untrained management and operatives and lack of capital, was more prosperous than that of the North. A New England estimate was quoted to the effect that fifty leading establishments of that section in the previous five years had paid average annual dividends of less than 7 per cent. 104 The profitableness of Graniteville was often pointed to; there came to be almost a Graniteville legend in this re- gard. Dividends under Gregg amounted, in various state- ments, anywhere from 7 to 12^4 per cent. The mill was in bad condition when Hickman became president in 1867, being in debt and paying 12 per cent interest. There is no 104 Baltimore Journal of Commerce and Manufacturers' Record, June 3, 1882. For minor variations of these statements, cf. ibid., Sept. 2, 1882. There was believed to be room for indefinite expan- sion of cotton manufacture in South Carolina, " inasmuch as the Carolina mills pay expenses when the New England mills run at a loss, make money when the New England mills only pay expenses and make still larger profits than the New England mill's when these pay well (Blackman, p. 19; cf. ibid., statement of Twitchell, p. 11). Cf. Baltimore Journal of Commerce and Manufacturers' Record, Nov. 4, 1882, quotation from Bradstreet's and Manufac- turers' Record, Baltimore, Feb. 1, 1883, statement of Hammett. The Boston Commercial Bulletin said : " The advantages of cotton manu- facture are with the South decidedly, the profits of the business there having shown an average dividend of 15Y2 per cent against 7% per cent in the North during the year 1882" (ibid., April 5, 1883). Re- viewing the generally unsatisfactory condition of Northern mills from 1890 to 1900, attributed in large part to successful competition of Southern enterprises, it was declared : " Prior to the close of the census year there had been scarcely any interruption of the exceed- ing prosperity of Southern spinners. They did not curtail produc- tion when many Northern manufacturers were in a state bordering upon despair; on the contrary, a large number of their mills were running day and night. They did not seek to dispose of their product by auction, but sold all they could make at prices which gave their stockholders handsome dividends " (U. S. Census of Manufactures, 1900, Cotton Manufactures, p. 20; cf. ibid., pp. 28-29). Cf. News and Courier, Charleston, Sept. 13, 1881, interview with Francis Cogin. 37 1 ~] THE ROLE OF CAPITAL 265 doubt about the improvement that took place — the plant was enlarged and bettered and stock was increased in value. 105 While competition was slight, mills run in any sort of way made money. 106 Stock in South Carolina mills in 1880 was worth, on the average, more than 125. Exclusive of a Clement Attachment mill, where 50 per cent was made, profits ranged from 18 to 25^2 per cent. It was said that in the next year any ordinary factory ought to pay as well. 107 Demonstrated fact encouraged one estimate that spinning mills in the South at large should make 50 per cent. 108 In 1882 "authenticated statistics" were declared to show that investments in Southern mills with good, bad and indifferent management were receiving average divi- dends of 22 per cent. 109 It was not unusual for mills in these years to make 30 per cent to 75 per cent profit. 110 105 Clark, in South in Building of Nation, vol. v, pp. 324-325 ; Blackman, pp. 4-5 ; News and Courier, Charleston, Feb. 23, April 25, 1 881 ; Boston Journal of Commerce, July 29, 1882 ; Augusta Trade Review, Oct., 1884; Tracy I. Hickman, int., Augusta, Dec. 29, 1916. As to profitableness of Georgia and Alabama mills in the seventies, see Clark, ibid., vol. vi, p. 256; Berney, Handbook of Alabama, p. 271 ; News and Courier, Charleston, Sept. 13, Aug. 18, 1881. 106 Benjamin Gossett, int., Anderson. When a new superintendent took hold at a North Carolina mill he found half the looms idle, and yet the plant was highly successful (William Entwistle, int., Rock- ingham). 107 Blackman, leading article and pp. 3, 8, 16, 18. 108 Daily Constitution, Atlanta, March 18, 1880. 109 Baltimore Journal of Commerce and Manufacturers' Record, June 24. Clark says that investments this year amounted to $10,- 000,000 ; this is not surprising when it is remembered that large, well- conducted corporations were paying dividends of 17 to 24 per cent. 110 Sterling Graydon, Charlotte; William Entwistle, Rockingham, interviews. The Augusta Factory made 17 per cent in six months (News and Courier, Charleston, Aug. 18, 1881 ; cf. Daily Constitu- tion, Atlanta, Jan. 6, 1880). Other mills made from 26 to 29 per cent (Baltimore Journal of Commerce and Manufacturers' Record, June 24, 1882). Even mills in bad situations and with poor equip- ment made large sums. A little factory eight miles from the nearest place and hauling over wretched roads paid 25 per cent on the investment; machinery was old, capacity limited and the mill ran only 278 days in the year. Yarns sold at 23 cents per pound cost 2.44 cents to manufacture, and the demand could not be nearly sup- plied — the total output might have been sold to one man (Blackman, pp. 11-12). 266 THE RISE OF COTTON MILLS IN THE SOUTH [372 If Vaucluse was the only mill established before 1880 that paid anything in the first year of operation, 111 its record was matched regularly after that date. Twenty-four per cent was made the first year by a Georgia mill, and a Spar- tanburg company after six months paid a 4 per cent divi- dend and proposed to increase its capital to $1, 000,000. 