College and Research Libraries B y M I L E S O . P R I C E The Need for a Subject Classifi- cation to Facilitate Research in Foreign Law Miles O. Price is librarian of the Law School of Columbia University. THE admirable paper by Professor Frederick C . H i c k s in the June Col- lege and. Research Libraries, o n t h e p r o f e s - sional aspects of l a w librarianship, w a s w h a t his colleagues have learned to expect from his scholarly pen. H o w e v e r , in his discussion of subject classification for l a w libraries he has overlooked a most impor- tant factor concerning the research type of l a w l i b r a r y — a factor which persuaded me to yield to the requests of my library committee here at C o l u m b i a to install such a classification. T h i s factor is the peculiar requirements of the research worker, whether teacher or graduate stu- dent, especially in the field of foreign l a w . I thoroughly agree w i t h Professor H i c k s that for most l a w libraries a subject classification w o u l d be an expensive nui- sance, more likely to confuse and delay than to help. Perhaps a brief explanation of the reasons back of that general feeling among l a w librarians in opposition to classification w o u l d be pertinent here. A n g l o - A m e r i c a n l a w is based upon the English common l a w as modified by stat- ute, and the common l a w is w h a t the lay- man w o u l d call u n w r i t t e n l a w . In other words, it has g r o w n up as a sort of crys- tallized custom, and its rules are found in no statute books. O f late years statute l a w has become increasingly important, but the foundation of the rules that govern most of our daily actions is still the com- mon l a w . T o make a too great simpli- fication, the l a w — c o m m o n and statutory — i s in the last analysis w h a t the appellate courts say it is, and the w a y a court says w h a t a rule of l a w is, is by a decision in a concrete case. T o avoid the endless confusion and consequent hardship of in- consistent decisions, there has g r o w n up a rule known as stare decisis, w h i c h means that a t r u l y decided case is to be regarded as authority in similar cases arising in the f u t u r e in the same court or in l o w e r courts in the same jurisdiction. F o r sufficient reason the same court may overrule or m o d i f y an earlier decision, by a later one, but it is not often done, especially w h e n real property rights are involved. T h e pertinency of all this for this dis- cussion is that the authority which a l a w - yer seeks is found in the statutes and in the decisions of appellate courts w h i c h de- cide rules of common l a w and interpret statutes, and not in textbooks. T e x t b o o k s in l a w are not authority, but merely more or less elaborate case-finders and indexes to or discussions of the l a w . T o be sure, certain commentaries, as Blackstone and 146 ' COLLEGE AND RESEARCH LIBRARIES K e n t , or monographs, as W i g m o r e on evi- dence and W i l l i s t o n on contracts, have achieved high prestige as secondary au- thority, but no j u d g e is bound to respect or f o l l o w them. Preponderance of Unclassifiable Material T h e result is that a conventional w o r k - ing l a w library, not of the research type, is composed of a great many l a w reports, some statutes, and several groups of guides to these reports and statutes; such as di- gests, tables of cases, citation books, ency- clopedias, legal periodicals, and textbooks or treatises. O f these, only the last two, and for practical purposes in A n g l o - American l a w , only the last, are suscepti- ble to subject classification. T h e others are grouped in a conventional fashion which, as Professor H i c k s has pointed out elsewhere, is really a classification dictated by the nature of large sets, but not a sub- ject classification as in history, chemistry, or geology. Session laws are bound tip in the order of their approval by the president or governor. Occasional revisions and con- solidations of statutes in force are classi- fied within themselves, but are of course shelved w i t h other statutes of that juris- diction. T h e l a w reports in each bound and consecutively numbered volume of re- ports are arranged chronologically by the date of each decision. T h e r e are probably three million decisions, and they pour forth from the courts at the rate of 30,000 a year in A m e r i c a alone. T h e y are made accessible by digests, which are arranged by subject classifications familiar to all l a w y e r s ; by tables of cases, all of w h i c h are shelved near the reports themselves; and by treatises, which cite cases in sup- port of their o w n views. T h e minimum library of 10,000 vol- umes required of member schools of the Association of American L a w Schools dis- misses classifiable material w i t h four words in its fifth specification: " ( 5 ) . L e a d i n g up-to-date publications in the w a y of gen- eral digests, encyclopedias, and treatises of accepted worth." A rather idealistic catalog for a l a w library of 15,000 vol- umes d r a w n up for my course in l a w library administration three years ago by Raymond Lindquist, librarian of the N e w Y o r k L a w Institute (a library of 125,000 v o l u m e s ) , contained only 781 volumes of legal periodicals and 1242 of textbooks, the remaining 13,000 volumes being strictly nonclassifiable. A n d the number both of legal periodicals and of texts as given by M r . Lindquist w a s considerably higher than w o u l d be found in the ordi- nary w o r k i n g library of 15,000 volumes. Eliminating legal periodicals, w h i c h are for the most part of a general nature in America and not classifiable by subject, w e have about one-twelfth of the library classifiable