College and Research Libraries


By P A U L S H A N E R D U N K I N 

Classification and the Scholar 

Mr. Dunkin is senior cataloger, Folger 

Shakespeare Library, Washington, D.C. 

LI B R A R I A N S have written much on classification; the scholar—the layman 
in the stacks—is yet to be heard from. 

T h e librarian, M r . Bliss no less than 

M r . Dewey, is essentially an ambitious 

theorist concerned with working out an 

elaborate general scheme for the organiza-

tion of all knowledge. T h i s because he 

assumes that a book is most easily found 

when its place on the shelf is determined 

by the logical relation of its subject matter 

to all subject matter. Like a modern 

Alexander he dreams of the fusion of all 

culture into an harmonious whole. But 

the scholar of today can no longer take all 

knowledge for his province. He is a busy 

man in a tiny corner, a princeling in the 

librarian's far-flung empire. L o f t y talk of 

sweeping organization for the accomplish-

ment of ultimate universal ends leaves him 

cold; he hopes only that the talkers will 

not prevent him from working efficiently 

in his own little world. 

T h e scholar's stack permit is useful to 

him only if the librarian's classification 

shelves the books that he needs for his 

research where he can easily find them. 

T h e moment he must walk out to the card 

catalog in order to locate a book whose 

general subject he knows, the classification 

scheme has failed him. It would be more 

practical for him in that case to have the 

books on the shelf in the simple order of 

their accession. T h e catalog card could 

give him an accession number as readily 

as a classification number, and once he had 

found the book he would know the exact 

place on the shelf where it would be for-

ever. Moreover, the money now spent 

on classification could buy him more books. 

T h e traditional library classification is a 

philosophical scheme; what the scholar 

wants is functional classification—an ar-

rangement of books according to needs as 

those needs appear as the result of experi-

ence. L a w , business, and even library 

administration are based on constant ex-

periment, but library classification is still 

based on a priori reasoning. Y e t func-

tional classification is not a particularly 

new notion. T h e bookseller has always 

shelved his books not by logic but accord-

ing to prospective buyers' wants. A few 

years ago Grace O . Kelley clearly demon-

strated that traditional shelving does not 

serve library patrons nearly so efficiently 

as has always been assumed,1 and in the 

Detroit Public Library Ralph Ulveling 

has for some time been urging that books 

on open shelves be arranged according to 

readers' interest.2 

T o the scholar good classification means 

only two things: ( i ) One section of the 

general stacks must be given over entirely 

to his books, and ( 2 ) T h e books in that 

1 K e l l e y , Grace O. The Classification of Books; An 
Inquiry iinto Its Usefulness to the Reader. 1937. 

2 " S h o u l d W e C l a s s i f y and Catalog from the Read-
er's Point of V i e w ? " A.L.A. Bulletin 32:55, Janu-
ary 1938. 

SEPTEMBER, 1942 
2 77 



section must be arranged in an intelligible 

(not necessarily logical) sequence. 

T h e implications of these general ob-

servations will become more apparent with 

a casual examination of the "language" 

and "literature" groups in classification 

schemes. 

Both the Dewey Decimal Classification 

and the Library of Congress Classification 

have held to the peculiar but common con-

ceit of a series of separate static languages 

and a series of separate static literatures. 

There are, indeed, such things as "the 

science of philology" and "great litera-

t u r e ; " and the scholar does edit a text 

and interpret a literature. But if he is to 

do these jobs well be must know the folk 

who spoke that language and wrote that 

literature, and he must exhaust every pos-

sible source in his endeavor to reconstruct 

their lives and their culture, their prob-

lems and their mistakes. 

Inscriptions 

A book of inscriptions, for instance, is 

an innocent enough affair. Normally a 

classifier would tuck it into a special nook 

where none but a linguist might find it 

with ease. But what of the classical 

scholar who traces the increase of prices 

and decrease of wages in Greece of the 

third century before Christ by means of a 

series of inscriptions? 

Erasmus' editor must know something 

of Luther and the Reformation and the 

Peasants' Revolt. Understanding of Ver-

gil's Aeneid or the poetry of d'Annunzio 

is not complete without some consideration 

of Augustus and Mussolini. 

