University of Illinois (Urbana-Champaign campus)..

The role of classification in the modern American library : papers presented at an institute conducted by the University of Illinois Graduate School of Library Science, November 1-4, 1959 (Volume 1959) online

. (page 10 of 15)
Online LibraryUniversity of Illinois (Urbana-Champaign campus).The role of classification in the modern American library : papers presented at an institute conducted by the University of Illinois Graduate School of Library Science, November 1-4, 1959 (Volume 1959) → online text (page 10 of 15)
Font size
QR-code for this ebook

keep up-to-date in the cataloguing of new acquisitions and
enable it to eliminate the 47,000-volume arrearage within
5 years.

The Faculty voted favorably on the adoption of the Library of Congress
classification system.

For the next three months while most Catalog Department staff
members were supervising some parts of the reclassifying of all
periodicals in the stacks into one alphabetical group, or completing
other projects, they were, in addition, studying the L.C. classification
system since none of our cataloguers had had experience with it. We
held a series of meetings with them in small groups for examination,
explanation, and discussions of the schedules. Each cataloguer accu-
mulated a file of L.C. proofslips in the class in which he was to work
and studied the class numbers as assigned by the Library of Congress.
In September 1954 we were ready to begin cataloguing again.

All book cataloguers but two were assigned to cataloguing the ap-
proximately 25,000 books for which L.C. cards with call numbers
were on hand. Catalogued next were the some 6000 books for which
there were L.C. cards without call numbers or with analytic call
numbers. Finally, most of the cataloguers were transferred to orig-
inal cataloguing, and the cataloguing of books with L.C. cards was
continued by a very small staff. By this time, there had been built up
a sizable shelf list which helped considerably in the classification of
books without L.C. cards.

Since we planned from the first to take full advantage of the


classifying done by the Library of Congress, we accept the class
numbers on the cards unless in the process of checking them with the
schedules we find them in error, as an error in printing, or an earlier
class number which L.C. has later revised or expanded. We have
carried over from work with Cutter no notions which we may have had
of the best placement of material (as subject bibliography which we
had thought previously must be with the subject). We do not dwell on
L.C. classification's weaknesses, which for our purposes are minor.
We are thankful for its many good features.

Reclassification of reference collections will soon be completed.
We do not worry about the number of books which will remain in
Cutter. In the basement of our library are space and compact shelv-
ing for half a million volumes. Eventually, perhaps in 10 years or so,
the "Cutter books" will be moved to the basement, except the "live"
titles which may then be reclassified. Perhaps we can in the mean-
time reclassify each "Cutter book" that a borrower returns, but
there is no plan for that at present. The 50,000-volume backlog has
now (1959) been reduced to 11,000 volumes, a great part of which is
in Hebrew, Russian, other non-roman alphabets and ideographic lan-

Five years ago we classified our first book by the Library of
Congress system. Now, five years and 130,000 titles later, we may
well ask: Should we have changed classification? Has the change to
L.C. classification in our particular library been a satisfactory one?
Since I did not think that my personal feeling in the matter would
make for a sufficiently impersonal answer (and furthermore one
should protect oneself against being accused of institutional chauvin-
ism), I questioned several cataloguers, reference librarians, depart-
ment and branch librarians and, through these librarians, faculty who
use the libraries a great deal.

Departmental librarians who responded were mostly from science
and technology libraries. They believe that L.C. classification is at
least as good as Cutter, though some miss the mnemonic feature of
Cutter. L.C. needs further subdivision in some parts of Science, and
scatters books on closely related subjects, notably when Chemistry
overlaps Physics or Medicine. It is better correlated than is Cutter
to the sequence of study in Biology. It does not result in long clumsy
numbers as in Dewey, does not break logical sets as in Bliss. It is
more flexible, in the opinion of one librarian, than Cutter, Dewey, or
Bliss. Several think that the faculties are not classification- con-
scious; one believes that faculty members consider it a good system
if it locates a book as quickly as possible "with little fussing." One
librarian believes that it is not the kind of classification that counts
but the consistent use of it that makes its application successful.

