That there was a time in history when books were in chains seems, at first glance, to give a very poor impression of the history of library science. Indeed, looking at the definition for chained books in Harrod’s Librarians’ Glossary, we get:
Books chained to shelves or reading desks in libraries of the fifteenth to early eighteenth centuries to prevent theft. 
While this definition is not wrong, per se, it provides a very limited view of a more complex situation. It can be shown, through consideration of the historical context, that the chaining of the books was part of the process of liberating the libraries.
Why were books chained?
Blades chronicles a rather unique reason for the chaining of a collection in Holland. It seems that the books “being all of a religious tendency, excited the animosity of the Devil, who, on several occasions, gained admittance and stole the best of them”. It was clear that this was, in fact, a visit from the dark master because “the marks of his cloven feet upon the flagstones showed plainly, not only the personality of the thief, but the very course he had taken in his sacrilegious visits”. To prevent further loss, the collection was chained and sprinkled with holy water — which seemed to have kept the remaining collection of 268 volumes intact at least until the time of Blades’ account (1892).
While this is, indeed, a compelling reason to secure a collection, more earthly reasons prevailed in most cases.
During the Middle Ages and the early years of the Renaissance, books were generally kept in bookchests known as armaria (also known as almaries or almeries). The chests were kept in a room that was usually locked, and would likely have included other valuables belonging to the institution. This room would not have been considered a library, as we understand the concept today, as it contained books, but did not allow free access to them, and certainly wouldn’t have included space for reading the books.
On the contrary, most books would have been found, at any given time, in the hands of various scholars. Loan systems were in place that allowed scholars to borrow books from the institution for a certain amount of time, often a single school year. At the end of the year, the books would have to be returned, at which point they would be inspected both for signs of actual use (to prove that the scholar hadn’t been wasting the institutions valuable resources) and for the amount of care that was taken in their preservation. Monasteries went so far as to initiate a ceremony for this transaction on the first Monday in Lent.
After Terce on the first Monday in Lent the librarian shall spread a carpet in the middle of the chapter house and place on it the books from the cupboard, and at a sign from the abbess she shall rise and sitting in the middle of the chapter house shall read out in the lists and indicate which person had which book in the past year. The reader should read slowly so that between each name there may be a pause until she who has been named shall return the book. When anyone shall hear her name called out she shall rise at once and bring her book to the carpet and if she has read the whole work she shall bow to the cross and retire. Those who have not read through their book shall prostrate themselves before the abbess and ask her pardon.
In addition, there was often a pledge system in place that required the scholars to offer goods of equal value as a guarantee towards the safe return of the books.
While this may have seemed like an altogether satisfying arrangement, it was generally ineffective for three reasons. First of all, books were extremely valuable, so the pledges for their safekeeping might have been prohibitively expensive. Consider the following appraisal.
The manuscripts were very valuable, more especially the missals which, since the lettering was necessarily large, used up a great deal of expensive parchment. They cost a fortune to buy. In 1074 a priest of Benediktbeuern received a vineyard in return for a missal and in 1120 the monastery of Baumberg was given a considerable piece of land, again in return for a missal.
Second, then as now, the books were susceptible to damage and loss. Samuel Johnson, for example, was a librarian’s nightmare. He borrowed books and used them as if they were his own, writing notes in the margins, and rarely returning them. Finally, if a scholar was interested in looking at a book, he might have to wait at least a year to gain access to it. In fact, if a student was not in favour with the right people, he might not be able to gain access to certain books at all.
Around the 14th century, institutions started to devote rooms to the storage of books alone. Often, there would be two rooms, going by various names: bibliotheca publica vs. bibliotheca secreta , magna libraria (or libraria communis) vs. parva libraria. The terminology varies, but the basic idea can be generalized to one room for books in common use (akin to modern reference sections) and another for the books in circulation.
The books in circulation were treated much like the books that were kept in chests in the older institutions. In fact, in some cases, they may have remained in their original chests, despite being given an exclusive room. They were lent out to the scholars for various lengths of time, and ordered to be returned for periodic inspections. This collection would often include duplicates and books that were, perhaps, more esoteric, and therefore not likely to be in high demand. This would generally be the larger of the two collections.