112 Without following this subject through remaining years, it may be mentioned that 1882 was in some respects not so easy for Southern cotton manufacturers as the two years previous, 113 and that, though experiencing general alarm and a few disasters in 1884, 1893 and 1896, the industry quickly recovered from these backsets. 114 Following the great activity of mill building which began about 1900, Southern earnings were approximating the low averages of New England plants twenty-five years earlier. 115 With some exceptions, 116 mills did only fairly well through the next decade, and many were in bad condition financially when the advent of the Great War lifted them all into pros- perity. The liveliest successes of the eighties were re- peated and surpassed. A right new mill at Gastonia with $150,000 capital after a short period of operation paid a stock dividend of 20 per cent and made $155,000 net profit for the year. Generally, mills at this place that did not make 75 per cent were thought poorly managed, numbers made their entire capitalization in twelve months and. some even higher. 117 Other localities throughout the South found themselves hardly less blessed ; with profits as the 111 Ibid., p. 6. 112 Baltimore Journal of Commerce and Manufacturers' Record, June 3, Aug. 26, 1882. For notice of typical factories paying un- evenly but averaging about 20 per cent, cf . Blackman, pp. 10, 13, 15 ; Deutsche Zeitung, Charleston, Feb. 28, 1881 ; News and Courier, Charleston, Sept. 7, 1881. For some less conclusive references to profits see ibid., Jan. 25, Feb. 26, April 4, 1881. 113 Manufacturers' Record, Baltimore, Feb. 1, 1883; A. B. Murray, int., Charleston. 114 Clark, in South in Building of Nation, vol. vi, pp. 281, 284-286. 115 Cf. Goldsmith, p. 6. 116 Murphy, p. 16; Law, pp. 23-24. • 117 S. N. Boyce and J. Lee Robinson, G. W. Ragan, C. B. Arm- strong, Gastonia, interviews; J. Lee Robinson, letter, Gastonia, Nov. 28, 1916. 373U THE ROLE OF CAPITAL 267 incentive, factories sprang up as suddenly and widely as in the Cotton Mill Campaign. 118 There is a relation between percentage of profit and size of plant. The magnitude of mills, of a part, of course, with degree of concentration of investment, is an interest- ing subject, indicating differences in development of the industry in various States. It is convenient to compare South Carolina and North Carolina in this regard. Fac- tories of the former State tended to be fewer in number but greater in capacity than those of the latter, and wove as well as spun. Furthermore, the impulse toward cotton manufacturing was felt later in North than in South Caro- lina. Several reasons may be assigned for these facts. The considerable capital of Charleston, as it had earlier been largely responsible for Graniteville and Langley, later played the leading part in the founding of such mills as Piedmont and Pelzer. These big weaving mills set a stand- ard; also, as has been noticed, Charleston money was a re- source to South Carolina local communities pretty generally. North Carolina bad no city the size of Charleston; Wil- mington was not so good a port and did not possess so much capital available for investment. Little neighbor- hoods were shut up to their own initiative and means. Moreover, there had always been less social unity in North Carolina; with much Scotch blood, the people were indi- vidualists. Most of the time small merchants had to be mill projectors, and this was 1 agreeable, too, because per- sonal control over modest units was preferred to a pool- ing of resources in the hands of an important capitalist. These things explain, also, why the development commenced later in North Carolina. More people had to be converted than in a State where a few could set a powerful example. Even where North Carolina had weaving mills, these were generally smaller than those of South Carolina. It is to be observed that whereas the principal mill mergers of South 118 W. J. Thackston, int., Greenville ; Literary Digest, N. Y., Dec. 9, 1916 ; cf . files of all trade papers, especially Manufacturers' Record, Baltimore. 268 THE RISE OF COTTON MILLS IN THE SOUTH ^374 Carolina showed concentration of management, in the out- standing case in North Carolina constituent mills remained semi-autonomous. When a tradition was established, it tended to maintain itself. 119 With local capital in greater supply in South Carolina, commission and machinery firms were more interested and engineers were more regularly engaged, so large plants were encouraged. In undertaking the development of power and manufacturing at Columbia, it was pointed out by engineers that a 16,000-spindle mill would cost $27 per spindle and yield 17 per cent profit; plant cost would be proportionately less if equipment was 20,000 spindles and the complement of looms, and earnings should be 21 per cent; 26,000 spindles ought to bring earnings of 25 per cent. 120 Later, a conscious, concerted movement toward mills, lifting spindleage above the 30,000 point, in which North Carolina patterned after South Carolina, was typified in the erection of the great Olympia plant at Columbia. The animating spirit was evidenced by a speaker before cotton manufacturers in 1903 : " I believe thoroughly in organiza- tions of such magnitude that will justify the employment of the very best skill to be obtained in systematic manage- ment." There was much to be saved in purchase of sup- plies and materials. "A weakly fitted up mill under poor management is worthless; the same mill under good man- agement is even then sadly handicapped." But merger with 119 David Clark, Hudson Millar, Charlotte ; E. A. Smyth, Green- ville ; Charles E. Johnson, Raleigh ; W. K. Boyd, Durham, N. C, Sept. 18, 1916; J. A. Chapman, Spartanburg; J. H. M. Beatty, Columbia, interviews. Georgia was much like South Carolina. Speaking of the limitations of North Carolina, it was said : " Our people hadn't the money; they all had to scratch to get anything to put into cotton mills, and then it wasn't much" (W. R. Odell, int., Concord). 