College and Research Libraries By T H O M A S F R A N K L I N C U R R I E R Preliminary Cataloging Thomas Franklin Currier is associate li- brarian, Harvard College Library. DU R I N G T H E past few years the term "preliminary cataloging" has, at the Harvard library, been on our tongues with ever-increasing frequency. It denotes a process introduced to free the expert cata- loger from certain mechanical and time- consuming work which simultaneously affords a preliminary survey of all books and pamphlets received, whether by gift or purchase, with a view to segregating duplicates. Briefly described the process is as fol- lows. Purchased books from the ordering department and all gifts are delivered to an assistant attached to the department of cataloging and classification, who checks them with the union catalog. T h i s catalog combines, in one alphabet, the Library of Congress depository cards with the cards for books in the Harvard College Library and the university's numerous depart- mental and special libraries; it serves also as the library's official catalog. T h e check- ing just mentioned ordinarily determines by a single"process: ( a ) if the piece in hand is a duplicate within the library sys- tem; (b) for non-duplicates, if there is a Library of Congress card that may be utilized for cataloging purposes; (c) if the form of name of the author (corporate or personal) has been determined for cata- loging purposes by our own previous in- vestigations or by the Library of Congress catalogers. T h e information thus ob-- tained is noted on slips which are placed in the books. T h e duplicates are segre- gated for further consideration and the rest passed on to a rather highly organized typists' department. Here, girls trained to this work, type, from the books before them, preliminary catalog cards, contain- ing the necessary data, transcribed mechan- ically without research. T h e books, with the cards inserted, are now distributed to the several supervisors1 in charge of the subjects concerned, who can thus attack the labor of placing them in the library's permanent collections with certain neces- sary or desirable information spread before them in convenient shape. Because of this preliminary process, the cataloger is re- lieved of the mechanical work of typing the initial or key cards, or (as an alterna- tive) of making complicated notes for the guidance of a typist. T h e advantage of this method, as against making notes in ad- vance for a typist to follow, rests in the fact that the cataloger does not have to project her imagination into the future and forecast what unexpected things a typist may do, but has before her an actual card, revealing just how the final card will ap- pear as respects length of title, arrange- ment of matter, and the like. T o the point when the books come into the hands of the subject supervisors, men- tioned above, the work has been done without research, the object having been to 1 Experiments that have been made in centralizing in the hands of "subject supervisors" the processes of acquisition and absorption of library material will be described in -anotheP, paper to be printed in Cojlege and Research Libraries. JUNE, 1940 23 7 push everything along as rapidly as ac- curacy w i l l permit, avoiding the delays incident to looking up lacking data and settling debatable points. I t is believed, after several years' experience, that this combination of processes, covering the no man's land between accessioning and cata- loging, is a notable step in reducing the cost of acquiring and absorbing library material and paves the w a y for placing greater responsibility on the cataloger, as w i l l be described. How Experiment Started T h e first conscious experiment in this direction w a s made in the year 1 9 2 5 w h e n the library acquired an unusually large col- lection of the w r i t i n g s of G e r m a n philoso- phers. A cataloger of moderate experi- ence, but a high degree of intelligence, had been assigned to the task of cataloging these books. Short-handed as w e w e r e at the time, and w i t h n e w accessions pouring in, some study of short cut methods of f o r w a r d i n g this important collection to the shelves instead of holding it until it could be taken up volume by vol- ume in the normal retail fashion, seemed desirable. B y our first experiment one of our best typists, w h o had some familiarity w i t h G e r m a n , copied the title pages on sheets of paper and these copies w e r e edited by the cataloger. It w a s immediately evident that a skilled typist, w i t h experience in copying catalog titles, could, by familiarizing herself w i t h the simpler rules for collation, prepare a title sufficiently complete and suitable in form to enable her to type it directly on regular card stock. T h i s w o u l d make it possible to use the card for the official catalog and thus save one recopying of the title. I t w a s found that a high percentage of titles so prepared needed little or no editing, provided that the correct heading could be determined in advance. F o r our G e r m a n philosophers, a list of correct headings w a s easily compiled, from which the typist worked. T h e notable success of this experiment—some 9 6 7 volumes w e r e handled at a cost of 3 4 2 cents a title — r e