CATALOGING FOR DEPARTMENT LIBRARIES
Before beginning the discussion of cataloging for department libraries, let me say that as it is a subject which is still in the experimental stage and not yet capable of generalization, the statements made in this paper are based, partly upon information collected from certain university libraries in which this problem is now being worked out, and partly on my own experience in organizing the department catalogs of the University of Minnesota. The other libraries quoted are those of the University of Chicago, Columbia, Illinois, Johns Hopkins, Michigan, Missouri and Wisconsin.
In considering the problem of cataloging for department libraries, we may start with a definition and an assumption. For the purpose of this discussion it may be said that a department library is not a mere handful of reference books on a subject, but a more or less comprehensive collection of books on the subject shelved and used separately from the collections of the main library; and it may be assumed that the necessity for a separate catalog of such a collection is admitted by all.
Assuming this, the first question that presents itself is that of the form of the department catalog. Shall it be an author, a classed or a dictionary catalog, or, since in most cases the department library is a small open-shelf collection, will it suffice to have a shelf-list only, serving also as a classed catalog? The shelf-list would offer the simplest and cheapest solution of the difficulty, but the day when it was accepted as a solution of the entire problem has passed. Not one of the libraries consulted suggests the shelf-list alone as a possible arrangement. An author catalog, at least, is needed in addition, and the majority of these libraries report dictionary catalogs in some of the department libraries, if not in all. Chicago University is to provide for the department libraries outside of Harper building an author catalog and a shelf-list, where printed cards are available, and an author catalog only for the department libraries within Harper building. Columbia, Michigan, Illinois and Minnesota have dictionary catalogs for all department libraries. Missouri has dictionary catalogs in 3, and Wisconsin in 2 department libraries, while Johns Hopkins is to have dictionary catalogs in all department libraries which are outside its main building.
A more difficult question is that of the scope of the catalog. How exhaustive is it possible, or even desirable, to make it? It must, of course, include all books in the department library itself, but shall it also record all books dealing with the same subject to be found elsewhere in the university? Such completeness of record would be the ideal arrangement, and would, undoubtedly, meet with the hearty approval of the university departments. But will not the cost be prohibitive to many libraries, even in this day of printed cards and multigraph? To be of value, such elaborate cataloging should be done thoroughly and systematically and above all, once undertaken, should never be allowed to lapse, or confusion will be the result. The fuller information about related materials in other parts of the library can always be obtained from the main library catalog, if that record is a union catalog of department libraries as well; and if the department librarian is in telephone communication with the reference librarian at the main library, the information can be obtained almost as quickly as if it were included in the department catalog. We may, therefore, conclude that the department catalog complete for its own library but not including related material in other libraries, is the most practicable form under present conditions, although the ideal form is the more complete catalog which expense at present generally prohibits.
The third point which our problem raises is that of variations in cataloging from the rules followed in the general library catalog. The first important variation which suggests itself as possible is in the treatment of analytics. Shall analytics be included in the department catalog, and if so, shall they be the same as those in the general catalog? On this point the practice of our eight libraries varies somewhat. Chicago University is not planning to include any analytics in its department catalogs, and Johns Hopkins includes only a few. Illinois, Michigan and Minnesota, in the main, duplicate for their department catalogs the analytics made for their main catalogs and, as a rule, include no additional analytics. The Columbia practice is more ambitious, as that library includes in its department catalog analytics (mainly articles in periodicals) which are not included in its general catalog. An article in the Columbia University Quarterly for March, 1911, states that the department catalogs have analytics for all important serials that bear upon the work of the departments whether shelved there or in the general library, that is, the department library catalog attempts to serve both as catalog and index. These cards are intended for temporary use only, to be removed when the demand for them ceases.
In most university libraries it would be impossible to keep up systematically such elaborate catalogs, and it is not clear that such indexing—for it is indexing rather than cataloging—would be desirable in all places. A catalog can never be made to take the place of a reference librarian, or of an intelligent use of the important annual and other subject indexes to the literature of a subject, such as Psychological Index, the various Jahresberichte, etc. Moreover, every reference or department librarian naturally does more or less in the way of keeping up card indexes or bibliographies, which are frequently revised and the old material discarded as new and better material takes its place. Such reference indexes are simpler and more practical than serial analytics in a department catalog, since they do not call for expert revision and absolute uniformity of subject headings. On the whole, the tendency of present opinion and practice seems to be that important analytics which are useful in the general catalog are useful in the department catalog also, but that beyond that it is better to encourage the use of the printed indexes and the keeping of an informal reference index for material not yet included in the printed aids.