120 News and Courier, Charleston, April 13, 1881 ; cf. Law, p. 19. As early as 1883 South Carolina had several mills which would be ranked as large even today — four companies with capitalization of half a million or more, with others of size (cf. Manufacturers' Record, Baltimore, Jan. 18, 1883). There were, of course, mills as small as any in North Carolina, but these dated from previous years (cf. as to Valley Falls and Reedy River, the former of only $5000 capital, Blackman, pp. 11, 13). 37 S2 THE ROLE OF CAPITAL 269 other plants would bring increased financial facilities. " The tendency to concentrate and build mills with a larger num- ber of spindles than formerly is a move in the right direc- tion." 121 But it was soon learned that very large separate mills and close mergers of several plants were of uncertain success, disadvantages more than offsetting advantages. 122 Small, isolated plants bought local cotton at a saving and paid no higher commissions on product; some could burn wood; operatives were few and individually known; a su- perintendent could be developed from the working force and did well enough on a limited number of standard yarns; living in a small place was cheaper. 123 Time has 121 " The record of the past three years shows a large number of plants erected in the South of from 25,000 spindles up to that grand specimen of push and enterprise — the Olympia Mills — which has 104,000 spindles in one mill and all in one room" (see Southern Cot- ton Spinners' Assn., proceed. 7th Annual Convention, address of E. W. Thomas, p. 149 ff.). Cf. Clark, in South in Building of Nation, vol. vi, pp. 287-288. Greenville had the example of the success of such large mills as Pelzer as contrasted with the smaller Huguenot and Camperdown factories ; there was the strong impression that individual mills of limited size were not easily financed (Clement F. Haynsworth, int., Greenville). "The Loray Mill in Gastonia was built about the same time as Olympia; small mills had succeeded, and they thought big ones would succeed even better" (S. N. Boyce and J. Lee Robinson, int., Gastonia). 122 " Attention is being paid to the danger of having too large units, the prevailing opinion in the South being that no special econo- mies from increased size are obtainable after say 50,000 to 60,000 are reached. A notable disaster to stockholders and near-disaster to creditors in recent years has taught the lesson that an unwieldy combination of plants scattered geographically has no advantage, through concentration of purchasing or selling, that can possibly offset the diminution of the personal equation in relations with em- ployees or scrutiny of details, usually given by the executive in charge of smaller units" (Law, p. 19; cf. Tompkins, Cotton Mill, Commercial Features, p. 55). The promoter of the chief amalgama- tion in South Carolina believed he would save in overhead expense; the main benefit was in financing, for much money was offered at 3 per cent when the merger went together, whereas the individual mills had never borrowed at less than 5 per cent. Any other savings were more than counterbalanced by expensively lax supervision. Failure resulted (J. H. M. Beatty, int., Columbia). In the principal North Carolina chain, while ownership is virtually identical, each mill has its own directors and must stand on its own bottom finan- cially. Some economies of combination are deliberately sacrificed to maintain efficiency of superintendence (James W. Cannon, int., Concord). 123 cf # Thompson, p. 90 ff. 270 THE RISE OF COTTON MILLS IN THE SOUTH £376 taken away some of these benefits, but the best present opinion is that well situated units of about 10,000 spindles are most economical. 12 * There was little buying and selling of mill stocks in the first part of the period, and for several reasons. Factories were so often looked upon as family affairs, part and parcel of the communities which established them. Local sub- scribers, small and large, put in their money as an invest- ment, and most of those who could purchase shares did so at first. Mills were successful, moreover, and brought divi- dends. To outsiders the industry was an experiment ; pri- vate investors were not attracted. There were few agencies in the South for handling the securities. Consequently, notices of value of stocks usually meant really book value. 125 It has been seen that in 1880 the shares of South Caro- lina mills were reported as being worth on the average $125. Three years later all were above par except five, which were at par; Langley was highest, selling at $I73. 126 The stock of the Wesson mill in Mississippi, paying 26 per cent divi- dends, stood at more than 300. 127 Shares in the Merrimack mills, in Alabama, par value $1000, sold for $i620. 128 124 Building since 1914 has shown this. " I had rather run four mills of 10,000 spindles each than one of 40,000 spindles" (C. B. Armstrong, int., Gastonia). This would have to be modified some in the case of cloth mills. 