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 9 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 9 of 15)
Font size
QR-code for this ebook

virtues which in practice outweigh our theoretical objections.
These are its accessibility and the ease with which it may be
applied in whole or in part to collections of books and other ma-
terial of any size, and expanded as these collections grow. Even
if the order of the main classes and of some divisions is unac-
ceptable to many minds, there is in ordinary general library
practice no obvious necessity for an optimum order, although
such an order is in some way necessary to the ideal scheme,
which should be one of logical classes in logical relations. Un-
fortunately all order is conditioned when applied to books, by the
size of the books, the physical shape and division of a library
into departments and branches, which make it impossible to run
all books in one sequence of class-numbers whatever they may

After a lifelong use of the Decimal scheme, in which I have
read and listened to thousands of comments, I am convinced that
the oldest and most persistent one comes from the expert who
wants all material together on his subject, whatever its verifi-
able place; it is the most understandable one and the least rea-
sonable. The notation was and remains the most obvious reason
for the world-wide use it enjoys; that is, an international 'lan-
guage' understood by all nations. Some day the Decimal scheme
may disappear, as do all human efforts, but now we look forward
to the seventeenth edition. 9


1. Francis Bacon. The Works of Francis Bacon, ed. James Sped-
ding, Robert Leslie Ellis, and Douglas Denon Heath (Boston: Taggard
and Thompson, 1863), Vol. VIII, pp. 407-408.

2. W. C. Berwick Sayers, A Manual of Classification for Librarians
and Booksellers (3d ed., rev.; London: Grafton &Co., 1959), p. 88.

3. J. W. Garner, Political Science and Government (New York:
American Book Co., 1928), p. 52.

4. Henry Evelyn Bliss, The Organization of Knowledge in Libraries
(2d ed., rev.; New York: The H. W. Wilson Co., 1939), pp. 227-229.

5. Ibid., pp. 201-202.

6. " Criteria and General Procedures for Decimal Classification,
Edition 16. Restatement, December 5, 1955," Journal of Cataloging
and Classification, XH (April, 1956), 91-92.

7. Fremont Rider, Melvil Dewey (Chicago: American Library
Association, 1944), p. 37.


8. It should be noted that the Universal Decimal Classification
which dispenses with the final zeros has ten one- symbol forms and
100 two symbol forms.

9. Sayers, op. cit., pp. 125-126. In the last sentence the word
"seventeenth" has been substituted for the words "necessary six-
teenth" to update Sayers' statement.


Library of Congress

Classification for the

Academic Library

Irene M. Doyle
Assistant Librarian, University of Wisconsin

Gabriel Naude", as early as 1627, advised on the arrangement of
books in a library as follows:

The seventh point .... is that of the Order and Disposition
which Books ought to observe in a Library; .... for without
this, doubtless, all inquiring is to no purpose, and our labour
fruitless; seeing Books are for no other reason laid and re-
served in this place, but that they may be serviceable upon such
occasions as present themselves; Which thing it is notwithstand-
ing impossible to effect, unless they be ranged, and disposed ac-
cording to the variety of their subjects, or in such other sort, as
that they may easily be found, as soon as named. I affirm,
moreover, that without this Order and disposition, be the collec-
tion of Books whatever, were it of fifty thousand volumes, it
would no more merit the name of a Library, than an assembly
of thirty thousand men the name of an Army, unlesse they be
martially in their several quarters, under the conduct of their
Chiefs and Captains; or a vast heap of stones and materials,
that of a Palace or a house, till they be placed and put together
according to rule, to make a perfect and accomplished struc-
ture. *

Three hundred years later classification of books is still a live
subject, and largely for the same reason: "that they [the books] may
be serviceable upon such occasions as present themselves".

Though it is a live subject, and one of the most powerful tools in
libraries, it is surprising how little seems to have been published
considering its long history on book classification, how little has
been published on the Library of Congress classification, how very
little on L.C. classification in the academic library, and how very,
very little on "L.C. Classification in the Modern Academic Library."

