If there is any subject which, more than all others, divides opinion and provokes endless controversy among librarians and scholars, it is the proper classification of books. From the beginning of literature this has been a well-nigh insoluble problem. Treatise after treatise has been written upon it, system has been piled upon system, learned men have theorised and wrangled about it all their lives, and successive generations have dropped into their graves, leaving the vexed question as unsettled as ever. Every now and then a body of savans or a convention of librarians wrestles with it, and perhaps votes upon it,

since the dissatisfied minority, nearly as numerous and quite as obstinate as the majority, always refuses to be bound by it. No sooner does some sapient librarian, with the sublime confidence of conviction, get his classification house of cards constructed to his mind, and stands rapt in admiration before it, when there comes along some wise man of the east, and demolishes the fair edifice at a blow, while the architect stands by with a melancholy smile, and sees all his household gods lying shivered around him.

Meanwhile, systems of classification keep on growing, until, instead of the thirty-two systems so elaborately described in Edwards's Memoirs of Libraries, we have almost as many as there are libraries, if the endless modifications of them are taken into account. In fact, one begins to realise that the schemes for the classification of knowledge are becoming so numerous, that a classification of the systems themselves has fairly become a desideratum. The youthful neophyte, who is struggling after an education in library science, and thinks perhaps that it is or should be an exact science, is bewildered by the multitude of counsellors, gets a head-ache over their conflicting systems, and adds to it a heart-ache, perhaps, over the animosities and sarcasms which divide the warring schools of opinion.

Perhaps there would be less trouble about classification, if the system-mongers would consent to admit at the outset that no infallible system is possible, and would endeavor, amid all their other learning, to learn a little of the saving grace of modesty. A writer upon this subject has well observed that there is no man who can work out a scheme of classification that will satisfy permanently even himself. Much less should he expect that others, all having their favorite ideas and systems, should be satisfied with his. As there is no royal road to learning, so there can be none to classification; and we democratic republicans, who stand upon the threshold of the twentieth century, may rest satisfied that in the Republic of Letters no autocrat can be allowed.

The chief difficulty with most systems for distributing the books in a library appears to lie in the attempt to apply scientific minuteness in a region where it is largely inapplicable. One can divide and sub-divide the literature of any science indefinitely, in a list of subjects, but such exhaustive sub-divisions can never be made among the books on the shelves. Here, for example, is a "Treatise on diseases of the heart and lungs." This falls naturally into its two places in the subject catalogue, the one under "Heart," and the second under "Lungs;" but the attempt to classify it on the shelves must fail, as regards half its contents. You cannot tear the book to pieces to satisfy logical classification. Thousands of similar cases will occur, where the same book treats of several subjects. Nearly all periodicals and transactions of societies of every kind refuse to be classified, though they can be catalogued perfectly on paper by analysing their contents. To bring all the resources of the library on any subject together on the shelves is clearly impossible. They must be assembled for readers from various sections of the library, where the rule of analogy or of superior convenience has placed them.

What is termed close classification, it will be found, fails by attempting too much. One of the chief obstacles to its general use is that it involves a too complicated notation. The many letters and figures that indicate position on the shelves are difficult to remember in the direct ratio of their number. The more minute the classification, the more signs of location are required. When they become very numerous, in any system of classification, the system breaks down by its own weight. Library attendants consume an undue amount of time in learning it, and library cataloguers and classifiers in affixing the requisite signs of designation to the labels, the shelves, and the catalogues. Memory, too, is unduly taxed to apply the system. While a superior memory may be found equal to any task imposed upon it, average memories are not so fortunate. The expert librarian, in whose accomplished head the whole world of science and literature lies coÖrdinated, so that he can apply his classification unerringly to all the books in a vast library, must not presume that unskilled assistants can do the same.

One of the mistakes made by the positivists in classification is the claim that their favorite system can be applied to all libraries alike. That this is a fallacy may be seen in an example or two. Take the case of a large and comprehensive Botanical library, in which an exact scientific distribution of the books may and should be made. It is classified not only in the grand divisions, such as scientific and economic botany, etc., but a close analytical treatment is extended over the whole vegetable kingdom. Books treating of every plant are relegated to their appropriate classes, genera, and species, until the whole library is organised on a strictly scientific basis. But in the case, even of what are called large libraries, so minute a classification would be not only unnecessary, but even obstructive to prompt service of the books. And the average town library, containing only a shelf or two of botanical works, clearly has no use for such a classification. The attempt to impose a universal law upon library arrangement, while the conditions of the collections are endlessly varied, is foredoomed to failure.The object of classification is to bring order out of confusion, and to arrange the great mass of books in science and literature of which every library is composed, so that those on related subjects should be as nearly as possible brought together. Let us suppose a collection of some hundred thousand volumes, in all branches of human knowledge, thrown together without any classification or catalogue, on the tables, the shelves, and the floor of an extensive reading-room. Suppose also an assemblage of scholars and other readers, ready and anxious to avail themselves of these literary treasures, this immense library without a key. Each wants some certain book, by some author whose name he knows, or upon some subject upon which he seeks to inform himself. But how vain and hopeless the effort to go through all this chaos of learning, to find the one volume which he needs! This illustration points the prime necessity of classification of some kind, before a collection of books can be used in an available way.

