A Cla.s.sification and Subject Index for Cataloguing and Arranging the Books and Pamphlets of a Library.

by Melvil Dewey.

PREFACE.

The plan of the following Cla.s.sification and Index was developed early in 1873. It was the result of several months' study of library economy as found in some hundreds of books and pamphlets, and in over fifty personal visits to various American libraries. In this study, the author became convinced that the usefulness of these libraries might be greatly increased without additional expenditure. Three years practical use of the system here explained, leads him to believe that it will accomplish this result; for with its aid, the catalogues, shelf lists, indexes, and cross-references essential to this increased usefulness, can be made more economically than by any other method which he has been able to find.

The system was devised for cataloguing and indexing purposes, but it was found on trial to be equally valuable for numbering and arranging books and pamphlets on the shelves.

The library is first divided into nine special libraries which are called Cla.s.ses. These Cla.s.ses are Philosophy, Theology, &c., and are numbered with the nine digits. Thus Cla.s.s 9 is the Library of History; Cla.s.s 7, the Library of Fine Art; Cla.s.s 2, the Library of Theology. These special libraries or Cla.s.ses are then considered independently, and each one is separated again into nine special Divisions of the main subject. These Divisions are numbered from 1 to 9 as were the Cla.s.ses. Thus 59 is the 9th Division (Zoology) of the 5th Cla.s.s (Natural Science). A final division is then made by separating each of these Divisions into nine Sections which are numbered in the same way, with the nine digits. Thus 513 is the 3d Section (Geometry) of the 1st Division (Mathematics) of the 5th Cla.s.s (Natural Science). This number, giving Cla.s.s, Division, and Section, is called the Cla.s.sification or Cla.s.s Number, and is applied to every book or pamphlet belonging to the library. All the Geometries are thus numbered 513, all the Mineralogies 549, and so throughout the library, all the books on any given subject bear the number of that subject in the scheme. Where a 0 occurs in a cla.s.s number, it has its normal zero power. Thus, a book numbered 510, is Cla.s.s 5, Division 1, but _no_ Section. This signifies that the book treats of the Division 51 (Mathematics) in general, and is not limited to any one Section, as is the Geometry, marked 513. If marked 500, it would indicate a treatise on Science in general, limited to _no_ Division. A zero occurring in the first place would in the same way show that the book is limited to _no_ Cla.s.s. The cla.s.sification is mainly made by subjects or content regardless of _form_; but it is found practically useful to make an additional distinction in these general treatises, according to the form of treatment adopted. Thus, in Science we have a large number of books treating of Science in general, and so having a 0 for the Division number. These books are then divided into Sections, as are those of the other Cla.s.ses according to the form they have taken on. We have the Philosophy and History of Science, Scientific Compends, Dictionaries, Essays, Periodicals, Societies, Education, and Travels,--all having the common subject, =NATURAL SCIENCE=, but treating it in these varied forms.

These form distinctions are introduced here because the number of general works is large, and the numerals allow of this division, without extra labor for the numbers from 501 to 509 would otherwise be unused. They apply _only_ to the general treatises, which, without them, would have a cla.s.s number ending with two zeros. A Dictionary of Mathematics is 510, not 503, for every book is a.s.signed to the _most specific head that will contain it_, so that 503 is limited to Dictionaries or Cyclopedias of Science _in general_. In the same way a General Cyclopedia or Periodical treats of no one cla.s.s, and so is a.s.signed to the Cla.s.s 0. These books treating of no special cla.s.s, but general in their character, are divided into Cyclopedias, Periodicals, etc. No difficulty is found in following the arithmetical law and omitting the initial zero, so these numbers are printed 31, 32, etc., instead of 031, 032, etc.

The selection and arrangement of the thousand headings of the cla.s.sification cannot be explained in detail for want of s.p.a.ce. In all the work, philosophical theory and accuracy have been made to yield to practical usefulness. The impossibility of making a satisfactory cla.s.sification of all knowledge as preserved in books, has been appreciated from the first, and nothing of the kind attempted.