125 " The stock of the company sold for $63 a share in 1867, and now is quoted at $123. Even this figure is not a fair estimate of what it is worth because nobody wants to sell. I could go in the market tomorrow and run it up to $130, or even $150, just by offer- ing that for it. This is not what we want, however" (Hickman, of Graniteville, quoted in Blackman (1880), p. 4. A Rockingham mill has been owned by the same stockholders for the forty years since its establishment (Charlotte News, Textile Ed., 1917, as to Roberdell Mill No. 1). Stock in the first mill at Salisbury could not be bought; 60 per cent of it was owned by women who received it by inheritance (O. D. Davis, int., Salisbury). Where there was a mar- ket at the opening of the period it was local, mills taking charge of their own sales (Tracy I. Hickman, Augusta; William Entwistle, Rockingham, interviews). 126 Manufacturers' Record, Baltimore, Jan. 18, 1883. The stock of Graniteville and Vaucluse had climbed to 170. For similar facts as to Augusta factories, cf. Baltimore Journal of Commerce and Manufacturers' Record, Sept. 2, 1882; News and Observer, Raleigh, Nov. 16, 1880. 127 News and Courier, Charleston, Jan. 14, 1882. 128 Observer, Raleigh, Aug. 26, 1880. 377] THE ROLE OF CAPITAL 27 1 Besides the usual causes, Southern mill stocks have varied in value because the business was subject to sharp fluctuations, companies were irregular in providing surplus to insure constant dividends and in offsetting deprecia- tion, 129 skill in management was so largely hit-or-miss, com- mission firms sometimes interfered hurtfully and, as will be remarked, machinery makers dumped their shares in large blocks. Pacolet once had to alter its product and so its ma- chinery; preferred stock was issued and common fell from 300 to below par. 130 Within two years after a commission firm had gained control of a South Carolina mill following a fight with local stockholders, shares that had been at 175 dropped to par. 131 An active market for the stocks developed in Charleston about 1890 and in the up-country somewhat later. A good many brokers made a specialty of these securities. The business was assisted by machinery builders disposing of their holdings at concessions. One firm handled in one year about $2,000,000 worth of securities thus thrown on the market. 132 Charlestonians had been heavy subscribers to new ventures in the State, but about 1900 stopped be- cause they could buy at less than par. 133 The financial history of Southern mills has exhibited physical differentials becoming less and less important, and skill in management becoming more and more important. Atkinson's admonition that success in cotton manufacture meant a small margin of profit on a large capital was, after 129 Qf. Tompkins, Cotton Mill, Commercial Features, p. 85. 130 A. N. Wood, int., Gaffney. Stock in Graniteville and mills at Augusta, which earlier led the field, went far below par (Tracy I. Hickman, int., Augusta). 131 VV. W. Ball, int., Columbia, Jan. 3, 1917. 132 w. J. Thackston, letter, Greenville. When machinery manu- facturers were taking part payment in stock, equipment was in great demand and high in price. Makers could therefore sell their shares quickly at 50 cents on the dollar and still make money (Washington Clark, int., Columbia). Commission men, retaining their shares, sometimes made money ; a firm that took stock when it received the agency of a mill and offered to sell at 50 later succeeded in selling at 300 (Walter Montgomery, int., Spartanburg, S. C, Sept. 5, 1916). 133 W. W. Ball, int., Columbia, Jan. 3, 1917. 272 THE RISE OF COTTON MILLS IN THE SOUTH [378 all, of only delayed applicability in the new industry. 134 At the opening of the period, as has been seen, "they didn't run mills, but just put them up and they made money. Long hours of labor and low wages made the difference between that time and this. But old superiorities have passed. Mills that stayed in the old rut went to the wall. It is necessary to operate mills in the South today." 135 Management of investments in land and negroes was not the best equipment for industrial control. As the South had grown a staple commodity, raw cotton, and grew too much of it, so it manufactured staple cotton goods, follow- ing the impulse mechanically. 136 Inexperienced men found- ing the industry in 1880 made money; the same type enter- ing the business twenty years later, as at Bessemer City, found they could not exist. 137 By this time, in the same mill in which average management would yield 10 per cent profit, superior management might bring 25 per cent and inferior operation a loss of 5 per cent. 138 The margin be- tween the price of middling cotton and of print cloth made from it between 1881 and 1910 worked down, though not without great irregularity, from 108.52 to S9- 2 4- 13& And 134 Cf. News and Courier, Charleston, Dec. 5, 1881, and the writer's " Factors in Future of Cotton Manufacture," in Manufacturers' Record, Baltimore, May 10, 1917. 135 \y_ j. Britton, int., Spartanburg. Social position and good intent too often had to serve in place of industrial ability, though after 1880 there were few instances approaching an episode during the Civil War, when, at reorganization of an Augusta mill, a gover- nor was given $100,000 in stock for his influence as a director (Charles Estes, int., Augusta). Often general capability, disregarding accustomed financial methods of corporate undertakings, succeeded through sheer force, but in other cases a slump in the business would take enterprises out of the hands of the original management (Chronicle, Augusta, Jan. 28, 1886; Henry E. Litchford, int., Rich- mond, Aug. 29, 1916). 136 Landon A. Thomas, int., Augusta. Initiative in the trend toward closely supervised plants making specialty products, already appearing, is fundamentally a problem of the common school, awak- ening public intelligence. Cf. Georgia Industrial Assn., proceed. 