(The books in circulation at the Sorbonne were stored in the parva libraria, which means “small” library. However, this collection was almost certainly larger than the contents of the magna libraria. The word “magna” was likely originally used to mean “great”, and thus “parva” came to be used as a match for “great” rather than an indication of size.)
It is the books of the communal collection which come into play in the discussion of chaining. There is a common misperception, likely due to the restraining symbolism conveyed by the chains themselves, that the chained books were being taken out of use. On the contrary, chaining certain books and placing them in a communal library gave all students access to books that would otherwise be locked up or constantly in some individual’s possession. Therefore, as will be shown in the following paragraphs, modern readers might consider chaining a “scarcely conceivable restriction”, when it was in fact “a glimmer of liberal thought”.
Which books were chained?
There is a common misperception that the most valuable books were chained in order to guard against theft or loss. This statement is not entirely false, but clarification needs to be given to the word “valuable”. In this case, the meaning of valuable cannot be monetary. Rather, the value in the books refers to their worth as sources of educational merit. In fact, the idea of chaining financially precious books is quite incongruous, as the act of chaining tended to damage the volumes. Prized tomes would more likely be kept out of the main collection, under lock and key. (In some cases a triple-key system was in effect, so a borrower would have to get the permission of three separate officials in order to gain access to a particularly pricey item.) Johnson suggests, however, that such costly books were often sold in order to purchase more useful books.
Libraries are generally dependent on the generosity of their benefactors, whether private or public. In the past, people would often leave their book collection to a university or specific library in their wills. These benefactors would often impose strict regulations on the borrowing and use of the books that they donated. A particularly interesting case is that of Samuel Pepys who donated his collection with the understanding that it would remain in its original order, that it never be moved, and that no new books be added to it. These instructions would also often include provisions for the chaining of the books. Blades quotes the will of Sir Thomas Lyttleton (1481):
I wull and bequeth to the Abbott and Convent of Hales-Oweun a boke of myn called Catholicon to theyr own use for ever, and another boke of myn wherein is contaigned the Constitutions Provincial and De gestis Romanorum and other treatis therein, which I wull be laid and bounded with an yron chayn in some convenient parte within the said Churche at my costes, so that all priests and others may se and rede it when it plesith theym.
It is clear, in this case, that chaining was not meant as a restriction, but as a warranty of free access.
So, it seems that libraries chained books for two reasons: either to ensure access, or to comply with the wishes of a benefactor. Librarians were not trying to hoard the books by chaining them, rather they were trying to provide fair access to them. In fact, the various new inventions of the era, such as the book wheel, the stationary or rotating circular desk, and the bookbox, show that librarians were trying to increase the use of the books rather than restrict access by chaining the books.
How were books chained?
As was mentioned earlier, books were originally kept in chests. As less restrictive policies became more prevalent, three different arrangements came into existence. Streeter termed these options the lectern, stall, and wall systems. As each system represents a historical jump from the last, the choice of which system to employ would generally correspond to the era in which the library was established.
The lectern system involved chaining books to a sloped board that served to support a number of books resting on a horizontal ledge (simliar to periodical shelving in some modern libraries.) A single volume would occupy a certain space on the lectern, which, with the addition of chaining, ensured that the item could always be found in the same place — and could always be used as long as no one else is sitting in that location. Lecterns were sometimes used alone, with the books resting on the slopes (as in Zutphen) or as part of a lectern-shelf combination, with a few books resting on a flat shelf under the lectern that could be transferred to the lectern to be read (as in Lincoln, Mialatestiana). A third option was to use both the lectern and the shelf extensively (as in Laurentian) and have a number of books resting on the slope while others stayed on the shelf below. This system would have required a great deal of jostling to get at the books one needed.
