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 2 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 2 of 15)
Font size
QR-code for this ebook

at classification, but my emphasis will fall on the last half of the
nineteenth century, that period in American library history when
many things were happening.

The complete history of classification in American libraries re-
mains to be written and our sources for even a summary of such a
history are all too few. There are some catalogues of colonial li-
braries, but these are seldom arranged by subjects. Of the twenty
catalogues which have survived, all but three are arranged alphabeti-
cally, either in a single list, or divided into three or four such lists
by size. As modern librarians we immediately concede the efficiency
of shelving folios, quartos, octavos, duodecimos, and smaller in
separate places but there is no evidence that books were shelved ac-
cording to size. It is assumed that the usual arrangement on shelves
was a fixed location. Probably a rough subject grouping was followed
when a collection was first arranged; with the addition of new titles,
or the movement of the library from one room to another, the subject
order was disturbed. Of the three surviving catalogues of the earliest
period, the Yale catalogue of 1743, the 1760 catalogue of James Logan's
library, and the partly classed 1764 catalogue of the Redwood Library,
only the Loganian catalogue reflected shelf arrangement. Thomas
Jefferson's books were placed on the shelves in the library at Monti-
cello in an order that matched the grouping of books in his catalogue.
However, the practice of Thomas Jefferson was not commonly followed


and as late as 1893 fixed location was a common arrangement of books,
although catalogues might be classified.

When librarians developed subject arrangements for books they
frequently borrowed ideas from the classifications of knowledge pre-
pared by philosophers and scholars of the past. All of us heard, early
in our courses in library school, that Dewey's classification scheme
was based on that of Francis Bacon, 1 and it has been pointed out that
Bacon's scheme was basically that of Aristotle. 2 Obviously then, to
study book classification one must begin with Aristotle and study the
various outlines of knowledge and the practical applications of these
outlines to the arrangement of books. During the years from Aristotle
to the period of colonial America hundreds of outlines were made, but
we will mention only two in addition to Bacon. Aristotle himself divid-
ed knowledge into three parts: practical or ethical; productive or
creative; and theoretical. Under practical he included the subjects of
economics, politics, and law. His productive or creative area included
poetry and the arts. His theoretical included mathematics, physics,
and theology.

Following Aristotle there were many philosophers who attempted to
equate the outline of knowledge with the various disciplines of educa-
tion. In Roman civilization the seven liberal arts were the preparatory
disciplines and the higher studies were theology, metaphysics, and
ethics. The seven liberal arts were divided into the trivium, consist-
ing of grammar, dialectics, and rhetoric, and the quadrivium, con-
sisting of arithmetic, music, geometry, and astronomy. Varro, who
lived from 116-27? B.C. made a classification of knowledge which
was merely a listing of the seven liberal arts with medicine and
architecture added. The writings of philosophers from Varro through
the middle ages frequently contained classifications of knowledge
which were nothing but this outline of studies, with higher studies of
medicine, jurisprudence, and theology added as they found their place
in the curricula of the universities.

The sixteenth century classification which Conrad Gesner used in
his Pandectarum sive Partitionum Universalum, the classified ar-
rangement of his Bibliotheca Univer salts, 3 was an expansion of the
schemes which represented the outlines of studies. An examination of
the twenty- one headings which Gesner used shows the familiar pattern
of trivium and quadrivium plus higher, studies and some rather mis-
cellaneous subjects. Gesner used the term Philosophy for the universe
of knowledge and thought of it as containing preparatory studies and
substantial studies. The preparatory studies were divided into nec-
essary and embellishing. The necessary included the seven liberal
arts, here expanded to nine by the addition of poetry to the trivium, or
conversational arts, and the use of both astronomy and astrology in the
quadrivium, or mathematical arts. These were the necessary prepar-
atory courses for advanced work. His embellishing courses have
puzzled classifiers by their variety: divination, geography, history, and
useful arts. His substantial sciences, or higher studies (natural


philosophy, metaphysics, moral philosophy, domestic philosophy, civil
arts, law, medicine, and Christian philosophy) are virtually the same
as the subjects which Aristotle listed as theoretical. Thus Gesner
combined the outline of Aristotle with the program of studies of the
university of his day and included other areas discussed in the books
that he examined for his universal bibliography.