4th Annual Convention, pp. 46-47. 137 S. N. Boyce and J. Lee Robinson, G. W. Ragan, Gastonia, interviews. 138 Tompkins, Cotton Mill, Commercial Features, p. 173. 139 Copeland, appendix, p. 394. 379] THE ROLE OF CAPITAL 273 training gained in manufacturing has made apparent a lack of development of commercial attributes which are as nec- essary a part of the mill man's equipment. 140 There has been a gradual evolution from first projectors, who were really transplanted slaveholders, through a somewhat later group composed of business and professional men, to the newer type of manufacturers who conceive it their work to make money on fabricated product and not in speculation on raw cotton or any other gamble, who are not afraid of competition with New England and the world, who relish technical information and know they had better manage a few plants well than many poorly. 141 A qualified observer has said that in the Southern indus- try the total losses on an investment of $100,000,000 have not amounted to 20 per cent, and that this is remarkable when it is remembered how few managers began with knowledge of the business. 142 Gregg assigned five main causes of failure of mills in South Carolina in his day. These were injudicious selec- tion of machinery and character of goods to be made, lack of steady and cheap motive power, poor location, lack of moral training of operatives, and want of sufficient capi- tal. 143 The first and last of these reasons are the only ones 140 See Georgia Industrial Assn., proceed. 4th Annual Convention, address of J. J. Spalding, pp. 46-47. 141 Cf. the writer's " Factors in Future of Cotton Manufacture," in Manufacturers' Record, Baltimore, May 10, 1917; Tompkins, Cotton Mill, Commercial Features, pp. 30 ff., 63 ; Plan to Raise Capital, p. 18; J. H. M. Beatty, Columbia, Jan. 3, 1917; Landon A. Thomas, Augusta; Joseph H. Separk, Gastonia, interviews. It is true that in this process young men technically trained have not yet made themselves available for large leadership, so that others without their advantages are still called upon (W. W. Ball, int., Columbia, Jan. 3, 1917). 142 W. J. Thackston, int., Greenville; cf. Law, p. 18. There is some truth in a statement that as a rule the local investor has not made much on dividends, but has received, with everybody, a large indirect, social benefit from the establishment of the industry (M. L. Bonham, int., Anderson). Small local shareholders, if dividends did not begin promptly, sometimes sold, very often to the mill promoter (J. W. Norwood, int., Greenville). 1 43 Quoted in Kohn, Cotton Mills of S. C, p. 18. 18 274 TH E RISE OF COTTON MILLS IN THE SOUTH {j80 that may be said to have held in the later period. 144 Mis- fortunes following untrained management were not men- tioned by Gregg, probably because he could not foresee competitive conditions that were to come. It has 'been remarked that a good many mills were sold just prior to the opening of the Cotton Mill Campaign. Some of these were old factories that had been run down, or their owners had died ; either they failed, or were bought up when the industry was receiving renovation and there was a demand for plants that could be improved. 145 The mills sold in the eighties were decidedly exceptions. 148 However, 1884 saw losses and partial shut-downs while debts accumulated. Graniteville went backward for the first time in seventeen years. Recovery in special cases was the slower because mills were just launching out. 147 Surrounded by cotton, the price of the raw material play- ing so large a part in coarse goods manufacture, and hav- ing some capital at their disposal, the temptation for mill executives to speculate in the staple has been an evil. Two men worked together in promoting manufactures at Gas- tonia; one was content to make or lose as a spinner, and succeeded, while the other after a time counted too heavily •on his skill in manipulation of cotton deals and met with -disaster. About 1900 the stock of an excellent South Caro- lina mill went to 150, and the promoter erected a second large plant. With good credit, mill president and town "were ruined in two years ; he gambled in cotton and the 144 Old machinery was always a bad bargain, but when it was bought the mills were making money and soon could scrap this equipment and profit by the experience ; standard goods were manu- factured and, as will be noticed presently, losses on these were be- cause of sudden change in the market rather than through mistaken choice of product. 145 Cf. Blackman, p. 12. 146 Cf . Savannah Morning News, July 7, 1882. The Charleston factory changed hands at a loss totalling $499,000 (Bird Memoranda). 147 Cf. Chronicle, Augusta, May 28, July 29, 1884; Jan. 29, April 23, 1885 ; Jan. 28, March 10, 1886, and other clippings arranged in the Raworth Scrapbook; there are printed reports of the president of the Sibley mill dated April 28, 1886, and Oct. 23, 1888. $Sl"] THE ROLE OF CAPITAL 275 market went against him. 148 A sympathetic 'critic of the Southern industry has said that " The principal occasion of financial disaster . . . has been that of speculation. It is true that in some instances this has been merely the final plunge of desperate unsuccessful management. In other cases, however, both directors and stockholders have known that earnings greater than possible from legitimate manu- facturing were being shown. They winked at the excessive profits and deserved little sympathy when they sustained losses." 