The lectern system did indeed provide access to the books, but it also had some disadvantages. First of all, the chains on the books had to be quite long if the lectern-shelf combination was used. The longer the chain, the heavier its overall weight, and the more damage it would therefore do to the book. Moreover, books that were left on the sloped surface would eventually become warped from supporting their own weight. And finally, books were only accessible if no one else happened to be sitting in front of them. As Oxford students complained in 1444:
[…] should any student be poring over a single volume, as often happens, he keeps three or four others away on account of the books being chained so close together.
The stall system evolved from the lectern system. Standing perpendicular to the walls, stalls were similar to modern bookshelves, but with a board jutting out at lap level to be used as a desk. Examples of this system can be seen at Hereford, Clare College, and St. John’s (the latter two are at Cambridge). The stall system was an answer to the overcrowding of the lecterns. As more volumes were added to a collection, the lectern system would become unwieldy and unmanageable. Stalls often included at least three shelves above the board, and eventually, as more books were added to the collection, the space below the board might also be used.
Stalls were an improvement in terms of space, but they still required the books to have very long chains. In fact, the books on the top shelves would have to have even longer chains than those in the lectern system. Furthermore, the problem of use being restricted to the number of people who can fit in front of the board was not at all improved as a result of this progression.
The final stage in chaining customs was the wall system, as seen in the Bodleian. Both the lectern and stall systems involved moveable structures arranged perpendicular to the walls. In the wall system, bookshelves were built directly into the walls of the building. Again, this did nothing to alleviate the chain-length problem, but perhaps it served to give more people access to books that happened to occupy neighbouring space on the shelves.
Books in modern libraries are invariably shelved with their spines facing out. Thus, it is often assumed that chains were attached to the spines of books, and that they were shelved in the same way that we do today. However, chains were added to various places on the book’s binding, and the shelving of the books depended on that. Chaining, thus, had to be done at each individual library, to suit the specific structures that the library had in place to hold the books. Blades describes the process:
[…] it was necessary, in the first place, to buy a chain, and, if the book was of especial value, a pair of clasps; secondly, to employ a smith to put them on; and, lastly, a painter to write the name and class-mark on the fore-edge.
In the case of Winborne Minster, the chains were made by children in a local orphanage and residents of a workhouse.
As stated above, the books were not necessarily shelved with their bindings facing out. For example, in the Hereford collection which remains today, chains were added to the right hand side of the front cover (if we imagine that the book is closed and sitting in front of us). The chains were attached somewhere between the middle and the top of that edge of the book. The other end of the chain was attached to a rod at the bottom of each shelf. Because of the location of the chain on the front edge of each book, the books were placed with the binding facing the back of the shelf. The Hereford process is described in Blades:
To attach the chain, a narrow strip of flat brass is passed around the left-hand board and riveted to it in such a manner as to leave a loop in front of the edge of the board, wide enough to admit an iron ring, 1¼ inch in diameter, to which one end of the chain is fastened. The book is placed on the shelf with the fore-edge turned outwards, and the other end of the chain is fastened to a second ring rather larger than the former, which plays along an iron bar.
In fact, there were many variations on this practice. Books could be chained anywhere along their binding and the chain could be attached to a rod either above or below the position of the book. The fore-edge could face in or out, and the books could either lie flat on top of a shelf or stand upright as we generally see now. Sometimes the chains would be placed at the top of a book, but the book itself would be shelved upside down (and binding to the back) to decrease the necessary length of the chain, and therefore its overall weight. In general, care would be taken in deciding the location of the chains and the method of shelving in order to best preserve the book itself and to inflict the least amount of damage on the surrounding books.
When were the books chained and unchained?
Chaining did not begin and end at strictly defined times. Certain libraries had already been through a chaining and unchaining cycle before other libraries started to chain their books. The Sorbonne library, which appears to have started chaining its books in the late 13th century, may have been one of the first to adopt this practice. At the other end of the spectrum, many libraries abandoned this practice, but some continue to this day (although the chains remain as museum pieces more than functional tools today). The following is an overview of some of the examples of chaining and unchaining.
- Peterhouse College Library in Cambridge had 380 volumes in 1418, and 220 of them were chained.
- The library of the Faculty of Medicine in France started chaining their books in 1509 and continued for approximately 300 years.