This brings us to Francis Bacon who settled himself down in 1603
to lament the sad state into which learning had fallen in his time. He
divided knowledge into divine and human and in setting forth what men
should learn he outlined knowledge as he saw it, relating it to the three
parts of man's understanding: his memory, his imagination, and his
reason. Memory covered history, including natural, civil, ecclesias-
tical, and literary history. Imagination, represented by poesy, con-
tained lyric, epic, and dramatic poetry, and fables. Reason was con-
cerned with philosophy which included science, mathematics, theology,
anthropology, physiology, psychology, and sociology. The development
of philosophical classifications did not end with Francis Bacon, but
his was the last such scheme to have a noticeable effect on nineteenth
century book classification.

The other scheme which exerted influence on book classification in
the United States during the early period was a practical scheme used
by Paris booksellers to arrange titles in their sale catalogues. That this
scheme was also influenced by Bacon's outline of knowledge is clearly
evident to anyone who places the two outlines together. In its final
form the bookseller's scheme became the model for numerous clas-
sification schemes used in American libraries. Presumably based
on the work of Jean Gamier, a Jesuit, who prepared a catalogue of
Clermont College in Paris in 1678, 4 or that of Ismael Bouillaud, who
compiled a catalogue of the library of Jacques -Auguste de Thou in
1679, 5 and altered by Gabriel Martin, 6 and Guillaume de Bure, 7 it was
best known in the form used by Jacques -Charles Brunei^, in his Manuel
du Libairie et de I' Amateur de Livres. 8 The scheme contains five
main classes: theology, jurisprudence, science and arts, literature,
and history. With very little adjustment these five main classes can
be fitted into the inverted Baconian scheme used by later classifiers.
Certainly Brunet's scheme owes much to Bacon's outline of knowledge
but because it was specifically adapted to the needs of a book classifi-
cation, it is customary to think of Bacon as a philosophical scheme
and Brunet as a practical one.

This then was the state of classification when America was settled.
What was known in Europe found its way to America in due course.
Books containing the outlines of philosophic classifications and cata-
logues of books for sale showing various ways of arranging subjects
would have been available to colonial librarians. It is possible that
the outlines used in the three surviving catalogues are based on out-
lines used elsewhere. It has been suggested that the classification
used in the 1743 catalogue of Yale College, prepared by the Rector,
Thomas Clap, was copied from an outline of knowledge presented by


Samuel Johnson, President of Kings' College, in an essay published in
1731, 9 but Rector Clap implied that the outline was his own when he

I have here with considerable Labour and Pains prepared a
Catalogue of the Books in the Library under proper Heads so
that you may readily find any Book, upon any particular sub-
ject. 10

Clap's classes represented subject areas and there was little attempt
to make a systematic arrangement of the subjects although geography,
history, and biography did fall in successive classes. 11

The 1760 catalogue of James Logan's library 12 was divided into
twelve classes 13 with each class subdivided by size. Within the size
groups the arrangement was alphabetical. No systematic order is
evident in the arrangement of classes.

The 1764 catalogue of the Redwood Library 14 was in two parts. The
books purchased with Redwood's gift of money were arranged by size.
The subject arrangement of the folios, quartos, and duodecimos was
roughly alphabetical by author. However the octavos were divided into
eight classes, 15 which in turn seemed to fall into sub-classes, although
no headings were used to mark these divisions. There were little
groups of titles on such subjects as painting, military science, carpen-
try, agriculture, sports, and electricity under the class Arts, Liberal
and Mechanic. The only reason for dividing the octavos into classes
that has been suggested is the number of entries. 16 They fill approxi-
mately twelve of the twenty-two pages required for listing the Red-
wood gifts. The books "given by other gentlemen" were listed in four
groups: folios, quartos, octavos, etc., and pamphlets.

