College and Research Libraries By C H A R L E S F. G O S N E L L Values and Dangers of Standard Book and Periodical Lists for College Libraries Charles F. Gosnell organized and is li- brarian of Queens College Library, Flush- ing, New York. ST A N D A R D S serve two purposes: as cri-teria for past accomplishments, and as bases for present and f u t u r e action. T h e present study of standard lists has been a part of an unusual experience of organizing a new college library, depend- ing heavily upon such lists. Queens College was created late in the spring of 1937 by fiat of the Board of Higher Education of the City of N e w York. Some buildings were renovated and a staff gathered during the summer; 400 freshmen were admitted to classes in October, 1937. T h r e e and a half years later there is a student body of over 2000 still growing, and a library of 45,000 volumes. T h e best way to begin gathering a li- brary in a hurry seemed to be to depend upon standard lists, and multiple copies of the leading lists were among the first books purchased for the new library. It was immediately apparent that the S h a w List of Books for College Libraries/ begun in 1928, contained a large propor- 1 Carnegie C o r p o r a t i o n of New York. Advisory Group on College L i b r a r i e s . A List of Books for College Libraries. P r e p a r e d by Charles B. Shaw. A . L . A . , 1931. 8 i o p . tion of out-of-date titles which the col- lege did not need and could not afford. Nevertheless one copy was cut into subject sections and sent out in sections for check- ing by members of the new departments of instruction. T h e library staff checked another copy. T h e checks were combined, and the checked entries clipped and mounted served as excellent order cards. M u d g e ' s Guide to Reference Books2 was similarly checked, but only a small proportion of these titles was found to be necessary or within the grasp of the library. All "basic" titles listed by Shores3 were clipped and added to the order file. C a r e f u l examination of the M o h r h a r d t List of Books for Junior Col- lege Libraries4 showed it to be excellent as a buying list. It was up to date, its titles were of a high quality, quite suited to all years of a four-year liberal arts col- lege. Accordingly all of the 5585 M o h r - hardt titles were added to the order list. T o the " s t a n d a r d " titles were added several thousand suggested for immediate purchase by members of the faculty. 2 M u d g e , I s a d o r e G. Guide to Reference Books. A . L . A . , 1936. 504P. 3 Shores, Louis. Basic Reference Books. P r e - l i m i n a r y ed. A . L . A . , 1937. 4o6p. (2d ed. 1939 is now available) 4 Carnegie Corporation of New Y o r k . Advisory Group on J u n i o r College L i b r a r i e s . A List of Books for Junior College Libraries. Compiled by F o s t e r E. M o h r h a r d t . A . L . A . , 1937. 386p. 216 COLLEGE AND RESEARCH LIBRARIES Nearly three-fourths of the faculty sug- gestions made independently proved to be duplicated in the standard group. T h e remaining fourth were incorporated into the buying list. Virtually all of these titles have now been secured. T h u s the Queens College collection is closely related to the standard lists. Some parts were so closely related that checked copies were used as subject guides until catalogers could begin to catch up with the flood of purchases. T h e lists were fundamental, and suggestive stimulants as w e l l ; but in no sense were they regarded as restrictive. T h e y were always subject to reinterpretation in the light of the curriculum. Periodicals As Queens never aims to be a great research library—it has too many rich and generous neighbors in M a n h a t t a n — i t s periodical subscription list is a modest one. In selecting the basic titles, there was heavy leaning upon the subject lists given in each section by Shaw and M o h r - hardt, and upon the independent lists compiled by Lyle,5 W a l t e r , 6 and Hilton. 7 T h e subscription list for the second year was extensively revised. M a n y of the more popular items, to be had in public library branches, were eliminated. O t h e r titles, especially foreign ones, and learned journals definitely related to specific courses were added. Periodicals require, and can be given, more attention per title than books. It is possible for one to be familiar with a periodical year after year as he cannot 5 L y l e , Guy R. Classified List of Periodicals for the College Library. 2d. ed., rev. a n d enl. F a x o n , 1938. 96p. 8 W a l t e r , F r a n k K . Periodicals for the Small Library. 6th ed. A . L . A . , 1932. 114P. ( 7 t h ed. 1939, is n o w available) 7 H i l t o n , E u g e n e , a n d W a p l e s , D o u g l a s . " P e r i o d i - cals f o r t h e College L i b r a r y . " North Central As- sociation Quarterly 8:425-44, A p r . 1934. be with each book of the month. T h e y can be subject to a continuous re-evalu- ation in the way that books cannot be. Each new year's selection is based on the past year's experience. The New Shaw Supplement T h e appearance of the Shaw Supple- ment8 in the spring of 1940 put Queens College Library back into the class of older libraries, enabling a check on past performance against a standard selection which was not available when purchases were made. Queens' holdings of titles in the supplement were checked complete- ly, except for periodicals, and it was found that the library had 52 per cent of the titles. Psychology was the highest class, with 78 per cent. Political science and geology followed with 7 1 P e r cent. T h e lowest group was religion, with 17 per cent. Queens is a municipal institu- tion, and does not offer courses in this subject. Holdings in the classics were low (small registration in this depart- ment) and also in philosophy (courses limited to juniors and seniors, and still being organized). I t was interesting to find a slight gen- eral tendency to have a larger proportion of the more recent titles than of the older ones in the 1931-38 list. T h i s tendency was not pronounced in any class (the time span was too short, and the number of titles too small) but it was clearly evident in history and sociology. Such are the practical values of the standard lists. By inference, the library that makes intelligent use of them ought to be admitted to the inner circles of accredited institutions. It is fairly sure, however, of getting a reasonably good 8 Shaw, C h a r l e s B., comp. A List of Books for College Libraries, 1931-38. A . L . A . , 1940. 