149 There has been little fraud on the part of mill men. In the beginning there could scarcely have been any, so inti- mately were communities acquainted with the enterprises. 15 * The scandal in the great Parker merger recently has been the conspicuous exception ; the experience did much to turn favor away from closely centralized financial control. This failure was a moral blow not only to the industry, but to the section. 151 Another cause of failure has been payment of too high salaries, with extension of plant to niake these seem plaus- ible. Also, superintendents have been accused of receiving commissions on machinery and supplies bought by them. 152 148 Cf . Columbia Record, Textile Ed., 1916, respecting Union and Buffalo mills. 149 Law, p. 21 ff. " Profits thus obtained are absolutely demoraliz- ing to efficiency in management or the working out of small econo- mies — the legitimate source of success — and are hurtful to the general industry, in that they create fictitious costs, apparently justi- fying sales of product at really destructive prices. . . . My belief is that the cotton manufacturer who now indulges in such speculation is the exception." There have been examples of what might be called speculation in finished product, too. A gingham mill at Rock Hill had been operating successfully; the market dropped, but pros- pects were thought to be good and cloth was stored in warehouses until it represented a value greater than the capitalization of the company. The style in ginghams changed, and the plant had to be sold (cf. Columbia Record, Textile Ed., 1916). 150 There seems to have been allegation of fraud in the case of the small Fork Shoals factory in 1881. This was an old and isolated mill (cf. News and Courier, Charleston, April 23). 151 W. J. Thackston, int., Greenville. 152 Charles Estes, int., Augusta. In the case of one Augusta mill, it is alleged, the president did not inspect matters narrowly; the 2^6 THE RISE OF COTTON MILLS IN THE SOUTH [382 Extensions involving debt, especially from the nineties forward, were a source of misfortune. The Gaffney mill after three profitable first years built a warranted addition, but then followed a big new plant and a finishing mill that saddled the company with obligations under which it could not succeed, business being depressed. 153 Coleman's mill at Concord, not unnaturally for a first enterprise by negroes, was badly managed and became in debt to local capitalists, who foreclosed. 154 Everyone had been willing to lend to the reliable Graniteville mill with- out anxiety as to payment of principal, until suddenly cred- itors became solicitous for their money, precipitating reor- ganization of the company in 191 5. Though there is complaint that too many mills were built in a short period, so that profits fell away, 155 it may be concluded that where enterprises have not succeeded their difficulties have been due to untrained management and lack of capital rather than to untoward conditions or lim- ited opportunities in the industry. 156 superintendent would send certified bills to him and he would make out New York checks for the amounts, the superintendent getting his benefit from such payments. 153 L. Baker, int., Gaffney. The Laurens mill borrowed $150,000 to give the plant 30,000 spindles and other enlargements ensued and contributed to embarrassments of the company later. Whaley in' Columbia built the little Richland mill and then Granby, and both did well. Then he proposed to build the greatest mill on earth under one roof, and exhausted the credit of his previous factories (W. W. Ball, int., Columbia, Jan. 3, 1917). "Many mills were built with a debt of $10 per spindle [the average cost being about $20], believing they could pay up in a few years at the high earnings of $4 or $5 per spindle. Many of these were caught with big debts and declining earnings" (Summerfield Baldwin, Jr., int., Baltimore). 154 Charles McDonald, int., Charlotte. 155 James D. Hammett, Anderson ; Mrs. M. P. Gridley, Greenville, interviews. 156 Julius Koester, H. R. Buist, Charleston ; Thomas Purse, Sa- vannah ; August Kohn, Columbia, interviews. INDEX Agriculture, exclusive devotion to, vii; character of, in ante- bellum South, 30 (note) ; use of primitive methods in, 27; indications of revival of, in N. C, 73, 81 ; improvement in, 86-87; at low ebb in seventies, 144-146; closely joined with industry, 173 ff., depressed condition of, 173-176. Atkinson, Edward, his views on proximity to raw material, 64-65 ; on preparation of cot- ton, 65 ; significance of Atlanta speech, 75 and note; his char- acter, purposes and influence, 117 ff. ; attitude of South to- ward, 237, 238. Baker, L., quoted, 101 (note) ; his leadership at Gaffney, S. C, 128-129. Banks, unimportant participa- tion of, 249, 260-261. Bonds, 258-260. Borrowing. See Capital and Commission Houses. Brown, W. G., quoted, vii. Capital, availability of, to ante- bellum South, 23, 24; in Southern industry 1850-60, 44; home, to be drawn upon, 83 ; investments as result of Atlanta Exposition, 124 and note; English, negligible, 149- 150; would come with immi- Igration, 205 ; investment off local, 232-237; attitude toward outside, 237-241 ; investment by machinery makers and commission firms, 241 ff. ; lack of working. 248 ff . ; of Charles- ton, 267; borrowing, 261 and note, 267 (note). Charleston, S. C, ordinance of, against use of steam engine, 33 and note; neglect of State industry by, 37-38; changed spirit' of, 81-82 and note, 111; concern of, for public welfare, 127; scarcity of operatives in, 193-196; negro operatives in, 216-218; capital of, 267. Charleston Manufacturing Com- pany, organization of, 71 ; reasons for inception of, 97, 127, 132, 133-134. 