The two most important human inventions are perhaps Speech and Writing. In no adequate way could man fully express his mental activities except by speech: nor could he adequately record them except by writing. And to some extent the development of the two ran a parallel course, for just as early Speech is largely an imitation of natural sounds (Onomatopœia), so early Writing is probably, in the first instance, entirely from pictures (pictograms), which from the need of acceleration became worn down to simpler forms, and finally to letters; the sense similarly declining from the plain or derivative ideas arising from pictures, to mere syllables with no intrinsic meaning, and finally to simple sounds corresponding to letters. But even this latest stage was reached in the Valley of the Nile by, or not long after, 3000 B.C., as is testified by a stelè representing the cult of Send, a king of the second Egyptian Dynasty (now in the Ashmolean Museum at Oxford), in which the letters S, N, D occur in a cartouche; while in 1750 the North American Indians used pictograms, and the practice is not yet extinct, for we use the symbol ☞

As soon as Writing occurs in a portable form, whether as cylinders of baked clay in Assyria, or as tablets of bark, wood or wax, or as sheets of parchment or paper, there is a volume, and there is the possibility of a collection of books which forms a Library. The earliest library of which a plan can be reconstructed was discovered at Nineveh by Layard and dates from about 700 B.C. Of early Egyptian libraries no trace has been found; but under the Macedonian dynasty of the Ptolemies the great library at Alexandria in Egypt was founded about 285 B.C. About libraries in Greece information is very scant until the time of Euripides (fifth century B.C.) In Italy, the earliest public library was founded soon after 39 B.C. by C. Asinius Pollio, and under Augustus at least two more were instituted at Rome; and by the fourth century at least twenty-five or so were there erected, but all of them are stated to have been closed down, “bybliothecis sepulcrorum ritu in perpetuum clausis” (Ammianus Marcellinus). Of course these public libraries had been preceded by private collections, of which records are few and far between.

Before the second century B.C. the contents of Greek and Roman libraries were not books in the form to which we are accustomed, but rolls of papyrus, and the fittings of libraries were accordingly shelves with vertical divisions forming pigeon-holes, into which a roll or rolls were thrust. But the advantages of parchment as a material to bear writing were so obvious that the long rolls wrapped round a stick (rotuli) gave way gradually to the well-known form appropriate to parchment or paper, namely sewn and bound leaves (codices) either quarto—which was found at first to be the most convenient size—or folio or octavo. The receptacles in libraries (at Rome) were no longer pigeon-holes (nidi, foruli) or long narrow boxes (loculamenta), but undivided shelves (either fixed to the wall, pegmata, or standing against it, plutei). The wall space above these various forms of shelving was usually filled with decorative portraits or busts of great authors. Perhaps it is not accidental that in the great library collected by Sir Robert Cotton (now in the British Museum) the various parts were called (as they are still referred to) by the names of the Roman Emperors whose busts surmounted each division, and in the Bodleian, after the Civil War, the wall divisions of the Picture Gallery were surmounted by a series of ornamental medallions and portraits above the ordinary pictures on the walls. The books, whether disposed in regular book cases or in cupboards (armaria) with doors, were in Roman times laid on their side.

The classical tradition concerning the care of books was apparently carried on with little change during the first five centuries of the Christian era. Ecclesiastical libraries were kept in churches, as classical libraries were usually connected with temples. It is usual to date the organization of mediæval collections from the Rule of St. Benedict in the sixth century; and the Order which he founded, with its direct offshoots, the Cluniacs, Carthusians and Cistercians, were foremost in assigning a place to literature in the daily duties of the monk. The gradual evolution of the library can be traced from the time when a few volumes could be accommodated in a press or presses in a recess or small room, usually at the north-east corner of the cloister near the chapter-house, to the fully developed collections of the fifteenth century, housed in a large separate room, which was almost always on the first floor, perhaps to avoid damp and to secure a good light.

The following figures will illustrate the size of some considerable mediæval libraries in various places and times.

The earliest fittings for the convenience of readers which still exist suggest that at any rate in collegiate institutions the first kind of desk was on what Mr. J. Willis Clark termed the Lectern system. One has to imagine a double lectern (that is, two plain lecterns placed back to back) so prolonged as to hold say five to fifteen volumes on each side, every volume lying on its side and being chained to the desk. The reader stood or sat, and the open book lay before him at a convenient angle, as in church lecterns. The system was obviously very wasteful of space, and was evolved from the time when one or two books only were brought out of the cupboard (armarium) for a reader’s use on a desk resembling a lectern, near a window. In the fifteenth century the normal type of library was a narrow room of considerable length, lighted by narrow windows at short intervals. From the wall-spaces between the windows there projected into the room at right angles to the wall the lectern desks described above, the desks by this time being probably fitted beneath with shelves on which the (chained) books could stand upright, and be pulled out on to the lectern for purposes of reading.

Next came the “Stall system.” In this, the cases projecting from the wall are just book-cases, of three or four or even more shelves on each side of the case, and the lectern part is superseded by a flat desk in front of, and attached to, the shelves at which a reader sits, and on which he places the (chained) book he takes down from a shelf above him. The earliest example of these is at Corpus Christi College, Oxford (1517), and they are still in use in the Old Reading Room (“Duke Humphrey”) of the Bodleian; and in the Chapter library at Hereford, the very chains of the books are still preserved.