Representative comments from cataloguers, reference librarians,
and faculty members follow:


(From the cataloguer's point of view)

It is unfortunate that "the better aspects of" the Library of
Congress classification are so intangible, while the limitations
are so obvious. However, the system is well suited to the or-
ganization of knowledge as practiced by the library classifier.
Although comprehensive, it is well indexed through the L.C.
subject heading list and the L.C. subject catalog. The ex-
amples given in the latter also aid the classifier in identifying
the particular aspect of the subject for which he seeks a class
number. The multiple volumes of the classification, though
intimidating to the beginner, are so organized that once the
overall pattern is comprehended, the specifics fall easily into
place. Such divisions as biography, study and teaching, etc.,
within any class come generally in the same progression,
whether the subject be comprehensive or minute. The detailed
expansions from general to specific allow for the ready iden-
tification of subjects with class numbers. The tables demand
a certain alertness, but repeated use of them soon brings

The principal disadvantage in the use of the classification is in
keeping the schedules up to date with regard to new numbers
and new expansions. The indexes, lists of individual authors in
literature and the personal bibliography numbers are impos-
sible to keep up to date, given the present format of the volumes.
With some volumes (BL-BX, for instance) it has been so long
since the last printed revision that there seem to be more cor-
rections and additions than original entries. Also the lack of a
comprehensive guide to the use of the system creates great
problems even for the more experienced classifier. These
limitations however are mere annoyances when balanced against
the general applicability of the Library of Congress classifica-
tion scheme to a large collection.

(From the reference librarian's point of view)

In many ways the preciseness of L.C. is not so useful to the
reference librarian as the broadness of Cutter. For example,
Cutter classifies French language, literature, literary bio-
graphy, and the apposite bibliography more or less together,
where they are easy to locate and to work with. The specificity
of L.C., though, scatters materials instead of bringing them to-

A principal criticism of L.C. by reference librarians and schol-
ars is that it separates bibliographies from pertinent subject


L.C. in general seems to be less popular with the faculty
library users than Cutter. Our first comment may have bear-
ing here, but the unpopularity in some measure can be dis-
countedin large part a question of getting used to a new

Many L.C. schedules are not kept up to date. Current L.C.
practices as well as specific numbers not known must be in-
ferred from new card numbers. In this respect, there is need
for a manual on the L.C. classification.

Though the narrowness of L.C. is in some ways a handicap (see
above) it is easier to pinpoint items, the classification adapts it-
self easily to new subjects and topics, and it is convenient to use
the L.C. list of subject headings as an index to the classification
and the materials classified.

Any complaints against L.C. are purely academic a matter of
simple economics, as long as libraries can cut cataloguing
costs by accepting numbers on L.C. cards.

(From the faculty point of view)

My general feeling might be that ANY system well administered
would be satisfactory .... and I find both the L.C. and Cutter
system quite satisfactory for my own purposes .... I prefer
the Cutter probably because the greatest part of our collection
is still classified in this way .... I have always felt that the
L.C. system tried to compress things too much with a relatively
small number of over-all divisions.

I would say that, as compared with Cutter, the new system is
superior in that the books now seem more carefully categor-
ized and more logically arranged on the shelves. I have not
noted, in the Library of Congress system, any cases where two
books of very similar subject and comprehensiveness were
widely separated on the shelves, a situation which too frequently
occurs in Cutter.

I suppose the best argument for the L.C. system is the conveni-
ence of using L.C. cards and in having eventually a more or
less uniform system throughout the country.

It seems to me that the problem of satisfactory classifications
lies more within the jurisdiction of the librarians who make and
work with these classifications and not with the users of the
library. Because regardless of the faults in the classification,
the value the user gets out of the system will probably depend
most upon the efficiency with which the card catalog is main-
tained. That is even if the system is bad, but if the card catalog
permits a person to find a book within a very short time, that is
all that really matters.


These somewhat extensive local comments will be recognizable, no
doubt, in their general tenor, to many librarians who have served in
academic libraries in which a change to L.C. has been made. The
Wisconsin change did not involve reclassification of past acquisitions.
A recent change involving complete reclassification of the entire col-
lection has been undertaken at Michigan State University at East
Lansing. There the change is from Dewey to L.C. It is organized as
a ten-year operation financed by a special appropriation of $250,000
which provides two full-time professional reclassifiers, four full-
time clerical workers and student help. The work was begun in the
late fall of 1955. Current acquisitions were put in L.C. very soon
after the initial authorization. There are six open-shelf divisional
reading rooms at Michigan State and the reclassification is being done
in one room at a time, current acquisitions in L.C. being placed at t'^e
beginning of the shelf ranges in each room.