157; labor troubles of, 193-196; negro operatives in old plant of, 216-218. Child labor, necessary and natural at first, 95. Civil War, not a fortuitous event, 43 ; as block to declared industrial beginnings, 46 (note) ; lack of manufac- tures to assist in, 53 ; opened door to Southern upbuilding, 53-55. 85. Clark, V. S., his views on period 1840-60, 22-26; on whole ante- bellum period, 41-43- Clay, C. M., on slavery as cause of Southern inaction, 51 (note). Clement Attachment, 74, 154. 263. Climate, modified by humidi- fiers, 67 (note). Coleman, Warren, 215-216. Columbia, S. C, peculiar sufferer in Civil War, 83 (note), 128 (note); canal project at, 128; advantage of location at, 224; self-help in, 234. Commission Houses, participa- tion of, 241 ff. ; as lenders of working capital, 248 ff; "Plaid Trust," 256 (note). Copeland, M. T., on location of mills, 184-185; on negro op- eratives, 220 (note). 277 278 INDEX [384 Cotton, course of price of, be- fore Civil War, 28 (note) ; increase in acreage in fifties, 49-So; increased production of, after Civil War, 75; low price of, stimulated manufac- tures, 144-146; an early view of benefits of, 161-162 (note). See Agriculture. Cotton Gin, effect of, upon price of staple, 10; cotton mill on on site of, 14-15 (note) ; re- sponsible for slavery, 28. Cotton Mills, ante-bellum ("Old Mills") antecedent in time but not necessarily in effect, 15- 16; character of, 16-22; sta- tus of, in 1840 and 1850, 21 ; V. S. Clark on those of 1840- 60, 22-26; evidences of rise of, in South, 59 ff. ; anticipated, not accidental, 78-79, 84-85; reasons for rise of, 96 ff . ; rise of, was through Southern ef- fort, 102. "Cotton Mill Campaign," 1880 as date of commencement' of, 59 ff.; its character, 151 ff. Dawson, F. W., his character and service, 113-114. DeBow, J. D. B., a representa- tive of minority opinion, 56. Dividends, 261 ff. Edmonds, R. H., on ante-bellum industrialism, 43-47. Enterprisers, from old aristo- cracy, 47-48, 55-56; doers, not talkers, 78; social exponents, 101 ff . ; ex-Confederates as, 102-105 ; cotton factors as, 105-106; merchants as, 106- 108; Northern, welcomed, 111- 112; as stimulated by town pride, 128-131 ; some who learned by example, 142; in- clined toward individualism in N. C, 267-268; evolution in type of, 271-273. Estes, Charles, 108. Failures, due to poor execu- tives, 108 (note) ; 273 ff. Finishing, 255. Gaffney, S. C. See Baker, L. Garfield, James A., South moved at assassination of, 91-92. Gastonia, N. C, G. A. Gray's influence at, 109, no. Grady, H. W., quoted, 77 and note, 80; character of his thought, 80 (note) ; on obli- gations to South, 94; advocate of cotton manufactures, 114. Graniteville Factory, 20 (note), 23, 142, 169-170, 264-265, 274. See Gregg, William. Gray, G. A, 109. Gregg, William, isolated advo- cate of manufactures, 26; on results of cotton culture, 28; proponent of self-sufficiency for S. C, 33; against current political leadership, 36-37; on advantages of corporate in- dustrial enterprise, 38-39 (note) ; not a social exponent, 43; on enterprisers, 107-108; on public benefits of mills, 125-126; on poor whites, 168- 169; on negro operatives, 210- 211; on causes of failures, 273-274. See Graniteville Fac- tory. Hammett, H. P., typical indus- trial leader, '58-59 (note) ; one of pioneers, 71, 109; his phil- anthropy, 133; influence of his Piedmont Factory, 143 ; on Southern operatives, 171 ; on negro operatives, 217; labor in mill of, 226. Hammond, J. H., his elevation of cotton culture, 48-49. Hayes-Tilden Election, 59, 62; as delay to "Real Reconstruc- tion," 88. Helper, H. R., quoted, 15 (note), 26, 34, 52 (note). See slavery. Hemphill, J. C, vii. Immigration, kept out of South by slavery, 23-25, 31 ; progres- siveness of districts having foreign traditions, 32; sought by South, 200 ff. Industrialism, opposition of ante-bellum leaders to, 35~37; 385] INDEX 279 importance of corporate enter- prise to, 38-39; unhealthy ad- vocacy of, 42, 83-84; no real movement toward, before Civil War, 41-47 ; precluded by sectionalism, 52; lack of, felt during Civil War, 53- Industrial Revolution in Eng- land, contrast with that of South, 78-79, 84-85, 86. Ingle, Edward, criticism of ante- bellum South, 28 (note) ; on poor whites, 165. International Cotton Exposi- tion, Edward Atkinson's pur- pose in, 65; significance of, 71- 73, 120 ff. Kohn, August, on share-tenants, 175 (note) ; on scarcity! of labor, 199 (note) ; on negro operatives, 211, 220 (note) ; on family incomes, 230 (note) ; on proportion of home-owned capital, 233 (note) ; on dividends, 262. Labor, now differentiated from management, 160-161 ; char- acter of poor whites, 161 ff. ; efficiency of, 171-172; plenti- fulness of, at first, 176 ff. ; proportions of classes of op- eratives, 180-181 ; local sup- plies of, 181 ff. ; floating, 186- 187; from mountains, 189- 191; scarcity of, 191 ff. ; oppo- sition to, 200, 205, 206 (note) ; soliciting of, 208-209; slaves in cotton mills, 209-213 ; negroes in after-war factories, 213-221. See "Poor Whites." Law, J. A., on extensions of plant, 263 (note) ; on manage- ment, 269 (note) ; on specu- lation, 275 and note. Location, only one mill within corporate limits in 1880, 62; alleged superior advantages of New England, 64-65 ; in re- sponse to social motive, 129- 130, 131 (note) ; with respect to labor, 184-187, 209; infe- rior, did not prevent profit, 265 (note). Losses, 273 ff. Machinery, new in South, 67; cotton manufacturing at At- lanta Exposition, 122, 124-125; need for reequipment with, 151; Southern demand for Northern, 244-246; use