s u l t e d in applying the method both to miscellaneous collections and to regular w o r k . A n accurate but moderately ex- perienced cataloger w a s , therefore, assigned to the initial process of checking n e w ac- cessions w i t h the union catalog in order to provide the typist w i t h exact headings, whenever these could be procured from the catalog w i t h o u t research. A large per- centage of titles thus came through to the subject supervisor bearing exact headings copied f r o m the c a t a l o g ; some had pre- sumably correct headings, also derived from the catalog, w h i l e for other entries, not found in the catalog, the title pages w e r e themselves adequate. A minor quan- tity, only, came through without a heading that w a s either perfectly, or else essen- tially, correct. ( T h e searcher n o w uses a special check to indicate that he believes he has found a heading absolutely correct.) T h e system w a s then perfected by rul- ing that the person in charge of the pre- liminary searchers should look over the day's influx and eliminate snags, unusual material, and rare books. T h e s e should not pass through the preliminary processes, but be distributed directly to the subject supervisor concerned in each case, thus pinning on this supervisor greater responsi- bility for the physical care of such ma- terial while it is in transit, and for its suitable and prompt treatment. A s the method developed, it became evi- dent that the subject supervisors should keep constantly in mind that the persons engaged in the preliminary checking w e r e 236 COLLEGE AND RESEARCH LIBRARIES not to be held responsible for anything except copying with perfect accuracy such data as headings and call numbers found in the catalog, comparing with care, title, date, and edition, for the purpose of re- vealing duplicates, and noting certain other matters that experience found useful. These checkers were not, when there was doubt in the matter, to attempt to prove identity of authorship, provide full names, search bibliographies, look up anonyms, or attend to other matters more properly the duty of experienced catalogers. It was not for the searcher to determine, for exam- ple, whether the book in hand by Fred- erick L a w Olmsted was written by father or son. So far as cataloging was con- cerned, the prime object of the preliminary processes was to prepare, without research, a provisional card for the use of the sub- ject supervisor. T o this card she could devote the energy of scholarly editing without being hindered either by the mechanical labor of typing the initial card herself or by writing out notes from which the card could later be made by a typist— the marking of title pages for this purpose being, of course, taboo. Matters had to be adjusted, however, so that the provi- sional card would, in a sufficiently large percentage of cases, prove good enough to serve as the permanent official card for the union catalog, thus sparing the danger of errors from a second copying and waste of time from additional proof reading. Other information found and noted by the preliminary searcher was to be regarded as a by-product to be used with discrimina- tion. When the subject supervisor received books by such well-known authors as Ab- bott Lawrence Lowell or Henry Wads- worth Longfellow, she would be confident that the spelling of the names and the date of birth (or of birth and death) had been correctly copied from the catalog. She would not herself have to go to the catalog to verify the data and, from her knowledge of books and authors in her subject, she would be absolutely sure of the identity of the author. When, how- ever, she received a book by a person with whom she was not familiar, even if the author's name were as distinctive as Fred- erick L a w Olmsted, it would be neces- sary for her to be constantly on her guard lest she (or, as usually happens, her suc- cessor) be humiliated, later, by a father- and-son confusion in the catalog. In fact, a competent supervisor will normally se- lect from her day's influx the books for which she deems it necessary to verify the searcher's findings. T h i s verification is the more essential if, as may happen when large gift lots are handled, some little time has elapsed between the time of prelim- inary searching and the cataloging, for a good catalog is perpetually in a state of flux. Some Results While the preliminary processes just described originated in a desire to simplify the routine of cataloging and decrease its cost, it became obvious that they would be of perhaps greater value in introducing a new routine in the matter of receiving, checking, and absorbing new accessions of library material. T h e two processes, (a) of determining the question of rejecting or retaining duplicates after a collection has been checked with the catalog, and (b) of making the necessary cataloging records after a decision as to duplicates has been reached, are, to a great extent, one and continuous. T h e person of good judg- ment, who has sufficient information about the book or collection to decide the first JUNE, 1940 23 7 matter is usually in a position, if trained and experienced as a cataloger, to attend to the second matter, often w i t h but mod- erate additional labor and expense. T h i s is particularly