Theoretical harmony and exactness has been repeatedly sacrificed to the practical requirements of the library or to the convenience of the department in the college. As in every scheme, many minor subjects have been put under general heads to which they do not strictly belong. In some cases these headings have been printed in a distinctive type, e. g., 429 =Anglo-Saxon=, under =ENGLISH PHILOLOGY=. The rule has been to a.s.sign these subjects to the most nearly allied heads, or where it was thought they would be most useful. The only alternative was to omit them altogether. If any such omission occurs, it is unintentional and will be supplied as soon as discovered. Wherever practicable the heads have been so arranged that each subject is preceded and followed by the most nearly allied subjects and thus the greatest convenience is secured both in the catalogues and on the shelves. Theoretically, the division of every subject into just nine heads is absurd. Practically, it is desirable that the cla.s.sification be as minute as possible without the use of additional figures, and the decimal principle on which our scheme hinges allows nine divisions as readily as a less number. This principle has proved wholly satisfactory in practice though it appears to destroy proper co-ordination in some places. It has seemed best in our library to use uniformly three figures in the cla.s.s number. This enables us to cla.s.sify certain subjects very minutely, giving, for example, an entire section to Chess. But the History of England has only one section, as our scheme is developed, and thus the two might be said to be co-ordinated. The apparent difficulty in such cases is entirely obviated by the use of a fourth figure, giving nine sub-sections to any subject of sufficient importance to warrant closer cla.s.sification. In history where the cla.s.sification is made wholly by countries, a fourth figure is added to give a division into _periods_. As the addition of each figure gives a ten-fold division, any desired degree of minuteness may be secured in the cla.s.sing of special subjects. The apparent lack of co-ordination arises from the fact that only the first three figures of these more important heads are as yet printed, the fourth figure and the sub-sections being supplied on the catalogues in ma.n.u.script. Should the growth of any of these sub-sections warrant it, a fifth figure will be added, for the scheme admits of expansion without limit.

The arrangement of headings has been sometimes modified to secure a mnemonic aid in numbering and finding books without the Index. For instance, the scheme is so arranged that China has always the number 1.

In Ancient History, it has the first section, 931: in Modern History, under Asia, it has 951: in Philology, the Chinese language appears as 491. After the same manner the Indian number is 2; Egyptian, 4; English, 2; German, 3; French, 4; Italian, 5; Spanish, 6; European, 4; Asian, 5; African, 6; North American, 7; South American, 8; and so for all the divisions by languages or countries. The Italian 5, for instance, will be noticed in 35, 55, 450, 755, 850, and 945. This mnemonic principle is specially prominent in Philology and Literature and their divisions, and in the _form_ distinctions used in the first 9 sections of each cla.s.s.

Materials, Methods, or Theory occurring anywhere as a head, bears always the number 1. Dictionaries and Cyclopedias, 3; Essays, 4; Periodicals, 5; a.s.sociations, Inst.i.tutions, and Societies, 6; Education, 7; Collections, 9. In the numerous cases where several minor heads have been grouped together under the head Other, it always bears the number 9. Wherever practicable, this principle is carried out in sub-dividing the sections.

For instance, the Geology of North America, which bears the number 557 is sub-divided by adding the _sections_ of 970 (History of North America).

The Geology of Mexico then bears the number 5578: mnemonically, the first 5 is the Science number; the second 5, Geology; the 7, North America; and the 8, Mexico. Any library attendant or reader after using the scheme a short time will recognize at a glance, any catalogue or ledger entry, book or pamphlet, marked 5578 as something on the Geology of Mexico.

Users of the scheme will notice this mnemonic principle in several hundred places in the cla.s.sification, and will find it of great practical utility in numbering and finding books without the aid of Catalogue or Index, and in determining the character of any book simply from its call number as recorded on the book, on all its catalogue and cross reference cards, on the ledger, and in the check box.

In naming the headings, brevity has been secured in many cases at the sacrifice of exactness. It was thought more important to have short, familiar t.i.tles for the headings than that the names given should express with fullness and exactness the character of all books catalogued under them. Many subjects, apparently omitted, will be found in the Index, a.s.signed, with allied subjects, to a heading which bears the name of the most important only. Reference to this Subject Index will decide at once any doubtful points.