of second-hand, 245-246, 250 (note). Machinery Manufacturers, not' speculators in South, 150-151 ; participation of, 241 ff. Management, 271 ff. Manufactures, dependence of ante-bellum South upon North for, 32-35, 97 (note). Mechanical Improvements, ab- sence of, through agency of slavery, 27 and note, 52-53. Mitchell, S. C, quoted, vii-viii; on Southern economic states- manship, 92 (note) ; on South- ern democracy, 126 (note). Murphy, E. G., his views on Southern industrial history, 46-48; on Southern democ- racy, 126 (note) ; on altruism of South, 132 (note). Negro, considered for mill op- erative, 25 ; opposition to em- ployment of, 27 (note), 200, 205, 206 (note) ; used in " old mills," 170, 209-213; in after- war mills, 213-221. New England, ability of South to compete with, 46; relative advantage with South, 64; losing in percentage increase in cotton manufacturing, 66; growth of special localities in, stressed, 67, 68; it's recognition of Southern development, 69; its interests guarded by Ed- ward Atkinson, 1 18-120; con- trast with South respecting labor, 180-181, 224, 225 ; re- specting operation, 238-239 ; respecting profits, 262, 264 and note. " New South," 45-46 ; it's her- itage from ante-bellum South, 47-48; constructive, not de- structive, 80-81 ; stimulation of, in Atlanta Exposition, 123. 28o INDEX [386 News and Courier, as advocate of cotton manufacturing, 71, 82 (note) ; 112-114. See Daw- son, F. W. Olmstead, F. L., on numbing ef- fect of slavery, 29 and note; on poor whites, 164-165 (notes). Operatives, from North and Europe contemplated, 25, 193, 199; social prejudice against, 196-198; sale of stock to, 261. Orr, J. L., his acquiescence in result of Civil War, 78; as type of ex-Confederate enter- priser, 103. Page, W. H., quoted, 126-127, 132 (note). Piedmont Factory. See Ham- mer:, H. P. Plant, size of, 267-270; exten- sion of, 275-276. Plunkett, Horace, on Irish agri- culture, 27 (note) ; on Irish politics, 30 (note), 51-52 (notes). Politics, profitless character of, in ante-bellum South, 29-31 ; growing manifestation of, in commercial conventions, 50>- 51 ; blow to), through Civil War, 53-54; losing power, 78; eschewed, 89-90, and 91 (note) ; 91-94; took all energies during Reconstruction, 98--99. " Poor Whites," effects upon of cotton culture. 10; ignored as possible operatives, 25, 192; concern for as motive in mill- building, 132 ff. ; character of, 161 ff. ; their eagerness to enter mills, 176-179. See labor. Presidential Election of 1876. See Hayes-Tilden Election. Presidential Election of 1880, issue of, as contributing to industrialism in South, 88 ff. Product, contributed to domestic industry, 18-10: advantage in coarse, accorded to South, 7^, 138, 139 and note; demand for, in 1880, 75 ; negroes useful only on coarse, 217, 221; selling of, 251 ff. See Com- mission Houses. Profits, growing certainty of, 58; realize total at home, 84; as motive to mill-building, 115-117, 147-148; easily made at first, 130 (note), 263-266: better in South than in North, 152 (note) ; extensions of plant from, 257; of Southern mills, 261 ff. Proximity to Raw Material, Edward Atkinson's views on, 64; as basis of Southern pro- gress, 66; importance of, as seen outside of South, 73; as insuring profits, 115-117; as responsible for rise of mills, 137-141, 157 (note). Railroads, William Gregg's op- position to subsidy to, 37; an- ticipated advantage from, 43 (note) ; renewed activity in building of, 74, 87; and labor supply, 185-187. Raworth Scrapbook, viii. "Real Reconstruction," 77 ff. ; comprehensiveness of, 86-88, 124 (note) ; spiritual and ma- terial promptings to, 94-95 ; newspapers as leaders in. 112 ff. See "Cotton Mill Cam- paign." Reconstruction, change of heart of South during, 56-58; lasted late in S. C, 62; progress pre- cluded during. 98-99. Revolutionarv Period, excellent start of Southern manufac- tures in, 11-12; domestic char- acter of industry in, 12-14. Salisbury, N. C. See Salisbury Cotton Mills. Salisbury Cotton Mills, signifi- cance of inception of, 134-136. Slavery, partly responsible for Appomattox, vii ; removal of. pqcisfpri industry, 10. 15 (note) ; hindrance of, to 387] INDEX 28l Southern prosperity, 15 (note), 23, 25, 26-29; as as- sistance to industrial leaders, 48; C. M. Clay's condemnation of, 51 (note) ; abolition of, freed the South, 54. Slaves, not wealth, 44-45. Stanwood, Edward, his descrip- tion of South's progress in cotton manufacturing, 65-68. States ' Rights, partly respon- sible for Appomattox, vii. Stock, sale of, by machinery makers, 247 and note; addi- tional issues of, 258; pre- ferred, 258-260; value of, 265; market for, 270-271. Tariff, as function of cotton, 40-41. Thompson, Holland, on negro operatives, 219, 220 (note) ; on proportion of Southern capital, 233 (note) ; on work- ing capital, 249 (note). Tompkins, D. A., his interpre- tation of Southern economic history, 10; on results of slavery, 26-27, 28 and note; on South's need for tariff, 41 ; his significance as industrial- ist, 109-110; plainness of his advocacy of manufactures, 117; on negro operatives, 220 (note) ; on working capital, 248; on profits, 262-263. Transportation, consequences of lack of, in ante-bellum South, 39. See Railroads. Wages, half those of North in forties, 169; as reflection of economic conditions, 221-222; in ante-bellum mills, 222-223; in plants of post-1880 period, 224 ff. ; variations in, 226-227, 228 (note) ; significance of in- creases in, 229-231. Watterson, Henry, his charac- terization of Southern eco- nomic history, 26. Woodward, Baldwin & Co., 243- 244 and note. " ttJXw %J'X^^~'lksu jtssJX&n hilt .