true of difficult material. T h e result then of this n e w method has been to pass on to the cataloger-classifier, w h o is responsible for a subject, a f a r greater share in deciding the question of whether duplicates in her field are to be rejected or retained. T h i s procedure has been particularly useful w h e n the staff is handling large collections devoted to some special topic. F u r t h e r m o r e , it cen- ters the responsibility for sane and efficient treatment on those individuals w h o , from knowledge of the subject, technical ex- perience, and familiarity w i t h the library's collections, can best bear such responsibility. B y having such additional duties placed on their shoulders, the persons concerned take greater interest in their w o r k and increase their knowledge of the subject and of the library's holdings. T h e y thus become daily more useful than if they confined their duties merely to recording w h a t is put on their desks to record, after some- one else has studied it and made essential decisions. "Subject Supervisors" D u r i n g the period of incubation, while the method thus described w a s being worked out, there w a s considerable ques- tioning by the heads of departments as to the wisdom of thus increasing the duties and responsibilities of those whose princi- pal duties had hitherto been confined to the processes of cataloging and classification. I t seems evident, however, both from the standpoints of economy and of satisfactory results that the decision as to retention of duplicates may, in large measure, be most appropriately apportioned to the catalogers and classifiers specializing in definite sub- jects, and that to these persons w e might w e l l assign the title "subject supervisor." I t is w e l l to emphasize the idea that not only must these persons have had past ex- perience in cataloging and classifying, but they must also be the actual persons w h o are daily engaged in, or closely supervis- ing, this w o r k on their own subjects. T h e distribution of the vast influx of books to several persons, each more or less expert in her special subject, is, on the v e r y face of it, a wiser practice than concentrating the responsibility on an accessions chief w h o attempts to cover the multitude of subjects within the scope of a large schol- ar's library and to deal w i t h the many languages involved. S u r e l y a policy is untenable whereby each department of the library (order, catalog, reference, etc.) has its own specialist in each field in which the library is active. R u n n i n g parallel to this increased re- sponsibility of the subject supervisors has come the annexation, to a greater or less degree, by the supervisor in charge of the preliminary searching process, of the man- agement of g i f t accessions, including large collections and the daily dribble. A minor, but essential, advantage resulting has been the diminution of books to which, on re- ceipt, the library's marks of ownership have been overhastily affixed, the item be- ing rejected after more careful scrutiny. A t present, the library stamps and book- plates are affixed almost as the last process, and thus valuable material may be more easily earmarked for special care in the mechanical, and at times abused, process of stamping and plating. I t has been found that the preliminary searching thus described can be done suit- ably by the better grade of student as- sistants, each of w h o m puts some twelve or 238 COLLEGE AND RESEARCH LIBRARIES fifteen hours a week into the w o r k . It is wiser not to keep a person at the job too long at one period, since it requires close attention to detail which cannot be given by a person physically tired from long standing at the catalog. I t is, moreover, necessary to have a sufficiently large num- ber of persons assigned to this w o r k so that current acquisitions may go through f r o m day to day w i t h the utmost promptitude, to save the occasional need of hunting for books that may be in process. I t has, at times, been our custom to establish a list of persons w h o may automatically be taken from their regular w o r k for an hour or t w o each, when there is danger of delay in handling the day's accessions. T h e chief of the typing department in turn sees to it that w o r k is so arranged that live and important books are pushed through her department immediately on receipt. I t is our intention to have these live and current books in the hands of the subject super- visors within t w e n t y - f o u r hours after they come into the department. In this w a y each subject supervisor is made responsible for the prompt handling of books in her field. N e w English and A m e r i c a n books have, in the past, usually been put into cir- culation from one to three days f r o m the time of receipt in the department, w i t h a f u l l set of cards in both catalogs within an additional day or two. T h i s speed de- pends largely on the exertions of the sub- ject supervisor. A w o r d as to "continuations." N o r - mally, these are distributed at once to the serial division and are not handled by the preliminary searchers. A s the serial divi- sion makes its record, it notes such titles (e.g. monograph series) as need analyzing. T h e s e titles are passed on to the prelim- inary