The modern practice of lining the walls of our rooms with book-shelves, though in one sense a reversion to an ancient type, in another is a definite change of system, of which examples are found in the sixteenth century abroad, but the first in England is in the Bodleian. In the Old Reading Room Sir Thomas Bodley reproduced the stall system. But when in his last days he erected an additional room (the Arts End) his choice fell on the newer idea that a wide space should be left in the middle of the room, and the books be all placed against the walls (1613). The lower books (folios) were still chained, and could only be used at the desks attached; such was the case, for instance, with the First Folio of Shakespeare, marked S. 2.17 Art. (see p. 46). The Gallery books were quartos and octavos, not chained, and given out to readers below by one of the officers of the Library. This arrangement was imitated in the Selden End, which was completed in 1640.

So far an attempt has been made to sketch briefly the evolution of a library in ancient and mediæval times. The plan of the present chapter of generalities now requires that we should consider the kinds of modern library, and the position which the Bodleian holds among them.

There are libraries and libraries. There are libraries of Reference or, as they are sometimes termed, Deposit; and there are libraries of the second rank, adapted for local and popular use. These two kinds are typical, and fall apart, as will be seen, by the application of a simple test.

The first public literary need of a district is a library of the latter type, one in which it is more important that the books should be in general use by lending and by free access to the shelves, than that they should be invested with a fictitious importance and too jealously guarded. This is the way to fan the faint sparks of literary instinct, and is still the common condition of most villages and even towns. But wherever literature has taken a long and firm hold there is a desire for a public library of another kind. By this time private libraries abound, and every serious student surrounds himself with the ordinary books on his special subject. And the public library shares in this general elevation of the personal standard; it becomes a necessary complement to individual effort. It supplies not only the ordinary works on every subject, but also the extraordinary and special books, and owing to the value of its contents has to restrict or abolish the custom of lending, in order that there may be continuous accessibility to its volumes. This division into two kinds is deep and real. The two fall apart according to their answer to the question, Do you in any real sense aim at being complete? All libraries of the second order confess that, in cases where they aim at any completeness, it is in a special subdivision or set of subdivisions of literature—incunabula, local books, and so on, or a working series of general reference volumes, or a departmental library rich in books on a particular science. But every library of the first class must have universality and completeness (within human limitations) as its theoretical aim, and must have made substantial progress towards its goal. In countries like Great Britain and the United States there are almost numberless libraries, many of them really large ones; but above them there tower the few, the very few, Libraries of Deposit. These are the super-Dreadnoughts of the literary world, and the Bodleian claims to be among them. It is an especial glory of Sir Thomas Bodley that he, a man of the world as well as a genuine scholar, planned from the first a library of the highest order.

Of this highest class it may be said that a really great library should have Universal scope, Independence, Size, Permanence, Wealth, and multiform Utility. Its Catalogue is not a mere book-finder, but a scientific work. Open access and lending are given up, but the Reference library is large. The aim of the British Museum to possess the largest French library outside France, and so with other foreign departments, and to store all local newspapers, is beyond praise.

Within the second class of libraries the manifold and various needs of modern students have resulted in a large number of kinds, which have been so seldom and so incompletely methodized that the following first attempt at their classification may be allowed, though cross-divisions cannot be entirely avoided.

LIBRARIES

• A. Public—general:
• e.g. National, Free, Lending, Circulating, Village.
• B. Public—special:
• e.g. Learned, Theological, Branch, Local.
• C. Institutional—general:
• e.g. Proprietary, Subscription, University, College, Travelling.
• D. Institutional—special:
• e.g. Technical, Professional, Departmental, Museum, Children’s.
• E. Private—general:
• e.g. Aristocratic or dilettante, Scholarly, working Reference.
• F. Private—special:
• Libraries of Collectors of special kinds of books.

It remains to indicate the place among libraries which the Bodleian Library, the Library of the University of Oxford, appears to occupy among the libraries of the world. In the absence of all proper standardization of statistics, size (as estimated by the reputed number of volumes, not works) must be first regarded, and next importance, such as that due to manuscripts, incunabula and the like. Mere size may not connote importance, for the million volumes claimed by Messrs. Foyle, of 121 Charing Cross Road, London, do not raise their twenty miles of shelving to the rank of a really great library. So too the large collections of printed books in the United States are as yet wanting in the special importance derived from the ancient and cardinal manuscripts which enrich some of the libraries of Europe.

The following libraries of the Old World contain a million or more volumes, in the opinion of Dr. Fortescue, late Keeper of the Printed Books in the British Museum, expressed in 1912, and compared with numbers given in the last edition of Minerva (1913), and with other sources of information.

The value of the MSS. (45,000 Min.) in the Vatican Library at Rome raises it to the first rank, though its printed volumes are only reckoned to be 400,000 (Min.). The size of the great libraries in the United States is undoubted, but, as has been remarked, the absence of large and valuable collections of MSS. reduces their importance. In the Library of Congress at Washington, where the printed volumes (“books and pamphlets”) are estimated in the Annual Report for 1915 at 2,364,000, the mileage of occupied shelves is believed to be eighty or more, which is not far from double that of the British Museum. The New York Public Library is estimated to contain about 1,920,000 volumes (Fort.) or 2,090,000 (Min.). The Boston Public Library is believed to possess 1,050,000 volumes (Min.).

A statistical survey of the Bodleian Library was made in 1915, with the following results:—

(Full details are printed in No. 9, Vol. I, of the Bodleian Quarterly Record, April, 1916. The total number of printed books in the world, i.e. separate works, has been estimated at about 12,000,000.)

The Bodleian may fairly claim to rank in size about ninth, and in size and importance (together) about eighth. It is the largest and most important University library in the world, and the largest (at any rate in the Old World) which is not aided out of State funds. It claims also to be one of the earliest public libraries of Europe, in the sense that it has always been open to those who bring a sufficient recommendation, practically without distinction of class or nationality. In early days a small charge was made on first admission.