16 JULY 1932, Page 24

Books in Chains

To add to, nay, even to correct, the late J. W. Clark's classic treatise on The Care of Books would have seemed very difficult Until one had read Canon Streeter's new book. But there can be no question as to his achievement in establishing the history of the English library on a sound basis, and showing more clearly than before how and why it evolved from a simple book-cupboard or chest or a single book chained on a desk into g Bodleian. Clark was unlucky in dating the Hereford Cathedral library two centuries too early. That picturesque collection, with its chained volumes, looks mediaeval enough at a first glance to be dated 1390, but it is, in fact, of the Shakespearean era and dates from 1590. The correction is fundamental, for the Hereford " stall-system," adopted a few years later in " Duke Humphrey's library " at the Bodleian, was, in fact, a new and late development in library furnishing. So far from chained libraries being a mediaeval pheno- menon, they persisted well into the reign of George III. Magdalen, Oxford, abolished its book-chains in 1799. Hereford Cathedral and Guildford Grammar School retain them still.

Canon Streeter has had the advantage of watching the restoration of the Hereford library, effected at the cost of a generous donor, and has examined all the old college libraries at Oxford and Cambridge, with others elsewhere. All these he describes and illustrates most charmingly. His evolu- tionary theory is simple and clear. The books of a monastery were kept in an Winery (cupboard) or chest, and lent out to readers, each of whom had a seat and desk in a " carrel " or study set against the cloister window. The books used in the choir were chained to lecterns. The early colleges continued the practice. Then the lectern with its chained books and the " carrel " were brought together, thus creating what J. W. Clark called the lectern system. Each lectern stood at right angles to a library window. Between each pair of lecterns was a double seat. Michelangelo's Laurentian library at Florence is the most beautiful and ornate example of the arrangement. A church at Zutphen offers another example, and a third is at Trinity Hall, where it was installed about the year 1600. But a further development came at Hereford, where the almery (cupboard) was placed on top of a modified lectern, and thus produced the stall system, best known in the oldest section of the Bodleian. In this system there were three shelves of books, all chained, above the hinged desk at which the reader sat. The innovation was popular at Oxford, but not at Cambridge, where the chains and desks were eliminated, leaving the bookcases free to receive books down to the floor, as in St. John's library. A Cambridge man might draw awkward conclusions from the fact that book-chains were abandoned at Cambridge early in the seventeenth century—except only at King's—whereas they were retained at Oxford for two centuries more.

The wall system, now most familiar, was not unknown in the Middle Ages, but the first great library to adopt it was the Escorial, completed by Philip II in 1584. Possibly Sir Thomas Bodley saw it. At any rate, when he extended his Oxford foundation at the " Arts End " in 1610-12 he put the shelves against the wall in arched recesses with a gallery above. Here, too, desks were provided and the books were chained. The same plan was adopted at the " Selden End " in 1634. St. Edmund Hall built its library in 1680, with shelves against the wall and the books chained. But chains and the wall system were found incompatible even at Oxford, when the great eighteenth-century libraries at All Souls, Christ Church and the Radcliffe " Camera " were installed. The rapid increase in the numbers of books made the old chaining system impracticable, and librarians had henceforth to trust to the honesty of readers. Canon Streeter's book, with its minute details about library planning and fitting, especially at Hereford, is of great interest and value, and hi, theory of library development must definitely replace the less securely founded theory of his eminent predecessor in this field, J. W. Clark.