In arranging books in the cla.s.sification, as in filling out the scheme, practical usefulness has been esteemed the most important thing. The effort has been to put each book under the subject to the student of which it would be most useful. The content or the real subject of which a book treats, and not the form or the accidental wording of the t.i.tle, determines its place. Following this rule, a Philosophy of Art is put with Art, not with Philosophy; a History of Mathematics, with Mathematics, not with History; for the philosophy and history are simply the _form_ which these books have taken. The true content or subject is Art, and Mathematics, and to the student of these subjects they are most useful. The predominant tendency or obvious purpose of the book, usually decides its cla.s.s number at once; still many books treat of two or more different subjects, and in such cases it is a.s.signed to the place where it will be most useful, and underneath the cla.s.s number are written the numbers of any other subjects on which it also treats. These _Cross References_ are given both on the plate and the subject card as well as on the cross reference card. If a book treats of a majority of the sections of any division, it is given the Division number instead of the most important Section number with cross references.

Collected works, libraries, etc., are either kept together and a.s.signed like individual books to the most specific head that will contain them; or a.s.signed to the most prominent of the various subjects on which they treat with cross references from the others; or are separated and the parts cla.s.sed as independent works. Translations are cla.s.sed with their originals.

The Alphabetical Subject Index is designed to guide, both in numbering and in finding the books. In numbering, the most specific head that will contain the book having been determined, reference to that head in the Index will give the cla.s.s number to which it should be a.s.signed. In finding books on any given subject, reference to the Index will give the number under which they are to be sought on the shelves, in the Shelf Catalogue, or in the Subject Catalogue. The Index gives after each subject the number of the cla.s.s to which it is a.s.signed. Most names of countries, towns, animals, plants, minerals, diseases, &c, have been omitted, the aim being to furnish an Index of Subjects on which books are written, and not a Gazetteer or a Dictionary of all the nouns in the language. Such subjects will be found as special chapters or sections of books on the subjects given in the Index. The names of individual subjects of biographies will be found in the Cla.s.s List of Biography.

Omissions of any of the more general subjects will be supplied when brought to notice.

In arranging the books on the shelves, the absolute location by shelf and book number is wholly abandoned, the relative location by cla.s.s and book number being one of the most valuable features of the plan. The cla.s.s number serves also as the location number and the shelf number in common use is entirely dispensed with. Accompanying the cla.s.s number is the _book_ number, which prevents confusion of different books on the same subject. Thus the first Geometry catalogued is marked 513-1; the second 513-2, and so on to any extent, the last number showing how many books the library has on that subject. The books of each section are all together, and arranged by book numbers, and these sections are also arranged in simple numerical order throughout the library. The call number 513-11 signifies not the 11th book on shelf 513; or alcove 5, range 1, shelf 3, as in most libraries, but signifies the 11th book in subject 513 or the 11th Geometry belonging to the library. In finding the book, the printed numbers on the backs are followed, the upper being the cla.s.s and the lower the book number. The cla.s.s is found in its numerical order among the cla.s.ses as the shelf is found in the ordinary system: the book in its numerical order in the cla.s.s. The shelves are not numbered, as the increase of different departments, the opening of new rooms, and any arrangement of cla.s.ses to bring the books most circulated nearest to the delivery desk, will bring different cla.s.s numbers on a given shelf.

New books as received are numbered and put into place, in the same way that new t.i.tles are added to the card catalogue.

The single digit occasionally prefixed to the book number, e.g. the 3 in 421-3-7 is the nearest height in decimeters of books too large to be put on the regular library shelves, which are only 2-1/2 decimeters apart.

The great ma.s.s of the library consists of 2-decimeter books, the size numbers of which are omitted. Books from 2-1/2 to 3-1/2 decimeters in height have 3 prefixed to the book number, and are found on the bottom shelf of each range. The larger sizes are prefixed with 4, 5, &c., and are found on the special shelves provided, in order to avoid the great waste of s.p.a.ce otherwise occasioned by the relative location. By this use of the size numbers a close economy of s.p.a.ce is secured.

Thus all the books on any given subject are found standing together, and no additions or changes ever separate them. Not only are all the books on the subject sought, found together, but the most nearly allied subjects precede and follow, they in turn being preceded and followed by other allied subjects as far as practicable. Readers not having access to the shelves find the short t.i.tles arranged in the same order on the Shelf Catalogue, and the full t.i.tles, imprints, cross references, notes, &c., on the Subject Catalogue. The uncatalogued pamphlets treating of any subject bear the same cla.s.s number and are arranged on the shelves immediately after the books of each section.