at the most. T h i s is a high percentage for l a w libraries, and the smaller the library, the lower the percent- age of texts. O f the approximately 631 A m e r i c a n l a w libraries listed in the Standard Legal Directory for 1939, 161 contained less than 5000 volumes; 146, between 5000 and 10,000 volumes; 1 5 1 , between 10,000 and 20,000 v o l u m e s ; 94, between 20,000 and 50,000 v o l u m e s ; 39, between 50,000 and 75,000 volumes; 28, between 75,000 and 150,000 volumes; and 12, over 150,000 volumes. ( O f the large libraries, several w e r e state libraries, containing more nonlegal material, such as history or economics, than is usual in l a w libraries.) T h a t is, 75 per cent of the organized l a w libraries listed contained less than 20,000 volumes, and by the same token, probably, MARC hi, 1941 147 only a negligible amount of material classifiable by subject. Because of the large proportion of l o n g sets, a l a w li- brary of 20,000 volumes is not considered large, and is likely to be a "one-man" library. Also, because of the relatively f e w texts, its catalog, if any, is probably rudimentary. I k n o w of one bar associa- tion library of 70,000 volumes, and a good one, which has no catalog, but only a short-form author list. A l l of this material, except possibly texts, is kept on open shelves for the l a w - yers' ready consultation, w h i c h is another argument against classification. E v e n w h e n the library increases in size, the material is probably for the most part sets of continuations, such as added reports and statutes, or printed appeal papers, re- ports of attorneys-general and bar asso- c i a t i o n s — a l l similarly unclassifiable by subject. N o r m a l l y , the l a w y e r is looking for a case, and other material is incidental to that search. T h a t is true even of good texts, which a f t e r all embody only the opinion of unofficial persons as to w h a t the l a w is. T h e l a w y e r w i l l sometimes cite texts as make-weights, but w h a t he wants is a case in point. T h e r e f o r e , he is searching for definite citations, as P r o - fessor H i c k s points out, and does not care where he finds them, and under the cir- cumstances a subject classification is a nuisance. I have so told many librarians of small collections w h o were interested in subject classification, and advised them not to attempt it. Problems of Closed Stacks and Research fV orkers H o w e v e r , w h e n a substantial part of the library o u t g r o w s the open shelves, and w h e n research workers are to be con- sidered, that situation loses some of its effect. Research workers are interested in citations in other books, it is true. M o r e important to this discussion, how- ever, is that they are pushing back the boundaries of knowledge and are soon be- yond the point where they depend exclu- sively or mainly on somebody else's cita- tions. T h e y are themselves giving the citations. Professor H i c k s is quite correct in stat- ing that legal literature is f u l l y covered by printed indexes w h i c h are probably far more complete and up to date than in any other field. T h a t is, he is correct as to A n g l o - A m e r i c a n laws, statutes, reports, and periodicals. T h e s e are covered by complete, speedy, and relatively inexpen- sive index services beyond those dreamed of in other fields. H o w e v e r , these do not cover foreign l a w w e l l , and treatises not at all. F o r example, in the C o l u m b i a U n i v e r s i t y L a w L i b r a r y there are over 25,000 volumes of treatises in A n g l o - A m e r i c a n l a w alone, not counting inter- national l a w and foreign relations. T h e s e are not on open shelves and are available as open-shelf books only to graduate stu- dents, faculty, and editors of the Law Review. H o w e v e r , real accessibility is gained only through our catalog, w h i c h is a considerable distance from the stacks. W e librarians must admit that our patrons hate to use a catalog, and the larger our collection and the more the cards, the more they hate it. It is an invaluable tool, but a nuisance to the reader, nevertheless. It is necessary to supplement a classification, too, of course, because authors do not w r i t e books w i t h a subject classification in mind. H o w e v e r , this is true of a classification covering any subject, as chemistry, engineering, or eco- nomics, all of which have been successfully 148 ' COLLEGE AND RESEARCH LIBRARIES classified for years. A n d in no science is it truer than in l a w , that authors are w r i t i n g "more and more about less and less." E v e n in A n g l o - A m e r i c a n l a w , for every general w o r k on a large topic, such as C o o k or Fletcher on corporations, w e have a dozen or more like Berle, Studies in Corporation Finance; D a v i s , Essays in the Early History of American Corporations; Doris, Corporate Meetings; L a t t y , Sub- sidiaries and Affiliated Corporations; Spellman, Corporate Directors; T r a c y , Corporate Foreclosures; a n d t h e l i k e . C e r t a i n l y there is little overlapping here. T h e librarian in chemistry or history or economics would be horrified at the sight of 25,000 unclassified treatises, ac- cessible only through even the best possible catalog. Since the success of our foreign l a w classification I am beginning to feel more and more that w a y myself. H o w - ever, w e are approaching the matter care- f u l l y , because of the practical unanimity of l a w librarians' opposition to subject classification, and of the cogency of the arguments marshalled by Professor Hicks. B u t there is authority on the other side, too. T h e late G . E . W i r e , longtime li- brarian of the W o r c e s t e r C o u n t y ( M a s s . ) L a w L i b r a r y , and the chief protagonist of the subject classification of l a w books, employed such a classification successfully for years, even w i t h open shelves and in a medium-sized library. T h e frequent re- quests from l a w librarians for a subject classification of A n g l o - A m e r i c a n l a w books indicate that the matter is by no means permanently settled. Subject Classification in Foreign Law W h e n it comes to foreign l a w w e at Columbia are in no doubt at all as to the value, if not necessity, of a subject classi- fication for treatises. W e have one and it works. Perhaps the makeup of the li- brary committee which persuaded me to consent to a, subject classification w i l l help to explain my point: its chairman is pro- fessor of Roman l a w ; another member is professor of comparative l a w , doing most of his w o r k w i t h foreign legal systems; and the third is professor of legal history. T h i s committee felt that while, as far as they knew, the needs of the w o r k e r in A n g l o - A m e r i c a n l a w might be sufficiently taken care of by the conventional alpha- betical arrangement of treatises, they w e r e so hampered in their research in foreign l a w that a subject classification was w o r t h its great cost. T h e r e are several reasons w h y a subject classification in foreign l a w in this coun- try stands on a different footing f r o m that of A n g l o - A m e r i c a n l a w . O n e is the much higher standing of the commentary and treatise, as opposed to the l a w report, in the civil l a w which is the basis of most foreign l a w . In f e w civil l a w countries does the l a w report have anything near the prestige it enjoys in this country or E n g l a n d . In some constitutions it is even provided that no court shall be bound by a prior decision. A treatise or commen- tary by a recognized scholar in a given field of l a w may have more authority than that of a line of decisions, and usually the courts w i l l not have the temerity to go counter to such a commentary. T h e r e - fore, the treatise is much more important in foreign l a w than in A n g l o - A m e r i c a n l a w . It may be stated here also that the legal periodical in civil l a w countries is considerably more likely to be devoted to a relatively small area of the l a w than in this country, and therefore classifiable by subject. Probably the impelling reason in our case, however, for adopting a subject class- MARC hi, 1941 149 ification w a s the fact that practically the only users of our foreign l a w collection are research w o r k e r s — g r a d u a t e students and f a c u l t y — a n d that the entire collection is in the stacks, at a considerable distance from the catalog. T h e r e w a s thus a vitally different setup f r o m that existing in A n g l o - A m e r i c a n l a w . Results of Subject Classification A f t e r three years of studying the pros and cons, I w a s convinced that my library committee w a s justified in its position, and the re-classification of our foreign l a w col- lection w a s begun. T h e classification, except for some minor matters of form, mnemonic features, etc., w a s devised by A . A . Schiller, chairman of the library committee, professor of R o m a n l a w , and expert on foreign legal systems. W e feel that the results have f u l l y vindicated us. Research workers now can go to the stacks and find most of the material they need in a closely and logically classified group, instead of having to w a n d e r through thou- sands of alphabetically arranged works. T h i s saves time and brings together in one place the actual material to be w o r k e d w i t h — w h i c h is much more satisfactory than even the best subject catalog (and at C o l u m b i a w e have one of the best). W h e r e formerly the research w o r k e r in French l a w , for example, had to p a w through 2500 alphabetically arranged treatises, a f t e r laboriously copying call numbers at the public catalog, he can now go to the stacks, either a f t e r getting the call number of his topic from the catalog, or after consulting the classification scheme posted in many places in the stacks, and there find his material together. T o be sure, he w i l l , as in other sciences, have to check w i t h cross-references in the catalog to be certain that he has missed nothing, but that is a much shorter process than depending w h o l l y upon the catalog. In our library considerable research is done in the fields of family l a w and indus- trial property (patents, trade-marks, copy- right, e t c . ) . T h e l a w of persons (including f a m i l y ) is embraced in num- bers 329-99 of our classification. M a r - riage, numbers 370-99, has 170 titles in French l a w alone. Industrial property is 670-90. W h e n a comparative study is being made of that subject, the w o r k e r does not have to spend hours at the cata- log, or searching through 2500 French, 2000 G e r m a n , 600 D u t c h , and 400 Italian titles of treatises to find his material. In- stead, he quickly finds nearly all of it in our 97 French, 120 G e r m a n , 20 Italian, and 17 D u t c h titles of treatises and special periodicals, in their proper places on the shelves. Professor H i c k s has pointed out the dis- advantages of a subject classification in the field of l a w , and I agree w i t h him. I firmly believe, however, that in the large research library, particularly w h e r e there is a very large group of treatises, and more particularly w h e r e there is a large collec- tion of foreign l a w used by research w o r k - ers rather than by the bulk of the library patrons, the subject classification is justi- fied and provision should be made for it. 150 ' COLLEGE AND RESEARCH LIBRARIES