A more important possibility of variation, where the department catalog is dictionary in form, is found in subject headings. Will the same headings that are found satisfactory in the main library catalog serve equally well in the department catalog as used by specialists? Too much emphasis can not be laid upon the fact that any variation of this kind greatly increases the cost of the cataloging, as the assigning and revision of two sets of subject headings, one for the general and one for the department catalog, will mean that that part of the work is greatly increased, though not doubled. The correct assignment of subject headings presents enough difficulties under any circumstances, and the catalog supervisor should hesitate to multiply these unless there is strong reason for doing so. In libraries which have adopted the Library of Congress subject headings, those headings, with minor variations, will, for most subjects, be found satisfactory in the department as well as in the general catalog. Law will at once occur to all as a subject for which it may be desirable to run two sets of headings. We have done this at the University of Minnesota, using the special Library of Congress law headings in the department catalog, and the regular Library of Congress headings in the general catalog. A point to be carefully considered in adopting more than one set of subject headings, moreover, is the possible confusion of mind that may be produced in the student, the exigencies of whose work require him to use more than one of the library catalogs. Such records are certainly much easier to use when there is uniformity of subject entries, and the adoption of several different sets of subject headings will certainly cause confusion, even to members of the library staff, much more to students.
After the questions of form, scope, and contents of the department catalog, comes the practical question of how best to get the work done. It can be done in either of two ways, by the regular cataloging force of the university or by the department librarians. In most university libraries the cataloging staff is small in comparison with the amount always to be done, and the work of keeping the general catalog up to date taxes all its powers, and leaves no time for extra records such as department catalogs. On the other hand, does not the department librarian have more or less time which, when properly arranged, could be given to cataloging under the direction of the head cataloger? We have found this to be the case at the University of Minnesota. Until three years ago our department libraries were all under the supervision of the various departments, and hence in a more or less chaotic state. Some of these have not yet emerged from chaos. In these three years, however, we have evolved a system by which this work is done by the department librarians, or, in one case, by an assistant in the department library. It has so far proved a perfectly workable system for our given conditions. All the department librarians so far appointed have been either library school graduates or people with equivalent library training, and in addition to that, in some cases, with special knowledge of the subjects of the departments. One of the first duties of the department librarian, on taking charge of his library, has been to organize it, classifying and cataloging it under the supervision of the head of the catalog department, but doing the work in the department library. The question has been raised as to how the department librarian could do the reference work and other work of his library, and at the same time catalog the department books for both the department and general catalog. Of course the cataloging will be intermittent and more or less interrupted, as our rule is that the work for the public must be done first. Until, however, the books of a department library are in order and properly listed, no satisfactory reference work can be done with them. Our own experience has certainly been that the reference work of our department librarians has been strengthened by their work of cataloging. The general library gains also from this work of the department librarian, as the latter does the cataloging of his books for the general catalog at the same time as that for the department catalog, and so the growth of the general library catalog is greatly promoted, without a corresponding tax upon the resources of the catalog department. In as far as possible the work is revised by the head cataloger or a reviser, in the department library, but in some cases of difficult revision it has been found necessary to transfer the books to the catalog department for revision there. At present, whenever printed cards can not be obtained, all cards are actually made by the department librarian, but as soon as we are able to adopt the multigraph, rough copy only will be supplied by that assistant.
After the department library has been thoroughly organized and cataloged, the department librarian goes on with the lighter task of cataloging the current accessions of his library for both the department and the general catalog.
Some of the advantages of thus having the work done by the trained department librarians are:
1. It adds several workers to the cataloging force of the library, and thus makes it possible to do much more in the way of providing needed departmental catalogs. This fact has been of great importance with us at the University of Minnesota, where, with the present cataloging force alone, it would have been impossible to provide these catalogs. Besides, there is the advantage to the general library of getting the cataloging of these same books done for the general catalog.