The story of classification must always be told in terms of the men
who produced the schemes and in discussing the development of clas-
sification in America we shall be concerned with librarians and other
scholars who were interested in achieving an orderly arrangement,
either systematic or practical, for the books in libraries. We shall
find some men dedicated to a single scheme, as Jefferson was dedi-
cated to the Baconian outline of knowledge. We shall find others, like
Jacob Schwartz, who could produce a number of quite different schemes.
Jefferson's first scheme was used in a catalogue of 1783; Schwarts
proposed his fifth scheme in 1893. This period of something more than
one hundred years is our immediate concern. The scores of schemes
that were produced during this period reflected the changing patterns
of knowledge and provided the foundation for the orderly arrangement
of books. In the early years the outstanding men of the profession
turned their talents to classification, but in our day, the last twenty or
thirty years, classification has been looked on as a necessary evil and
the talented members of the profession have often concentrated on other
aspects of librarianship.

The first classifier of note in the post-Revolutionary period was


Thomas Jefferson, third president of the United States, who shelved
the books in his library according to a Baconian order and provided
a catalogue in that order. Jefferson's book collecting activities had
begun when he was a young man. His first library, destroyed by fire
in 1770, was valued by the owner at 200. 17 Following the fire he
renewed his bookcollecting activities and by 1783 had assembled 2640
books which were arranged in his library according to a classification
scheme based on "An Outline of Human Knowledge" found in L'Ency-
clopedie ou Dictionnaire Raisonne des Sciences, des Arts et des
Metiers. Diderot and D'Alembert had altered the Baconian classes of
History, Poesy, and Philosophy to read History, Philosophy, and Im-
agination and had expanded Bacon's outline. Jefferson's scheme fol-
lowed the French version closely but he called his main divisions
History, Philosophy, and Fine Arts. 18 Jefferson explained to his friend
James Ogilvie 19 that books were arranged on the shelves divided into
twelve subjects 20 with the arrangement beginning "behind the partition
door leading out of the Bookroom into the Cabinet" and proceeding
from left to right. The catalogue follows this same order but is ex-
panded to forty-six chapters. The arrangement of the catalogue which
Jefferson provided to accompany his library after its purchase for the
Library of Congress differed only slightly from this classification of
1783; there was some reorganization in Fine Arts and the total number
of chapters was reduced from forty-six to forty-four. After the sale
of his books to the government, Jefferson, inveterate collector that he
was, began assembling another library. However, his third classifica-
tion scheme is found not in the catalogue of his third library but in an
acquisition list for the Library of the University of Virginia, Jeffer-
son prepared this catalogue of items to be purchased for the Univer-
sity in September 1824, arranging the titles in a classed order following
his usual Baconian form. The scheme was virtually the same as that
employed in his two earlier catalogues. More space was given to law.
and subjects which had had separate chapters in the 1815 classification
were combined. Of the influence of Jefferson's scheme on other li-
braries we have little evidence, but the 1815 catalogue of the Library
of Congress 22 was printed in an edition of 600 copies and these must
have been rather widely distributed. Moreover, the Jefferson classi-
fication remained in use at the Library of Congress until 1898.

The early Harvard catalogues, beginning in 1723, were arranged in
alphabetical order within size groups. From press marks which are
included in the first volume it appears that the books were originally
grouped on shelves in a rough subject order but we have no record of
a classification as such. However, the 1830 catalogue provided a sys-
tematic index which was designed to serve the purpose of a classed
catalog. This 1830 catalogue was prepared by the librarian Benjamin
Pierce. 23 It was a three volume work, containing the alphabetic file
under authors in two volumes and a systematic index in volume III.
The arrangement was that of Brunet, 24 with a sixth class for works
relating to America. The purpose of this class is not clear since


section IX of the class V, History, was assigned to American History.
There was some duplication of entries in the two places but not every-
thing in Works Relating to America appeared in section IX of class V.

Another librarian as devoted to the Baconian outline as Jefferson
was Edward William Johnston, who became the librarian of the College
of South Carolina in 1835 and a year later produced a classified cata-
logue which was strongly reminiscent of the Diderot-D'Alembert
adaptation of Bacon used by Jefferson. Johnston went from South
Carolina to the New York Mercantile Library Association and pro-
duced for that library a catalogue which used virtually the same clas-
sification as that adopted for his earlier catalogue.