Then comes in the skilled bibliographer, to convert this chaos into a cosmos, to illumine this darkness with the light of science. He distributes the whole mass, volume by volume, into a few great distinct classes; he creates families or sub-divisions in every class; he assembles together in groups all that treat of the same subject, or any of its branches; and thus the entire scattered multitude of volumes is at length coÖrdinated into a clear and systematic collection, ready for use in every department. A great library is like a great army: when unorganized, your army is a mere undisciplined mob: but divide and sub-divide it into army corps, divisions, brigades, regiments, and companies, and you can put your finger upon every man.

To make this complete organization of a library successful, one must have an organising mind, a wide acquaintance with literature, history, and the outlines, at least, of all the sciences; a knowledge of the ancient and of various modern languages; a quick intuition, a ripe judgment, a cultivated taste, a retentive memory, and a patience and perseverance that are inexhaustible.

Even were all these qualities possessed, there will be in the arrangement elements of discord and of a failure. A multitude of uncertain points in classification, and many exceptions will arise; and these must of necessity be settled arbitrarily. The more conversant one becomes with systems of classification, when reduced to practice, the more he becomes assured that a perfect bibliographical system is impossible.

Every system of classification must find its application fraught with doubts, complications, and difficulties; but the wise bibliographer will not pause in his work to resolve all these insoluble problems; he will classify the book in hand according to his best judgment at the moment it comes before him. He can no more afford to spend time over intricate questions of the preponderance of this, that, or the other subject in a book, than a man about to walk to a certain place can afford to debate whether he shall put his right foot forward or his left. The one thing needful is to go forward.

Referring to the chapter on bibliography for other details, I may here say that the French claim to have reached a highly practical system of classification in that set forth in J. C. Brunet's Manuel du Libraire. This is now generally used in the arrangement of collections of books in France, with some modifications, and the book trade find it so well adapted to their wants, that classified sale and auction catalogues are mostly arranged on that system. It has only five grand divisions: Theology, Law, Arts and Sciences, Belles-lettres, and History. Each of these classes has numerous sub-divisions. For example, geography and voyages and travels form a division of history, between the philosophy of history and chronology, etc.

The classification in use in the BibliothÈque nationale of France places Theology first, followed by Law, History, Philosophy and Belles-lettres. The grand division of Philosophy includes all which is classified under Arts and Sciences in the system of Brunet.

In the Library of the British Museum the classification starts with Theology, followed by 2. Jurisprudence; 3. Natural History (including Botany, Geology, ZoÖlogy, and Medicine); 4. Art (including Archaeology, Fine Arts, Architecture, Music, and Useful Arts); 5. Philosophy (including Politics, Economics, Sociology, Education, Ethics, Metaphysics, Mathematics, Military and Naval Science, and Chemistry); 6. History (including Heraldry and Genealogy); 7. Geography (including Ethnology); 8. Biography (including Epistles); 9. Belles-lettres (including Poetry, Drama, Rhetoric, Criticism, Bibliography, Collected Works, Encyclopaedias, Speeches, Proverbs, Anecdotes, Satirical and facetious works, Essays, Folklore and Fiction); 10. Philology.

Sub-divisions by countries are introduced in nearly all the classes.

In the Library of Congress the classification was originally based upon Lord Bacon's scheme for the division of knowledge into three great classes, according to the faculty of the mind employed in each. 1. History (based upon memory); 2. Philosophy (based upon reason); 3. Poetry (based upon imagination). This scheme was much better adapted to a classification of ideas than of books. Its failure to answer the ends of a practical classification of the library led to radical modifications of the plan, as applied to the books on the shelves, for reasons of logical arrangement, as well as of convenience. A more thorough and systematic re-arrangement is now in progress.

Mr. C. A. Cutter has devised a system of "Expansive classification," now widely used in American libraries. In this, the classes are each indicated by a single letter, followed by numbers representing divisions by countries, and these in turn by letters indicating sub-divisions by subjects, etc. It is claimed that this method is not a rigid unchangeable system, but adaptable in a high degree, and capable of modification to suit the special wants of any library. In it the whole range of literature and science is divided into several grand classes, which, with their sub-classes, are indicated by the twenty-six letters of the alphabet. Thus Class A embraces Generalia; B to D, Spiritual sciences (including philosophy and religion); E to G, Historical sciences (including, besides history and biography, geography and travels); H to K, Social sciences (including law and political science and economics); L to P, Natural sciences; Q, Medicine; and R to Z, Arts (including not only mechanical, recreative and fine arts, but music, languages, literature, and bibliography).