2. The department librarian should have, and generally does have, special knowledge of his subject, which is of assistance in cataloging, especially in classification and the assignment of subject headings.
3. As the work is done in the department it is easy for the department librarian to consult the professors whenever necessary or desirable.
4. There is a real advantage to the department librarian in the added familiarity with the department books which he has gained in cataloging them. This is particularly true in the case of the librarian who, in the beginning, is not a specialist in his subject, but even the specialist may gain some knowledge from this handling of the material which will help him in the service of his readers. Moreover, if he has actually made the catalog, he can use it more intelligently himself and instruct his students better in the use of it.
Our scheme has certain disadvantages as well as advantages. Some of these are:
1. There is danger that not enough cataloging research work will be done when the cataloging is done in the department library, because many of the important catalog and bibliographical aids are not accessible outside the catalog department—for example, the depository or union catalog of printed cards.
2. There is danger that the existing records will not be consulted enough, because the general catalog is not easily accessible and can only be consulted on special trips to the main library.
3. When the work is thus decentralized, there is much greater difficulty in obtaining from the various assistants work which is even fairly uniform. No one who has had experience in trying to manage such work will minimize this difficulty. For this reason, the revision is more difficult, and must be done with the greatest care, especially in the matter of subject headings.
4. It is sometimes more difficult to get good cataloging from those whose first interest does not lie in this branch of the work, and who are not closely associated with the regular catalogers, and familiar with the many traditions of a catalog department. For this reason we have found at Minnesota, that it is an advantage to have a newly appointed department librarian work in the catalog department for a time before taking up the work in his library.
We have found, however, that with us the advantages outweigh the disadvantages, particularly the one great advantage that it has been a practical way of accomplishing work which could not have been done by our present cataloging department.
Summary
A tabulated summary of the replies received from the libraries circularized has been prepared, but as it is too detailed for reading here, I will omit it, and present, instead, a few conclusions which may fairly be drawn from this summary. While practice is not uniform on any one point of department cataloging, certain tendencies toward uniformity are clearly evident.
1. In the matter of department librarians there is clearly a tendency towards the appointment of trained workers having, whenever possible, some special knowledge of the subject of their departments as well. This, of course, is important, if the cataloging is to be done in the department libraries. All these libraries feel, also, the need for some kind of department catalogs, although the number of such catalogs already established varies from three at the University of Wisconsin to twenty-three at Columbia.
2. There is a pretty general agreement that the dictionary catalog is the most desirable for department libraries. Columbia, Illinois, Michigan, and Minnesota report dictionary catalogs in all organized department libraries, and Johns Hopkins in all department libraries outside the main building. Wisconsin and Missouri report dictionary catalogs in certain department libraries, and Chicago reports author catalogs and shelf-lists. All the libraries using dictionary catalogs report the use, in the main, of the same kind of subject headings in department as in general catalogs, except for certain special subjects, such as law, or for certain highly specialized collections, such as the Avery Architectural library at Columbia.
3. There is a somewhat greater variation in the scope of material to be included. So far, only two libraries, Columbia and Michigan, report any department catalogs covering more than the material in the department libraries, but Missouri and Minnesota report that they intend, eventually, to have their department catalogs include all books on the subject in the university. In the matter of analytics the majority practice is to include the same analytics in both general and department catalogs, although Chicago uses no analytics at all in department catalogs, while Columbia, at the other extreme, includes more analytics in department catalogs than in the general catalog.
4. Present opinion seems to be pretty evenly divided on the subject of whether the actual work of cataloging should be done by the department librarians or by the regular cataloging force, although there is perhaps a tendency to have this work done by the department librarians wherever there are trained workers in charge of the department libraries. Universities in which the work for department catalogs is done by the department librarians, report that the department librarians catalog these same books for the general catalog as well.
In conclusion, let me repeat that if the cataloging is done by the department librarians, too much emphasis can not be laid upon the fact that it is absolutely essential to have all this work done under the supervision of the head cataloger, with the most careful revision. Otherwise there will be as many varieties of cataloging as there are department librarians.
In behalf of Dr. W. Dawson Johnston, of Columbia, Miss Isadore G. Mudge read the following