In both the Authors' Catalogue and the Subject Index, brevity has been studied because of the economy, but more because of the much greater ease of reference to a short t.i.tle catalogue. The custom of giving full t.i.tles, etc., under authors, and only references or very brief t.i.tles under subjects, has been reversed. A reader seeking a book of a _known author_, in the vast majority of cases, wants simply the number by which to call for it, and can find it much sooner in a brief t.i.tle catalogue.

In the rare cases where more is needed the cla.s.s number refers instantly to all these facts on the cards. On the other hand, a reader seeking books on a _known subject_, needs the full t.i.tle, imprint, cross-references, and notes, to enable him to choose the book best suited to his wants.

The Subject Catalogue is a full t.i.tle Shelf List on cards and is for the use of the public. The Shelf List is a short t.i.tle Subject Catalogue in book form, made of separate sheets laced into an Emerson binder, and is for official use. We thus have without extra labor, both full and short t.i.tle Subject Catalogues and Shelf Lists. The public Authors' Catalogue is a printed volume; the official Authors' Catalogue or Index is on cards. As a result each of the public Catalogues is checked by an official Catalogue; each of the card Catalogues by a book Catalogue; each of the brief t.i.tle catalogues by a full t.i.tle catalogue--an advantage that will be appreciated by all librarians desiring accuracy of administration and catalogues.

The Arabic numerals can be written and found more quickly, and with less danger of confusion or mistake, than any other symbols whatever.

Therefore the Roman numerals, capitals and small letters, and similar symbols usually found in systems of cla.s.sification are entirely discarded and by the exclusive use of Arabic numerals in their regular order throughout the shelves, cla.s.sifications, indexes, catalogues and records, there is secured the greatest accuracy, economy, and convenience. This advantage is specially prominent in comparison with systems where the name of the author or the t.i.tle must be written in calling for or charging books and in making references.

Throughout the catalogues the number of a book shows not only _where_ it is but _what_ it is. On the library accounts the character of each person's reading is clearly indicated by the numbers charged, and the minutest statistics of circulation in any subject are made by simply counting the call slips in the check box, and recording the number against the cla.s.s number in the record.

By the use of size numbers the greatest possible economy of s.p.a.ce may be secured, for the size distinction may be made for every inch or even less if desired, and this without additional labor, as it will be seen that the size figure, when introduced, requires one less figure in the book number, and so does not increase the number of digits as would at first appear.

Parts of sets, and books on the same or allied subjects, are never separated as they are sure to be, sooner or later, in every library arranged on the common plan, unless it be frequently re-arranged and re-catalogued. The great expense of this re-cataloguing makes it impracticable except for a few very wealthy libraries. In this system the catalogue and book numbers remain unchanged through all changes of shelving, buildings, or arrangement. In addition to its own peculiar merits, this plan has all the advantages of the card catalogue principle and of the relative location, which have been used and very strongly approved by prominent libraries.

As in the card catalogue system, there is room for indefinite expansion without devices or provisions. s.p.a.ce is the only requisite and if the shelf room is exhausted, the floor s.p.a.ce is equally good, except for the inconvenience of stooping.

Some prominent opponents of cla.s.sed catalogues have admitted that the Subject Index, in deciding where to cla.s.s a book at first, and where to look for it ever afterwards, has removed their strongest objections.

Certainly it would be impossible to make an Index more cheaply or more easy of reference, it being a single alphabet, of single words, followed by single numbers.

These cla.s.s numbers applied to pamphlets have proved specially satisfactory. The number is written on the upper left corner and the pamphlets are arranged either in pamphlet cases with the books on the same subject or on special shelves divided every decimeter by perpendicular sections. As each pamphlet is examined when received into the library, it is the work of a single moment to pencil on it its cla.s.s number. There is no expense whatever incurred, and yet the entire pamphlet resources of the library on any subject can be produced almost instantly. The immense advantages of this plan over those in common use, both in economy and usefulness, will be appreciated by every librarian caring for a pamphlet collection. A catalogue of authors may be made on slips if desired. The pamphlets themselves are the best Subject Catalogue.