In 1858 Johnston became librarian of the St. Louis Mercantile Li-
brary Association. This library, which had been founded in 1845, had
published its first catalogue, prepared by William P. Curtis, in 1850.
The arrangement under six headings was clearly the Harvard arrange-
ment of Brunet. 25 The compiler of the catalogue thus identified the

With respect to the arrangement of the Classification, it may be
well to state, that it is the same, with little exception, as that
which is used in the Catalogue of the Harvard University Library
of 1830; and, as to a subject upon which the rules are so arbi-
trary, and opinions so various, it is believed that this arrange-
ment is as perfect as any heretofore published, and it is hoped
that it will be as satisfactory to the mass of our readers as any
which could be adopted. 26

But the French scheme did not long survive the arrival of Johnston.
Johnston was too ardent a Baconian to accept it and shortly after his
arrival in 1858 embarked upon a new catalogue. It was a classified
catalogue because he was convinced that no other kind of catalogue
was satisfactory. In the introduction to the catalogue he said:

There is but one real method of arranging the contents of large
libraries; and this is the Systematic the regular classing of
books, each under the subject which it treats, so as to bring
together in one body all that the collection affords as to each
separate matter; while every matter, of course, finds its own
due place in a right intellectual arrangement of all human
knowledge. A mere alphabetical method (if indeed it can be
called such) can never, no matter how well executed, supply the
place of a true one. There is nothing to recommend it except
the facility of execution. For to make its (so-called) Classified
Index at all accomplish what it assumes to do, it would have to
be as large and minute as a regular systematic one, while total-
ly destitute of its advantage of rational arrangement. 27

The scheme, as usual in his catalogues, was a modification of Bacon,


following the Diderot-D'Alembert version. A third catalogue of the
library produced in 1874 28 was also Baconian but showed slight chang-
es from the earlier works. The Baconian scheme was retained until
about 1892 when Horace Kephart began reclassification to the sixth
expansion of Cutter.

The influence of the Harvard version of the Brunet classification is
reflected in the first catalogue of the San Francisco Mercantile Library.
The catalogue, prepared by Horace Davis in 1854, was quite similar to
the catalogue which was prepared for the St. Louis Mercantile Asso-
ciation Library in 1850. The 1861 catalogue 29 of the San Francisco
Mercantile Library made some changes but the Brunet scheme was
still clearly evident. 30 This remained the scheme in use until about
1891 when the library was reclassified using the Decimal Classification
of Dewey.

The scheme 31 used for arranging the Boston Public Library, al-
though not important in itself, is mentioned for two reasons. It is
often referred to as an early decimal scheme and it was prepared not
by a librarian, but by a member of the library board of trustees.
The scheme, which was decimal only in the method of placing books
in a room which extended three floors in height and had ten alcoves of
ten ranges each, with ten shelves to a range, on each floor, was a
fixed location arrangement. Alcoves were assigned to the various
subjects and ranges were assigned to the subdivisions. The call num-
ber showed alcove number, range number, shelf number, and number
of the work, not the volume, on a shelf. This ingenious scheme was
devised by Nathaniel Shurtleff, but the details of putting it into opera-
tion were carried out by C.C. Jewett. It may have had some vogue in
Massachusetts libraries. The public library of Haverhill, Massachu-
setts, reported in 1893 that the scheme had been in use there for ten
years. At the same time the public library of Cambridge, Massachu-
setts, reported the use of the Cutter classification, with Shurtleff
notation, and gave the time it had been in use as eighteen years.

Jacob Schwartz, librarian of the Apprentices' Library in New York,
was one of the most versatile of classification makers and one of the
most ardent and vocal speakers on the subject of classification at
meetings of librarians. Schwartz was concerned with the practical
arrangement of books on shelves. He established his divisions into
classes on the basis of the number of books in each division rather
than on the importance of the division as a field of knowledge. His
notation was designed to arrange books by subject, by size, and alpha-
betically by author. The three main classes (Cosmology, or Natural
Science; Anthropology, or Human Science; and Theology, or Divine
Science) were divided into twenty-five general classes with nine sub-
classes for each. He began to apply the system to books purchased for
the New York Apprentices Library in 1871, and in 1874 printed a cata-
logue of the library 32 with a classified index.