More's Utopia, no less than Swift's 

Gulliver, is a product of its author's 

milieu and may properly be studied only 

with that entire milieu in mind—histori-

cal, social, and economic. T h e student of 

modern literature must be acquainted with 

M a r x and Darwin and Freud as well as 

Sinclair Lewis and Kipling and George 

Bernard Shaw. 

A l l this, of course, is obvious. 

But classifiers have ignored it. Inscrip-

tions are language, while prices and wages 

are economics; Erasmus is literature, but 

Luther is religion; Vergil and d'Annunzio 

are poetry, but Augustus and Mussolini 

are history. 

T h i s is not just captious criticism. T h e 

following table shows how D . C . and L . C . 

treat certain subjects of interest to the 

scholar in a specific field, the study of the 

classics: 

Language 
and 

Literature 

Epigraphy 
and 

Palaeography 

Early Christian 
Literature 

Greek and Latin Subjects 

L.C. 

Epigraphy: 

Palaeography; 

P — P A 

C N 

Z i 14 

BR60-67 

D.C. 

Language: 470-480 

Literature: 

Latin: 

Greek: 

870-880 

471.7 

481.7 

281.1 

334 COLLEGE AND RESEARCH LIBRARIES 



L.C. D.C. 

Geography, Antiq- Greek: 
uities, and His- Roman: 
tory 

Numismatics 

Bibliography 

Periodicals and 
Society Publica-
tions 

A r t 

L a w 

Philosophy 

Mythology and 
Religion 

DF10-289 
D G i l - 3 6 5 

CJ201-1143 

Z7016-7026 

Classed according 
to subject 

N5605-5896 

K 

B165-708 

BL700-820 

Antiquities: 
History: 

9I3-37-9I3.38 
937-938 

737.37-737.38 

016 

Classed according 
to subject 

709.37-709.38 

349.37-349.38 

180 

292 

T h e consolidation of the major group 

of books, " L a n g u a g e and Literature," is 

much better than D . C . ' s curious assump-

tion (common though it is in classification 

schemes) that all languages belong to-

gether and all literatures together. M o r e -

over, L . C . ' s internal arrangement is 

excellent: In t w o large alphabetical files 

every Greek author and every Latin author 

has, or can easily find, a place. D . C . , on 

the other hand, insists upon its "mnemonic" 

singsong of nine points in Greek and Latin 

literatures. T h i s produces a curious re-

sult. In each literature " M i s c e l l a n y " is 

heavily laden while the other eight groups 

have only one or two reputable authors 

apiece to justify their existence. D . C . ' s 

shelving of " E p i g r a p h y " and "Palaeogra-

phy" in the language section is more use-

ful than L . C . ' s treatment, but both systems 

separate classical Greek and Latin litera-

ture from " E a r l y Christian Literature." 

T h e second major group, "Geography, 

Antiquities, and H i s t o r y , " is split by both 

systems, although the theory of splitting 

differs. Probably the D . C . arrangement is 

more useful, but neither is satisfactory. 

Books on the minor allied subject, "Numis-

matics," are to be found in still a third 

section. 

T h e importance of "Bibliography" and 

"Periodicals" cannot be overemphasized. 

In a careful analysis Grace O . Kelley has 

demonstrated that not more than one third 

of the material brought out under a spe-

cific subject in a dictionary catalog is 

shelved under that subject's specific class 

number.3 O f all literature upon a subject, 

the share brought out by classification is, 

of course, considerably less than one third. 

"Bibliography" and "Periodicals" are the 

scholar's key to the great bulk of material 

which classification cannot locate for him. 

But in neither scheme are classical bibli-

ographies shelved with any group of classi-

cal books, and in both schemes classical 

periodicals are scattered through the stacks 

3 K e l l e y , op. ext., 100-25. 

SEPTEMBER, 1942 
2 77 



according to specific subject. 