284P. JUNE, 1941 217 Year of pub!ication F I G U R E I . D A T E S O F P U B L I C A T I O N O F T I T L E S I N T H R E E S T A N D A R D L I S T S O F B O O K S FOR C O L L E G E L I B R A R I E S collection of books in an economical way. But I believe that these standards ought to have more lasting values and broader implications than a mere listing of indi- vidual titles. W e ought to be able to take a list apart and to see what makes a good library book collection. I like to think of the lists as a momentary freezing of the stream that is a good college library book collection. T h e college library is in a state of continual flux. T h e lists are permanent only in the sense that they are not constantly remade. W e ought to get some clues to enable us to determine the proper size of a col- lege library collection—assuming that there is an optimum size. H a v i n g fixed the size, we should be able to determine the yearly additions required to keep it up to date. Conversely we should have an 218 COLLEGE AND RESEARCH LIBRARIES active policy of discarding to avoid stag- nation. Obsolescence and Mortality From the interchange of new and old, we should derive a rate of obsolescence and mortality which would be useful for budget and accounting purposes. If there are mortality tables for hu- mans, why not for books ? Like their authors, books grow old and die. T h e flesh may remain in-mummified form, but the spirit goes. A few books may seem immortal, even in spirit, but they approach immortality through the successive re- incarnations of translations and editions. I refer to the life of a book as reading matter in the college library—as an active everyday participant in the educative proc- ess. I exclude 42 line Bibles and First Folios—which are to be looked at and not read, and which are not generally found in college libraries anyway. Randall9 and Eells10 have already done some work in the field. W i t h more data, covering a longer period of time, the stim- ulus of immediate budget problems, and the desire to keep the new collection at Queens alive, I propose to go further. W h a t follows is in the nature of an in- terim report. In Figure 1 are shown on logarithmic scale the distributions by date of publica- tion of titles in three standard lists, Shaw, Mohrhardt, and the Shaw supplement. It is immediately apparent that the rela- tively recent books predominate in each list. No adjustment has been made for fluctuations in actual book production, but such fluctuations are certainly not 9 R a n d a l l , William M. " T h e College L i b r a r y Book B u d g e t . " Library Quarterly 1:421-35, Oct. 1931. 10 Eells, W a l t e r C. " R e c e n c y as a M e a s u r e of Book Collections." Junior College Journal 8:308-10, M a r . 1938. sufficient to account for the predominance of recent titles, particularly in the last two decades. T h e drop in production between 1915 and 1920, due to war conditions, is reflected in the curves, but otherwise their slope is nearly a straight line. Ad- justments for such fluctuations will smooth the curves to a great extent. It may be inferred that consciously or unconsciously the compilers of these lists preferred the more recent books. If a compiler were to do his work over again, in later years, but with the same standards of selection, the curve would remain ap- proximately the same, except for dis- placement in time. T h e peak would advance a number of years corresponding to the time elapsed. Book Obsolescence Figure 2 illustrates such a conjecture in simplified form. A hypothetical list of books issued in 1920 had its peak be- tween 1910 and 1920, and contained almost no books published prior to 1870. A similar list, of the same size, and com- piled on the same basis twenty years later, in 1940, shows a peak between 1930 and 1940. It has almost no books older than fifty years, or published before 1890. Area A represents the new books in the 1940 1920 1900 1880 F I G U R E 2 . B O O K O B S O L E S C E N C E JUNE, 1941 219 194° list n o t included in the 1920 list. Area B includes books common to both lists. Area C represents books in the old list but dropped from the new. Area C is equal to Area A. Area A represents the births and C the deaths in the passage of twenty years. Similar curves, plotted for individual subject classes, show wide differences in their slopes or rates of decline. T h e peak for classics is not high, and the drop is very slow. Chemistry has a high peak, and drops sharply. O t h e r subjects tend to vary with the speed of research and development in the field. T h i s family of curves resembles the generally recognized "decay" curves, in- cluding the curve of organic decay, certain obsolescence, depreciation and mortality curves, and even the classic curve of for- getting developed by Ebbinghaus. T h r e e mathematical formulae may be considered: Pearson's type I and I I I curves, and the exponential curve: y = be~ax T h i s last is easiest to use. But it does not take care of the lag in selection during the time the list is being compiled. H e r e y = number of titles x = years elapsed T w o parameters may be regarded as keys to the character of the collection or se- lection : b = maximum number of books for any one year (i.e., when x = o ) a = the index to rate of obso- lescence in the group If this formula and these parameters can be established for general collections and subject subdivisions, a library can be tested at any time by taking a sampling v^of its titles. A glance at the resulting curve will show whether the library is steadily adding new material, and whether there is progressive discarding of old. It will be possible to substitute objective rat- ings in place of the intangible terms "live" and "up-to-date." Presentation of objective data on obso- lescence in terms which budget makers and accountants can understand should be of great help when the librarian seeks funds for the new books each year. 220 COLLEGE AND RESEARCH LIBRARIES