Though designed wholly for library use, the plan has proved of great service in preserving newspaper clippings in large envelopes arranged by cla.s.s numbers; and more especially in taking the place of the common note-book and Index Rerum. Slips of uniform size are used with the cla.s.s number of the subject written on the corner. Minute alphabetical headings are used under each cla.s.s number, the slips being arranged in numerical order like the Subject Card Catalogue. Clippings and notes arranged in this way are at all times their own complete index, and have the same advantages over the common sc.r.a.p and note-books that the Subject Catalogue has over the Accessions Book, in looking up the resources of the library on any given subject. Those who have tried this method are so enthusiastic in its praise that it seemed worthy of mention in this place.

The plan was adopted in the Amherst College Library in 1873, and the work of transferring the entire library to the new catalogue at once commenced. It was found entirely practicable to make the change gradually, as means allowed, without interfering in any appreciable degree with the circulation of the books. The three years trial to which it has been there subjected has more than justified the claims of its friends, and it is now printed with the more confidence on this account.

It has been kept in ma.n.u.script up to this time, in order that the many minor details might be subjected to actual trial and modified where improvement was possible. The labor involved in preparing the Cla.s.sification and Index has been wholly beyond the appreciation of any who have never attempted a similar task. Much valuable aid has been rendered by specialists in many departments, and nearly every member of the Faculty has given advice from time to time. Among the many to whom thanks are due, special mention should be made of Mr. C.A. Cutter, the librarian of the Boston Athenaeum, and Mr. John Fiske, of the Harvard University library, for valuable suggestions and appreciative criticism.

While these friends are in no way responsible for any remaining imperfections in the scheme, they should have credit for many improvements which have been made during these three years of revision.

The essential character of the plan has remained unchanged from the first. Doubtless other improvements are still possible, and it is hoped that users of the scheme will call attention to any proposed change in the naming or arrangement of the headings, or to any omission which should be supplied in the Subject Index.

Before printing, the plan was submitted to quite a number of librarians for criticism. Among the hundreds of points raised as to its practical workings and usefulness there was only one in which it was not shown to be equal or superior to any other system known. This objection applied only to the arrangement on the shelves; not at all to the catalogues or indexes. It was, that in this relative location, a book which this year stands, e.g., at the end of a certain shelf; may not be on that shelf at all another year, because of the uneven growth of the parts of the library. This slight objection inheres in any system where the books are arranged by _subjects_ rather than by windows, doors, shelves, and similar non-intellectual distinctions.

In this hurriedly prepared account of his plan, the author has doubtless failed to meet many objections which may be raised and which he could easily answer. He would therefore ask the privilege of replying personally to any such objections, where they arise, believing that it will be possible to answer, if not all, at least a very large proportion.

In his varied reading, correspondence, and conversation on the subject, the author doubtless received suggestions and gained ideas which it is now impossible for him to acknowledge. Perhaps the most fruitful source of ideas was the _Nuovo Sistema di Catalogo Bibliografico Generale_ of Natale Battezzati, of Milan. Certainly he is indebted to this system adopted by the Italian publishers in 1871, though he has copied nothing from it. The plan of the St. Louis Public School Library, and that of the Apprentices' Library of New York, which in some respects resemble his own, were not seen till all the essential features were decided upon, though not given to the public. In filling the nine cla.s.ses of the scheme the inverted Baconian arrangement of the St. Louis Library has been followed. The author has no desire to claim original invention for any part of his system where another has been before him, and would most gladly make specific acknowledgment of every aid and suggestion were it in his power to do so. With these general explanations and acknowledgments he submits the scheme, hoping it may prove as useful to others as it has to himself.

AMHERST COLLEGE LIBRARY,

June 10th, 1876.

Those interested will find fuller explanations and remarks in the Library volume now being printed by the Bureau of Education at Washington.

(GENERAL) PHILOSOPHY, THEOLOGY, SOCIOLOGY, PHILOLOGY, NATURAL SCIENCE, USEFUL ARTS, FINE ARTS, LITERATURE, HISTORY.

DIVISIONS.

(GENERAL)

= 0= 10 BIBLIOGRAPHY.

20 BOOK RARITIES.

30 GENERAL CYCLOPEDIAS.