So far, we have dealt with the large university library. Is the
small academic library using L.C. and how satisfactorily? A study
on "Classification in College and University Libraries" by Eaton was
reported in College and Research Libraries for April 1955. " Its
purpose was primarily to collect accurate figures on the number of
institutions using the classification schemes commonly taught in
library schools. Of the 744 college and university libraries replying
to Miss Eaton's questionnaire, fifty-four libraries of 100,000-or-less
volumes were using L.C. classification. Of these, ten would prefer
Dewey. Four hundred and eighty-seven libraries of 100,000-or-less
volumes were using Dewey, and seventy of them would prefer L.C.
Surely Dewey has control of classification here.

It seems to be an accepted fact, in the literature, that L.C. classi-
fication is not for the small library. "Few small libraries have ever
adopted L.C."; "Since it lacks general numbers for many areas, it
will never serve very well in the small library needing broad classi-
fication"; "Does not lend itself easily to abridgment for use in librar-
ies with small collections"; "The large library will probably find the
L.C. scheme more satisfactory than will the small library."

In order to find out how some smaller libraries which had used
L.C. classification for some time were faring classification-wise, a
brief questionnaire was addressed to college and university libraries
listed as using L.C. classification in the 1936/37 and 1940/41 annual
reports of the Librarian of Congress, but limited to those libraries
which, in the latest American Library Directory, showed holdings of
100,000-or-less volumes. 18 In all, twenty-nine questionnaires were
sent. Replies were received from twenty-four libraries. Of the
twenty-four who answered, four reported that they used Decimal
Classification and one librarian reported that L.C. had been used but
that his predecessor had changed to Dewey in his small combined
college-high school library.

In answer to the question, "Are schedules followed as printed?"
all answered in the affirmative except one library which used the


term, "Mostly." The question, "Have schedules been abridged?"
was answered "No" by all except one library which said, ". . . . in
Cutter numbers." To "Do you make alterations in parts of classifi-
cation?" there were answers of "No," "Rarely," "Seldom," "Few,"
"Once in a while." One library has an expansion for Lutheran church
material, and another uses Lynn classification for Catholic theology;
another classes fiction prior to 1930 in PR, PS, etc. rather than in
PZ with the idea that older fiction if worth keeping should be in Liter-
ature, and if not worth keeping should be discarded. At some future
time the "1930" line will be moved up so that PZ will always be
fairly recent fiction. Several expressed dissatisfaction with L.C.'s
classification of biography by subject.

All but one had student access to the stacks. All believed that the
faculty and students liked the L.C. classification, one adding "when
they become familiar with it." One reported that the faculty was be-
coming interested in classifications developed by professional socie-
ties for special fields and also in the Universal Decimal Classification.

The following are selections from comments which were made by
the librarians who replied:

No one has ever mentioned another classification. Our Refer-
ence library in the city uses L.C. and the Public library uses
Dewey. Our clientele are familiar with both schemes and use
them quite casually.

Our experience has been that students almost never question or
comment on the classification, even though they have been ac-
customed to Dewey in high school libraries. Once they learn to
use the card catalog, they accept the number as a matter of

Personally, I like it better than Dewey and I have done classify-
ing in both systems and worked as reference assistant with both

I prefer it even for the medium sized college library in spite of
the fact that Dewey is easier to keep in mind. Our staff is
pretty generally glad we have L.C.

It has seemed to me that 'size* of the collection is not so great
a factor in deciding whether or not to use L.C. I think it is a
matter of how detailed a classification is needed.

Dewey is simpler and more economical to use in small librar-
ies than L.C., but L.C. works just fine in small libraries too.

From my point of view the L.C. system is very satisfactory,
and our faculty members and students have not complained at
all, except the Freshmen who were used to Dewey. Basically,
I believe one of the most important advantages for the small
academic library to adopt the L.C. system lies in the fact that


L.C. gives suggested call numbers in most instances. This
makes it possible for small libraries to save money on staff.
For example, we have only one professional cataloger, but she
can do about 5,000 volumes a year without any help, and this is
done on top of her teaching duties which amount to three hours
a week, in a subject field.