I like to believe that the Library of Congress classification had its
beginning at the University of Wisconsin. Mr. J.C.M. Hanson, cata-
loguer at the University of Wisconsin 1893-1897, reported on its
beginning as follows:


During several informal discussions on classification and nota-
tion which I had about 1896 with Miss Olive Jones, librarian of
Ohio State University Library, the defects of both the D.C. and
the E.G. were gone over quite thoroly. We were both agreed
that a new classification with a notation representing a compro-
mise between the two would be desirable, especially for colleges
and university libraries. As for notation, we had in mind one or
two letters to indicate classes, subdivisions to be indicated by
numerals, either in regular or decimal sequence.

In 1897, therefore, when confronted by the necessity of sub-
mitting plans for a classification for the Library of Congress,
the rough sketches drawn up in 1895- 1896 were again brought
out and expanded. Fortunately, the Library of Congress had
secured, about this time, the services of Charles Martel, the
present chief of the Catalog Division. Mr. Martel was in sym-
pathy with the simplified notation suggested and the main work
of developing both notation and schedules was assigned to him.
It is mainly due to his indefatigable zeal and interest that the
classification developed as it did during the next fifteen years. 2

After leaving the Library of Congress, Mr. Hanson, at the Univer-
sity of Chicago, worked with the L.C. classification for many years.
Based on this additional experience with it, he wrote, "The advan-
tages have seemed to outweigh the disadvantages to such an extent
that personally I have no hesitation in recommending the adoption of
the L.C. classification for college libraries, large and small, as
against any other system in the field." 3

We now have L.C. classification at the University of Wisconsin.
Our own experience in changing to it is so recent, and it has been
such an absorbing experience, that perhaps I have failed to see the
woods for the trees in including in this paper such a full report of a
single institution. It may seem from these opening remarks that the
title of this paper should be: "The Library of Congress Classifica-
tion in One Academic Library."

In 1953, when classification became a very important topic with
most of us in the University of Wisconsin Library, we had just moved
from very crowded quarters in a building which we shared with the
State Historical Society to a new University Library building. Cata-
loguing was being done centrally for eleven department and school
libraries on the campus and for several reference collections within
the new library as well as for the general collection. There were in
the new building ten floors of stacks and, in the basement, stacks
providing compact storage for half a million books. The libraries
contained about 800,000 accessioned volumes, of which 50,000 were
uncatalogued. The cataloguing staff had not increased with the book
budget and preparations for moving to the new library (including a
series of projects which required almost the entire time of most of


the cataloguers and many of the clerical staff) had taken priority over
regular cataloguing hence the backlog. The 50,000 uncatalogued books
were not unavailable, however, for a multiple slip system was used so
that on the day a book reached the Catalog Department a card was
placed in the Public Catalog supplement and the book was passed on to
the Circulation Department. The volumes could circulate, and indeed
a great many of them did circulate.

For some time we had wanted to change to a different classification
system. Wisconsin was using the Cutter Expansive Classification. In
1893 Cutter's system was chosen over Dewey's because the notation
was more elastic and it seemed likely that Cutter's seventh classifi-
cation, then in the making, would profit from some of the errors of
Dewey, and that it would be more modern and more scientifically
developed. At the time the Cutter Expansive system was chosen for
Wisconsin, the first six classifications for small and medium librar-
ies were printed with an index covering all six classifications. The
seventh, planned for the large library, was not yet finished. Unfor-
tunately for Wisconsin it was never finished, and the Cutter Expansive
Classification, which continued to be used until 1954, was a combina-
tion of the 6th, with the index to the first six classifications, and part
of the 7th with an index to each class used. Some classes of the 7th
were printed too late for Wisconsin to adopt, or so it seemed to the
cataloguers, since they had already expanded parts of the 6th. Miss
Eliza Lamb, who worked with Mr. Hanson at Chicago and became
head of the Catalog Department at Wisconsin in 1930, described the
work of expanding the 6th classification as follows:

The librarian usually made an outline based on the best avail-
able authorities. This was referred to an expert in the field,
generally a member of the faculty. The results have not always
been continuously pleasing, even to those responsible. Such was
the case for the Botany scheme which was criticized adversely
by the very professor who had worked it out, he having forgotten
his connection with it. 4

Mr. Hanson remained at Wisconsin four years only, but within that
period he discovered that the classification was far from perfect:

Four years with the Expansive Classification convinced me that
no mistake had been made by the University of Wisconsin in
selecting the Expansive in preference to the D.C. classification.
However, the irregular sequence of letters, the preliminary
numbers for form classes, and other features, combined with
the slow progress in furnishing additional schedules, proved a
serious disadvantage. 5

Forty years after Cutter classification was adopted at Wisconsin,
Miss Lamb published an explanation and a defense of it:


The Expansive notation has been criticized as cumbersome,
but there has been little if any trouble .... Although the
younger generation has the reputation of being unfamiliar with
the alphabetical sequence, books are both found and shelved with
ease .... [It] has proved adequate to the required amplification
of passing years, avoiding the labor and expense of reclassifica-
tion which has been found necessary for many libraries adopting
other classifications before that of the Library of Congress was
available. 8

But twenty years later, in 1953, the number of volumes catalogued
per year had trebled, the staff had increased considerably, the revis-
ing time required taxed the abilities of the staff. There was little
time for the research, study, contemplation, and experimentation
necessary for the expansion of many of the classes, the placing of
new subjects, and the new approaches to old subjects. There was little
time for the instruction needed for the new and inexperienced cata-
loguers. For the most part they had not heard of the Cutter Expansive
Classification. To most of them "Cutter" meant only "Cutter author
tables." There were many inconveniences for example we had only
three copies of the classification. We spent years trying to locate
copies of the 6th and 7th classifications, finally finding someone who
had a small stock for sale. Negotiations were quickly underway but
when the signatures were received and checked against our copies we
could use less than one-fourth of the pages. The rest of it had to be
typed, the equivalent of two rather large volumes.

Not only in the Catalog Department but throughout the library there
was dissatisfaction with Cutter, particularly among the new staff
members. Faculty members who had studied in other research librar-
ies had become familiar with and recognized the merits of the Library
of Congress classification. New faculty members were completely
unfamiliar with Cutter. When at last we were settled in our new
building, it seemed a propitious time to change from Cutter, particu-
larly with 50,000 volumes awaiting cataloguing. We were not only
willing, but in fact eager, to give up Cutter in spite of its good, endur-
ing qualities.

There was one factor which deterred us from deciding immediately
in favor of the Library of Congress classification: the notation. Both
Cutter and L.C. consist of combinations of letters and figures. What
confusion there would be if the classification could not be recognized
as one or the other! Cutter class numbers, as assigned at the Uni-
versity of Wisconsin, consist of a combination of from one to five
letters. Wherever there is a geographical division, the letters are
followed by figures (used decimally). L.C. class numbers consist of
one or two letters only, followed by figures 1 to 9999 (used as integers).
In practice, with no exceptions, Cutter class numbers had been written
as one line (both letters and figures) except when there were more
than four letters. Only then were figures which followed the letters


written on a second line. 7 L.C. class numbers would not have more
than four letters in fact, not more than two. The figures which fol-
low the letters could always be written on the second line. Thus the
problem was resolved very easily! Sufficient differentiation was
provided to guard against confusion. To forestall any misunderstand-
ing that might possibly occur, and to help the Circulation Department,
we planned to stamp every University card, which included a Cutter
call number, with the word "Cutter" below the call number.