searchers and go through the same routine as other monographs, being special- ly marked by a characteristic slip in order to procure prompt attention and speedy routing. Since the serial record notes them as received, they must not be delayed in the hands of the subject supervisors. Being already recorded as received on the serial records, there is the greater need that they should not be delayed in the later processes of acquisitioning and cataloging. Duplicates t h r o w n out by the prelim- inary searchers must undergo scrutiny, first by the supervisor in charge and, when desirable, by the subject supervisors or by specialists available, w i t h a v i e w to spot- ting valuable items not obvious to search- ers inexperienced in such matters. T h e searchers are, however, instructed to no- tice, and set aside for inspection, all books w i t h annotations on fly leaves or margins, that contain inserts, or that bear seemingly interesting autographs. Savin ffs A t H a r v a r d , this combination of the preliminary search to determine the cor- rect catalog heading and find possible L i - brary of Congress cards, w i t h the survey of material to reveal duplicate copies, has reduced a process, which in some libraries requires three separate searchings in t w o card catalogs, to a single comparison of the book w i t h the union catalog. W h e n it is remembered that the rate of such search is less than twenty titles an hour, the sav- ings for a yearly accession of 5 0 , 0 0 0 titles might be as high as 5 0 0 0 hours when three searchings are necessary. In the year 1 9 3 8 - 3 9 , at the H a r v a r d library, it took 2 3 7 0 hours to perform the preliminary searching for 3 9 , 9 8 2 titles, an average of only 1 6 . 9 titles an hour. A t this rate the saving is greater, i.e. some 3 0 0 0 hours for one searching saved, and 6 0 0 0 hours for two. T h e saving in time is more than that JUNE, 1940 23 7 of t w o full-time assistants, or the equiva- lent of from $2000 to $ 3 0 0 0 annually. A s s u m i n g , however, that for one-third of the titles f u r t h e r verification is necessary on the part of the cataloger, the saving might still be as high as f r o m $ 1 7 0 0 to $ 2 5 0 0 , for one as against three searchings. Additional saving results f r o m this n e w method in the f a c t that the mechanical checking of titles w i t h the card catalog and the typing of the first card are done by persons of less pay than the cataloger or subject specialist. S o many elements go into this figure that it w o u l d be a d i f - ficult or impossible task to reduce it to exact statistics, but if w e assume a saving of but t w o cents a title on these t w o proc- esses w e get an additional saving of $ 1 0 0 0 on a yearly output of 5 0 , 0 0 0 titles. A n - other notable, but indeterminate, amount is saved by the elimination of conferences and discussions between subordinates and chiefs of the accessions and catalog depart- ments, when questions arise as to in- dividual items or lots. T h i s , I am sure, w i l l be acknowledged by those w h o have watched intimately the progress of w o r k in the staff of a large li- brary. T o sum up, the H a r v a r d library, bene- fited by its combination official, union, and depository L . C . catalog, n o w unites t w o processes in its w o r k of accessioning books. T h e s e t w o processes are, ( a ) the survey of new material w i t h a v i e w to discover- ing duplicates, and ( b ) a preliminary checking to prepare the w a y for cataloging. T h i s checking reveals, as occasion offers, the correct entry heading, the existence of L . C . cards that may be used for cataloging, the location of copies in department li- braries, and other useful data. F o l l o w i n g this preliminary w o r k , a skilled typists' de- partment prepares provisional cards for the cataloger, thus relieving the latter of the time-consuming labor of preparing her key card herself or of making complicated notes w i t h a v i e w to having the result re- duced to card form by a typist. T h e an- nual savings for 5 0 , 0 0 0 titles added by this method might be f r o m $ 2 5 0 0 to more than $ 4 0 0 0 ; moreover, it surely reduces the time needed to make new books avail- able. F u r t h e r , the subject supervisor w h o has a larger share in the duty of com- paring duplicates and deciding w h a t shall be rejected, acquires a more intimate con- nection w i t h the development of the sub- jects assigned her, frequently gets more closely in touch w i t h those of the faculty whose fields are concerned, and becomes increasingly the center of information for her o w n field of w o r k . T h e placing of all books in her subjects promptly in her care pins down her responsibility for these books—a most useful procedure w h e n a question suddenly arises as to the disposi- tion or whereabouts of a given item or a special lot. T h e r e has been v e r y little excited search for a missing item in our staff of classifiers and catalogers, and su- pervisors have become w a r y of disclaim- ing knowledge as to the whereabouts of a book or pamphlet in their particular fields, for it is becoming an axiom that a missing uncataloged book is a l w a y s found just where it ought to be—in the subject super- visor's hands. 240 COLLEGE AND RESEARCH LIBRARIES