If there is anything we feel badly about, it is the fact that some
one in times past decided to alter the Cutter numbers given by
L.C. in order to make them shorter. Now that we have more
books than anticipated back in the 1930's, we find ourselves in

I firmly believe that any small library wishing to adopt the L.C.
system would gain more than it could lose, provided it does not
proceed to change the L.C. classification.

As the writing of a paper progresses, many by-paths open up be-
fore one and many vistas beckon to lure one away from the main sub-
ject. There is one which I would have liked to explore, and that is:
With such a close classification as L.C., do we need full subject cat-
alogues? Can we defend this duplication of effort?

This paper was opened with a quotation from Gabriel Naud and I
am bringing it to a conclusion with another quotation from him, made
in 1627, in which he gives an opinion on this very matter:

After all which, it shall be very requisite to make two Cata-
logues of all the Books contained in the Library, in one whereof
they should be so precisely dispos'd according to their several
Matters and Faculties, that one may see & know in the twinkling
of an eye, all the Authors which do meet there upon the first
subject that shall come into ones head; and in the other, they
should be faithfully ranged and reduced under an Alphabetical
order of their Authours, as well to avoid the buying of them
twice, as to know what are wanting, and satisfie a number of
persons that are sometimes curious of reading all the works of
certain Authours in particular. 19


1. Gabriel Naude". Instructions Concerning Erecting of a Library
trans. John Evelyn (London: For G. Belle & T. Collins & J. Crook,
1661), pp. 74-75.

2. J.C.M. Hanson, "Library of Congress Classification for Col-
lege Libraries," Library Journal, XLVIII (February 18, 1921), pp.


3. Ibid., p. 153.

4. Eliza Lamb, "The Expansive Classification in Use," Library
Quarterly, IV (April, 1934), pp. 268-269.

5. Hanson, op. cit., p. 151.

6. Lamb, op. cit., pp. 268-269.

7. Example: HF39, Economic History of France, but

HECOA, Coal trade in Australia

8. B. I. Palmer, "Classification," Library Trends, II (October,
1953), pp. 236-248.

9. Margaret Mann, Introduction to Cataloging and the Classifica-
tion of Books (2d ed.; Chicago: American Library Association, 1943),
p. 83.

10. H. E. Bliss, The Organization of Knowledge in Libraries (New
York: H.W. Wilson Co., 1934), pp. 242-278.

11. S. R. Ranganathan, "Colon Classification and its Approach to
Documentation," Bibliographic Organization; ed. J.H. Shera and Mar-
garet E. Egan (Chicago: University of Chicago Press, 1951), p. 63.

12. Hanson, op. cit., p. 152.

13. Mortimer Taube, "Functional Approach to Bibliographic
Organization," Bibliographic Organization; ed. J.H. Shera and Mar-
garet E. Egan (Chicago: University of Chicago Press, 1951), p. 63.

14. J.H. Shera and Margaret E. Egan, The Classified Catalog
(Chicago: American Library Association, 1956), p. 24.

15. Wisconsin, University, Regents, "Laws of the Regents, 1914,
Chap, in, sec. 2, Bd. min., 6/20/16; Bd. min., 6/14/51" Laws and
Regulations Governing the University of Wisconsin. (September,
1951), pp. lv-2.

16. Wisconsin, University, Faculty, "Adoption of the Library of
Congress Classification system for Books in the University Library
System," Calendar, University Faculty Meeting, Monday, May 3,
1954, Document 1133 (Mimeographed).

17. Thelma Eaton, "Classification in College and University
Libraries," College and Research Libraries, XVI (April, 1955), pp.


18. Questionnaire sent to librarians of twenty-nine academic
libraries containing less than 100,000 volumes.


Name of library Number of volumes

How long has L.C. classification been used in this library?

Are schedules followed as printed?

Have schedules been abridged?

Do you make alterations in parts of classification?

Approximate size of collection classified in L.C.

Do students have access to stacks?

Does faculty have access to stacks ?

Do you believe that the following are satisfied with the classifica-



Library Staff

If they would prefer another classification, what one?