The second problem which we had to consider was: Assuming that
L.C. classification is the best existing classification for this library,
is it enough better than Cutter for us to give up Cutter for it? The
weaknesses and advantages of Cutter were well known to all of us.
The literature for L.C. classification was examined for criticisms
and reports of experience of other libraries. The familiar arguments
in its favor, occurring over and over again in the literature, are, in
part, as follows:


Particularity (topics are logical subdivisions of general sub-




Simplicity of notation

Individuality (made for L.C., for an actual collection of books, a
very large collection)


Each main schedule is preceded by a synopsis

There are tables which permit of very precise classifying,

particularly the "floating" geographical tables in Class H
There is an index to each schedule
Classifiers who made the classification and who revise it are

competent classifiers
It is a "close" classification

Since Library of Congress is behind it, there is reasonable
assurance that it will be kept up to date; also that the
schedules, printed as government documents, will be
reasonably priced

It undergoes continuous amplification in those fields in which
there is a concentration of material

Printed schedules are reprinted with additions and changes

Printed schedules are revised

Additions and changes are distributed quarterly

Class numbers are printed on L.C. cards in the majority of


Notation is elastic

Each class is printed as a separate book, Language and Litera-
ture (Class P) in several volumes

L.C. list of subject headings can be used as an index, in lieu of
an index

Not many general adverse criticisms were found. As Palmer has
said "The Library of Congress classification has been approached
with a certain measure of restraint."" Typical of the unfavorable
comments found are those from Mann:

No directions for its use
As yet, no complete index
Lack of mnemonic features
The magnitude of the scheme 9

and from Bliss:

Order of main classes unscientific and unecomonic

Five letters unused, but many important subjects without dis-
tinctive literal notation

Notation is of excessive length, in many cases far beyond the
economic limit

Too complicated and cumbersome!

Ranganathan, also, supplied an adverse criticism of L.C. In com-
menting on rigidity in the notation of some classifications, he said that
this rigidity can be broken by numbering the known specific subjects
by integers that are not consecutive, leaving unused integers between
them a "gap- notation." But the difficulty is that while some gaps
remain, others get filled up and it is in these filled-up gaps that more
and more new specific subjects must be inserted. He commended
Melvil Dewey for breaking this rigidity in gap-notation by using a
"pure decimal-fraction-notation," and continued:

It is a great pity that this master-stroke was lightheartedly ig-
nored and the rigid, primitive, gap-notation of integers was
adopted by the most influential scheme of classification in ex-
istence - the Library of Congress classification - which has all
the influence, resources, and backing of a mighty government.
The world is all the poorer for this. 11

Much earlier Hanson wrote as follows on this same matter of in-
tegers versus decimals in the L.C. notation:

Mr. Spofford, Librarian of Congress since the early sixties, and
assistant librarian after 1897, had personally supervised the de-
velopment of the Jeffersonian Classification, then in operation.
Mr. Spofford realized as fully as anyone the need of a new sys-
tem and was most generous and friendly in his attitude toward


our plans. Only on one point was he inexorable: there must be
no decimals.

This was one of the reasons why decimals were not more freely
used at the outset. Later on, while it would have been a rela-
tively simple matter to convert the numbers for subdivision in-
to decimals by writing them 0000-9999, the advantage of shorter
numbers for many thousands of books was thought to be of great-
er importance than the slight gain in symmetry and regularity,
resulting from the decimal arrangement. K

In connection with the Army Medical- Library of Congress discus-
sions concerning a proposed Army Medical Classification, Taube, in
1950, made this comment concerning the weaknesses of L.C.:

Even within the structure of the Library of Congress itself, this
conflict between general and special interests is a constant and
recurring phenomenon. Special consultants in various fields
have found that the library classification brought together unre-
lated materials and tore asunder materials which [naturally]
belonged together. Much more serious is the feeling of some of
the special divisions that the general cataloging and classifica-
tion system neglects and subverts their special interests. Many
of these divisions have set up special collections and special
bibliographic keys not provided by the general bibliographic
organization of the library. The degree of unification to be
achieved in the Library of Congress is a matter of internal ad-
ministrative policies, but the reality of the problem is additional
evidence that the specialist is not content with the by-product of
a universal organization .... What is required is the recogni-
tion that the Library of Congress system, for all its complexity
and detail, is not a tool for specialists but a general system for
the non-specialist's approach to knowledge as a whole. 13

Finally, in our consideration of the L.C. classification, we had to
ask ourselves, and answer, this question: "Does this classification
effectively meet the demands of the University of Wisconsin library?"