In 1879 Mr. Schwartz produced a mnemonic system of classification,
consisting of an alphabetico-subject arrangement of classes. In 1882


he presented a second alphabetic scheme which used a quite different
set of terms for the classes. Both of these schemes were accompan-
ied by elaborate author marks that separated books into four sizes
ranging from duodecimo to folio, and arranged them alphabetically
within the size groups. In 1885 he produced another scheme contain-
ing ten main classes. Except for the general works the classes were
in alphabetic order. The three digit notation which accompanied this
scheme was not decimal in nature. In answer to Kephart's question-
naire of 1893 he produced a variation of this scheme. There were still
ten classes but the alphabetic order was abandoned. The last four
schemes produced by Schwartz were not applied to the Apprentices'
Library, nor is there any record that they were adopted elsewhere but
libraries which were developing local schemes may have used some
of the ideas found in the many articles which Schwartz wrote for li-
brary journals. St. Benedict's College at Atchison, Kansas, used, until
1926, a scheme consisting of forty broad classes, each class divided
into five size groups: folio, quarto, octavo, duodecimo, and sexto-
decimo. Some libraries still use Schwartz book notation.

At about the time that Jacob Schwartz was beginning his work at
the Apprentices' Library in New York a scholar and philosopher was
turning his attention to classification in the distant city of St. Louis.
William Torrey Harris had been born in Connecticut and educated at
Andover and Yale. He went to St. Louis in 1858, as a teacher in the
public schools. In 1866 he was elected assistant superintendent of
schools and two years later he became superintendent of schools. He
left St. Louis in 1880 to assist in founding a school of philosophy in
Concord, Massachusetts. In 1889 he became United States Commission-
er of Education.

As superintendent of schools, Harris was ex officio one of the
"Managers" of the library. The public library, which was maintained
by the school district of St. Louis, was established in 1865. In 1870
Harris published in the Journal of Speculative Philosophy a scheme
for classing books in libraries. 3a In the same year the scheme was
applied to the St. Louis Public Library.

In explanation of the scheme Harris wrote:

It uses Bacon's fundamental distinction (developed in the De
Augmentis Scientarium, Book II, chap. I) of the different facul-
ties of the soul into MEMORY, IMAGINATION, and REASON,
from which proceed the three great departments of human
learning, to wit: History, Poetry, and Philosophy. Without
particularly intending to classify books as such Lord Bacon at-
tempted to map out "Human learning" as he called it, and show
its unity and the principle of development in the same. But his
deep glance seized the formative idea which distinguished dif-
ferent species of books. 34

Harris made no claim of originality in using Bacon's outline. He


had examined other Baconian schemes and had developed an organi-
zation somewhat different from theirs. He acknowledged his indebted-
ness to Johnston in the following words:

I should not omit this opportunity to refer to the Catalogue of that
excellent collection, the St. Louis Mercantile Library, which i's
based on the Baconian system. In fact, it was the eminent, prac-
tical success of that system of classification considering both
its usefulness to the reader and the convenience to the librarians
that led to this attempt at a Classified Catalogue of the Public
School Library. This form of the Baconian system adopted in
the Catalogue of the Mercantile Library is substantially that of
D'Alembert (Encyclopedic Methodique 1787); but it has numer-
ous modifications introduced by the fertile mind of the librarian,
Edward Wm. Johnston, Esq ....

Many of the subdivisions in the present Catalogue have been
borrowed from this system, but his [Johnston's] system lacks
proper subordination, and there is consequently much confusion
in the second department, or "Philosophy."'

There has been considerable discussion of the influence of the
Harris scheme but there is little that can be proved. The Peoria
[Illinois] Mercantile Library soon adopted the scheme as the best that
was available at that time. As the Board of Directors explained:

To arrange such a system of classification, however, one that
shall be complete and exhaustive, is an effort of the highest
philosophy, for it implies no less than a classification of all

2 4 5 6 7 8 9 10 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 2 of 15)