Further comments: (use verso of this sheet, if necessary)

19. Naude", op. cit., pp. 90-91.


One Mathematician Looks at

the Classification of


Robert G. Bartle

Associate Professor of Mathematics
University of Illinois

At the very outset I want to warn you that I am here in the role of
a mathematician who is interested in books and that I am entirely in-
nocent of library procedures and terminology, the theory of classifi-
cation, or the actual classification of anything but mathematical books.
I am not sure that words of wisdom have ever come from the mouths
of infants, but I am very strongly relying on that possibility. If this
hope proves wrong, then I can only apologize and point out that every
carnival should have a freak show and that I am only trying to do my

I am also aware of the extensive use of the vertical pronoun in my
talk, but I know of no alternative. I have spoken with a number of my
mathematical colleagues, but I do not pretend that my remarks are
really an accurate statement of the ideas of the mathematical com-

I shall be more than satisfied if I can act as a gadfly and provoke
some discussion. Many of my remarks are very frankly critical.
However, it is my earnest hope that they will not be taken offensively,
but that they might be turned to constructive use. If this can be done,
I shall be most pleased.


I should like to make a few remarks about mathematical termino-
logy which may distinguish mathematics from certain other fields.
Unlike chemistry which has a large supply of artificial technical
words which is constantly augmented, the tendency in mathematics is
to use homely words and to attach a new, technical meaning to these
words. Thus, for example, the nouns "group," ''ring," ''ideal,"
"lattice," "field," "neighborhood," "measure," "sheaf," "fiber
bundle," "place," etc., denote definite mathematical concepts whose
exact meaning cannot be guessed in fact, it is not even apparent in
which area of study these words are used. Also, modifiers such as
"regular," "normal," "absolute," "proper," "analytic," etc., are
used in a quite technical fashion. (The meaning is not necessarily
unique, however the words "regular" and "normal" have well over


a dozen totally different usages.) This is not to say that we mathe-
maticians do not have our words such as "homeomorphic," "iso-
metric," "automorphism," "eigenf unction," but I can think of nothing
in mathematics as dramatic as chemistry's word "dichloro-diphenyl-
trichloro-ethane" (DDT).

In addition to attaching technical meanings to old, familiar terms,
mathematicians often take over proper names; thus we get "Euclidean
geometry," "Riemannian geometry," "Riemann zeta function,"
"Riemann integral," "Riemann surface," "Hilbert space," "Fourier
series," "Chebyshev polynomial," and many others.

I am sure that these terminological practices (which occur in
mathematical writings in every European language), complicate the
job of the non-specialist classifier. Nevertheless, this practice is
not likely to be discontinued if the alternative to using the word
"measure" is to employ the far more cumbersome equivalent phrase
"non-negative extended real-valued countably additive set function
which vanishes at the empty set," which, in addition to being unwieldy,
itself employs many technical terms. As a matter of fact, the word
"measure" is fairly descriptive if one realizes that it is intended to
generalize the notion of length, area, volume, mass in short, the
measure. However, it is easy to see that a book entitled Measure
Theory, by Paul R. Halmos, will cause difficulty to the average li-
brarian. I should like to take an imaginary trip with this book as it
leaves the publisher, in 1950, and finds its way into the mathematics
library. There is a joke among mathematicians that this book was
actually classified, in some unnamed library, with the books on
carpentry. 1 A more likely classification would be to put it in 510 V2,
since it is published by Van Nostrand in their "University Series in
Higher Mathematics." This number does put it in the mathematics
bracket, so certainly is to be preferred to carpentry, but I do not
feel that it is a very good classification as I hope to make clear later.
Let us suppose, then, that the book has managed to elude the Serials
Department (which might be possible since the so-called series to
which it belongs is not numbered and this is only the second one of
the Van Nostrand to be bound in blue.) What is in store for the book
now? The answer might be that the Library of Congress card is ob-
tained and the book is classified 513.83, since the book is declared to

1 2 3 4 5 6 7 8 10 12 13 14 15

Online LibraryUniversity of Illinois (Urbana-Champaign campus).The role of classification in the modern American library : papers presented at an institute conducted by the University of Illinois Graduate School of Library Science, November 1-4, 1959 (Volume 1959) → online text (page 10 of 15)