" A r t " and " L a w , " on the other hand, 

are examples of minor subjects whose 

segregation is merely irritating. Finally, 

"Philosophy" and " M y t h o l o g y " represent 

borderline groups of books to which the 

classical scholar can lay claim only if they 

have been purchased with funds at his 

disposal.4 

T o the classical scholar this scattering 

is senseless chaos; to the librarian it is 

justified by the very good reason that in 

the theories of the organization of univer-

sal knowledge upon which these systems 

are based Latin and Greek culture is not 

a unit. 

Theory Costly 

In a large library, theory costs the classi-

cal scholar many a weary mile. N o w it is 

true that H . E. Bliss presented the field 

of the classics as a "peculiar" problem, 

"one of the most difficult that the classifier 

has to face," because "the philological 

study of the culture has largely coalesced 

with the archeological study of the civiliza-

tion."5 But this "peculiarity" is in reality 

typical of the study of every literature. 

For the present it need be pointed out only 

that the Elizabethan scholar (to take the 

specific case of a modern literature) works 

with Elizabethan handwriting (palaeogra-

phy), Francis Bacon (philosophy and 

l a w ) , Elizabeth and Essex (history and 

biography), "rogues and vagabonds" 

(sociology and economics), and Thomas 

Cartwright (religion), as well as with the 

4 I n d e e d , the distinction seems a hit h a z y in the 
systems themselves. I n D . C . , f o r instance, u n d e r 
" P l a t o " in the 180's t h e r e is a note, " C l a s s his 
works p r e f e r a b l y in 888.4, hut discussion of his 
p h i l o s o f y h e r e . " I s P l a t o ' s Republic G r e e k litera-
ture, but P a u l S h o r e y ' s What Plato Said, G r e e k 
" p h i l o s o f y " ? 

8 Bliss, H . E . System of Classification. 2d ed , 
rev. ( 1 9 3 6 ) 38. 

plays of Shakespeare (literature). 

T h e librarian treats of one vast world 

whose parts he calls language, art, science, 

literature, history, philosophy, economics. 

T h e scholar busies himself, not with some 

small atom of these larger units, but with 

a cross section of that entire world—lan-

guage, science, literature, history, and all. 

T h e librarian's classification is, so to speak, 

vertical; the scholar's, horizontal. 

T h e significance of this conflict cannot 

be overemphasized. T h e stacks are where 

the scholar comes into most intimate con-

tact with the library. If that contact day 

after day invariably perplexes or infuriates 

him, he cannot fail to entertain some 

peculiar notions about librarians. And the 

scholar is a powerful library patron. Both 

the small college and the large university 

find that books, no less than salaries, keep 

able men on their faculties. By the same 

token, the librarian of the small college or 

the large university finds that the scholar 

often dominates faculty library committees. 

T h e A . L . A . figures on salary and tenure 

in university libraries as compared with 

the salary and tenure of university instruc-

tors can have no great significance so long 

as this fundamental difference about classi-

fication remains. T h e scholar will con-

tinue to feel, and with some justification, 

that the librarian whose stacks cause him 

all this trouble must persist in error be-

cause he is at worst mentally inferior or 

at best stubbornly pedantic. O n the 

faculty library committee and in his con-

ferences with other faculty members and 

with trustees, he will certainly voice his 

dissatisfaction. 

Clamorous Minority 

By way of defense the librarian may 

urge that the scholar represents only a 

336 COLLEGE AND RESEARCH LIBRARIES 



clamorous minority of the library's patrons. 

T h i s is to forget that he is the only patron 

with direct access to the stacks. So far as 

the stacks are concerned, a classification 

scheme which serves the scholar cannot 

discommode other patrons; their books 

from the stacks are secured for them by 

library-trained stack attendants who could 

easily find their way about stacks arranged 

for the convenience of a layman. 

A more valid objection is expense. 

Practically every large library in the 

United States has long since been saddled 

with some form of D . C . or L . C . shelving. 

T o change now—even to a perfect system 

of classification—must involve great ex-

penditure of money and time. Moreover, 

in a functional classification determined by 

experiment, revision would have to be con-

tinuous as new needs develop and old 

needs disappear. But classification exists 

only to serve; any system of classification 

which does not serve the reader is itself a 

tremendous expense for which little is re-

ceived in return. 