The sub-divisions of these principal classes are arranged with progressive fullness, to suit smaller or larger libraries. Thus, the first classification provides only eleven classes, suited to very small libraries: the second is expanded to fifteen classes, the third to thirty classes, and so on up to the seventh or final one, designed to provide for the arrangement of the very largest libraries.

This is the most elaborate and far-reaching library classification yet put forth, claiming superior clearness, flexibility, brevity of notation, logical coÖrdination, etc., while objections have been freely made to it on the score of over-refinement and aiming at the unattainable.

What is known as the decimal or the Dewey system of classification was originally suggested by Mr. N. B. Shurtleff's "Decimal system for the arrangement and administration of libraries," published at Boston in 1856. But in its present form it has been developed by Mr. Melvil Dewey into a most ingenious scheme for distributing the whole vast range of human knowledge into ten classes, marked from 0 to 9, each of which sub-divides into exactly ten sub-classes, all divisible in their turn into ten minor divisions, and so on until the material in hand, or the ingenuity of the classifier is exhausted. The notation of the books on the shelves corresponds to these divisions and sub-divisions. The claims of this system, which has been quite extensively followed in the smaller American libraries, and in many European ones, are economy, simplicity, brevity of notation, expansibility, unchanging call-numbers, etc. It has been criticised as too mechanical, as illogical in arrangement of classes, as presenting many incongruities in its divisions, as procrustean, as wholly inadequate in its classification of jurisprudence, etc. It is partially used by librarians who have had to introduce radical changes in portions of the classification, and in fact it is understood that the classification has been very largely made over both in Amherst College library and in that of Columbia University, N. Y., where it was fully established.

This only adds to the cumulative proofs that library classification cannot be made an exact science, but is in its nature indefinitely progressive and improvable. Its main object is not to classify knowledge, but books. There being multitudes of books that do not belong absolutely to any one class, all classification of them is necessarily a compromise. Nearly all the classification schemers have made over their schemes—some of them many times. I am not arguing against classification, which is essential to the practical utility of any library. An imperfect classification is much better than none: but the tendency to erect classification into a fetish, and to lay down cast-iron rules for it, should be guarded against. In any library, reasons of convenience must often prevail over logical arrangement; and he who spends time due to prompt library service in worrying over errors in a catalogue, or vexing his soul at a faulty classification, is as mistaken as those fussy individuals who fancy that they are personally responsible for the obliquity of the earth's axis.

It may be added that in the American Library Association's Catalogue of 5,000 books for a popular library, Washington, 1893, the classification is given both on the Dewey (Decimal) system, and on the Cutter expansive system, so that all may take their choice.

The fixed location system of arrangement, by which every book is assigned by its number to one definite shelf, is objectionable as preventing accessions from being placed with their cognate books. This is of such cardinal importance in every library, that a more elastic system of some kind should be adopted, to save continual re-numbering. No system which makes mere arithmetical progression a substitute for intrinsic qualities can long prove satisfactory.

The relative or movable location on shelves is now more generally adopted than the old plan of numbering every shelf and assigning a fixed location to every volume on that shelf. The book-marks, if designating simply the relative order of the volumes, permit the books to be moved along, as accessions come in, from shelf to shelf, as the latter become crowded. This does not derange the numbers, since the order of succession is observed.

For small town libraries no elaborate system of classification can properly be attempted. Here, the most convenient grouping is apt to prove the best, because books are most readily found by it. Mr. W. I. Fletcher has outlined a scheme for libraries of 10,000 volumes or less, as follows:

A. Fiction (appended, J. Juvenile books); B. English and American literature; C. History; D. Biography; E. Travels; F. Science; G. Useful arts; H. Fine and recreative arts; I. Political and social science; K. Philosophy and religion; L. Works on language and in foreign languages; R. Reference books.

Numerous sub-divisions would be required to make such a scheme (or indeed any other) fit any collection of books.

In arranging the main classes, care should be taken to bring those most drawn upon near to the delivery desk, or charging system of the library.

The alphabet is usefully applied in the arrangement of several of the great classes of books, and in many sub-divisions of other classes. Thus, all English and American fiction may be arranged in a single alphabet of authors, including English translations of foreign works. All collected works, or polygraphy, may form an alphabet, as well as poetry, dramatic works, collections of letters, and miscellanea, arranged by authors' names. In any of these classes, sub-divisions by languages may be made, if desired.

The class biography may best be arranged in an alphabet of the subjects of the biographies, rather than of writers, for obvious reasons of convenience in finding at once the books about each person.