In fields where the instruction offered includes doctoral programs,
as in the arts, the collections have to be represented in considerable
depth and necessitate large volume holdings. We convinced ourselves
that the Library of Congress classification does provide a serviceable
arrangement for books in these fields where research needs necessi-
tate voluminous holdings. An examination of its quarterly "Additions
and Changes" convinced us that an effort was being made to keep the
classes represented in these disciplines up-to-date.

We made our decision in favor of changing to L.C. classification
knowing full well that it would not be entirely satisfactory in all sub-
jects, and that we were definitely influenced by the fact that we could


make certain advantageous applications of the system. Later we read
in Shera and Egan's The Classified Catalog: "The first principle to
be remembered in either choosing or constructing a classification is
that there is no single universal system that will serve all purposes
in all fields. The second principle is that there are no absolute values
in classification other than those of utility in the particular situa-
tion." 14

Various studies on Cutter versus L.C. classification, and on re-
cataloguing and cataloguing costs, were made for our Library Com-
mittee, of which the Librarian was a member. The Committee de-
cided against the proposal of the Library Administration to reclassify
the books already classified in Cutter, a project with which we had
hoped to combine some badly needed subject heading revision. It
approved the proposal to classify all new accessions (i.e., all titles
not previously catalogued) according to L.C. classification. The
President of the University agreed with the Committee that changing
over to the Library of Congress classification was desirable. Then,
on May 3, 1954 the Committee brought a proposal to the University

At Wisconsin, the University Faculty has a very important part in
academic affairs. It "has charge of all matters which concern more
than one college, school, or division, or are otherwise of general
University interest. . . . Subject to the laws and by-laws of the Regents,
under the laws of the State, the Faculty shall have general charge of
those questions of scholarship which pertain to more than one college,
school, or division; and they may make needful rules for the enforce-
ment of scholarship and discipline .... In case of conflict of juris-
diction between University Faculty and the faculty of any college,
school or division, the decision shall rest with the University
faculty." The Wisconsin Faculty is in charge of questions concern-
ing the educational interests or educational policies of the University;
requirements for admission to colleges, etc. and for graduation; rec-
ommendation of candidates for honorary degrees; regulation of social
affairs and athletic sports; investigation of cases of alleged infraction
of University rules; elections of Library, Nominations, University and
other committees, as well as many other matters.

The proposal brought to the Faculty by the Library Committee

The University Library Committee and the librarians of the
School and College libraries recommend that the Faculty ap-
prove the use of the Library of Congress classification system
in lieu of the Cutter Classification for books in the University
library system, except those in the Law Library. w

The Committee further called the Faculty's attention to several points:
1) Disadvantages of Cutter


2) Advantages of L.C. classification, especially the fact that

"classification number and subject entries on the printed
cards can be used almost automatically." Also that, in using
the classification number on the L.C. card there would be in
the U.W. library a saving of 42- cents per title in cataloguing

3) Reclassification was not feasible because of the cost alone, the

estimated cost being not less than half a million dollars.

4) Discontinuing the use of Cutter classification and adopting that

of the Library of Congress meant that, with a few possible
exceptions, most of the books classed in Cutter classifica-
tion would never be reclassified, but would be shelved as far
as possible on the same stack levels as the corresponding
classes in the Library of Congress classification.

5) Periodicals would be taken out of Cutter classification and all

periodicals shelved together alphabetically. Current serials,
except periodicals, would be gradually reclassified into L.C.

6) Books in the reference rooms would be reclassified into L.C.

7) The saving in cost of cataloguing would enable the library to

1 2 3 4 5 6 7 9 11 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 9 of 15)