Finally, the librarian may contend that 

the scholar proposes to enrich the class of 

books in which he is interested at the ex-

pense of other classes in which other people 

are interested. It is, however, the funda-

mental principle of functional classification 

that classes are built up or weakened 

only as experiment shows reader interest in 

those classes is strong or weak respectively. 

N o two libraries can use exactly—or even 

nearly—the same system. It is well 

known that one school, for instance, spe-

cializes in the humanities, another in sci-

ence ; one in arts, another in social sciences. 

And within each major group there are 

weak and strong classes. Every librarian 

will have to build his own scheme about 

the major interests of his patrons. Cer-

tainly he will do well to invite and care-

fully consider the scholar's suggestions. 

Functional shelving, then, is the only 

classification with which the scholar will be 

content. H o w secure it? 

Special Reading Room 

T h e simplest expedient is, of course, 

the special reading room. T h e classics 

reading room of the University of Illinois 

is an excellent illustration. Into one room 

have been gathered nearly all books and 

periodicals relating to classical civilization. 

T h e shelf arrangement follows D . C . in 

general with the notable exception that 

the literature classification has been dis-

carded. A l l Latin authors are classed 

straight 871, all Greek authors, 881. 

Thus, the authors of each literature form 

one large alphabetical group, as is the case 

with L . C . ' s treatment of the classical 

literatures. Philosophy, church fathers, 

economics, the languages, the arts, the 

literatures, antiquities, history—the order 

of things is readily learned and the num-

bers which bring it about may be readily 

forgotten. T h e card catalog has been 

shoved into its proper subordination, for 

even first-year graduate students after a 

short time begin confidently to ignore it 

and to "feel their way around." In all 

this the classifier notes only one disquieting 

fact: the assembling of books is achieved, 

not by classification, but in spite of classi-

fication. T h e special reading room is a 

classification scheme's final confession of 

failure. 

Something might be gained in D . C . by 

a further application of the theory behind 

the Illinois revision of the classical litera-

ture groups. Confining the literatures to 

two numbers makes eighteen numbers 

(Continued on page 341) 

SEPTEMBER, 1942 2 77 



largely eliminate both. W e know, because 
it has been done. But it takes courage, 
determination, realization of need, and 
vast patience to overcome the problems 
and obstacles which lie on the road to 
success. Chief among them would seem 
to be the inability of administrators to see 
the advantages of cooperation and their 
unwillingness to enter into agreements, the 
fear of librarians that they will lose prestige 

or authority, and the difficulty of making 
legal and financial arrangements. 

T h e two volumes which have served as 
pegs upon which to hang these notes should 
be required reading for everyone inter-
ested in higher education. Their contents 
are, if the writer is any prophet, signposts 
of the f u t u r e . — J . Periam Danton, librar-
ian, Sullivan Memorial Library, Temple 
University, Philadelphia. 

Classification and the Scholar 
(Continued from page 337) 

available in the emptied 870's and 88o's, 

possibly enough to take care of all phases 

of classical study. By such a scheme 

classification, unaided, might bring to-

gether books according to their use. Simi-

lar revisions might be worked out in 

various other classes, both in D . C . and in 

L . C . O f course such a system of revision 

might soon cost more than to devise and 

install an entire new scheme of classifica-

tion. 

Certainly, reshelving without a revised 

classification can never succeed. "Objec-

tions to the order of the D . C . tables," 

Dorcas Fellows argued, "can be largely 

and easily overcome by adjustments in 

shelving, e.g., English philology (420) 

may be shelved next to English literature 

(820), travel in Italy (914.5) next to 

Italian history ( 9 4 5 ) , etc."6 T h e same 

"solution" could, of course, be worked 

out in L . C . But if classification does not 

indicate where a book may be found and 

if the stacks are to be a maze of jumbled 

letters and figures penetrable only to the 

initiated—why classify? 

8 F e l l o w s , Dorcas. " L i b r a r y of C o n g r e s s Classifi-
cation vs. D e c i m a l C l a s s i f i c a t i o n . " Library Journal 
50:292, A p r . 1, 1925. 

SEPTEMBER, 1942 
2 77