- In Eton, books were first chained around 1519-20 and the chains were removed 200 years later.
- The first general unchaining at Cambridge took place in 1627 at Clare College, but some books were unchained as early as 1574. Cambridge libraries progressively followed Clare’s example, except for King’s College which kept its chains until late in the 18th century.
- In the 17th century, all new and refurbished libraries at Oxford were chained, but in the 18th century, all new libraries (except Lincoln) were not chained. Oxford libraries kept their chains until late in the 18th century, but that decision is said to have been motivated by security more than conservatism.
- Some chained libraries still exist today: Erfurt in Germany, Hereford in England and a number of existing chained collections are listed in Blades’ 1892 research (although it is doubtful that many remain today). The library at Wimborne Minster was chained at its inception in 1686 and it remains chained today, although a few volumes have been unchained for display.
Why were books eventually unchained?
Except for a few late stragglers, the lifespan of chaining had run its course by the late 1800s. There are a number of reasons why the practice of chaining fell out of use.
First of all, with the introduction of the printing press, the value of books dropped significantly and duplicates became more readily available. This also meant that books were available in unforeseeably large numbers, which strained the capacity of the libraries and made chaining increasingly tricky. Eventually, it became impossible to chain all books in common use.
The chains themselves were never a model of convenience. They served a purpose for the library, but they were not particularly user-friendly. The chains would often get tangled, to the point where library rules had to be made to address this problem. Scholars at King’s College, Cambridge, were given the following warning:
For the rendering his business about the library more easy, each person that makes use of any books in the said library is required to set them up again decently, without entangling the chains.
And, as mentioned above, throughout the progression from lectern to stall to wall system, a workaround for the problem of one person blocking the access of another was never successfully implemented.
Furthermore, as libraries evolved, it was noticed that more efficient use of the space could be made if the chains were removed and the shelves were redesigned. The seating did not have to be next to the books, and taller shelves could be employed.
In fact, fashion also played a part in this change as the “gentleman’s library” came into vogue in the 17th century. This kind of library was modeled after the private libraries of the wealthy, which would naturally not need chains. Once this idea caught hold, certain other libraries followed the trend, as was seen in Cambridge.
Wormald gives several compelling reasons for the end of the chaining era around the 1700s. Books were getting cheaper, smaller, and more readily available. A change in the social setting meant that rich people populated the universities, not poor, so the library evolved from a dark chamber into an upper class salon. And finally, spaciousness and display were valued over security.
While chaining enjoyed a history of approximately 500 years, it was bound to be superseded as societal developments, one by one, removed both the need and the desire to chain the collections. The chains may have prevented stealing, but they caused irreparable damage to the books. The slope of the lecterns caused the books to warp and the chains tugged on the bindings, often enough to separate the binding entirely from the contents. And even though the books were chained, the odd one still managed to go missing. In fact, in surveying several institutions in the late 1800s, Blades found that the only part left of some books was the binding or the chain.
From the preceding argument, it can be seen that chaining does not entirely deserve the negative image that it conveys. On the contrary, it is clear that chaining was an inevitable step in the evolution towards modern libraries. Universities or monasteries, for example, started off with collections of books, but if most of them had stayed locked up in chests or scattered around in the hands of a few individuals, the concept of the academic library, as we know it today, might never have been born. The need to equalize access to the books created the phenomenon of chaining. But in order to chain the books, a permanent space needed to be provided, staff had to be employed to perform the chaining and other maintenance, and a significant amount of money had to be set aside for the purchase of the requisite chaining paraphernalia. In short, chaining solidified the integrity of the academic library and gave it cause to grow. When it came time to set the books free from their chains, the practice may have seemed to embody restriction, but when they were first introduced, it is undeniable that chains brought an unprecedented amount of liberation to the libraries of their time.
1 R. Prytherch. Harrod’s Librarian’s Glossary, 8th ed. Aldershot, England: Gower. 1995. (p. 120.)
2 W. Blades. Books in Chains. London: Gale. 1892. p27.
3 ED. Johnson. Libraries in the Western World. New York: Scarecrow. 1965. p114.
4 http://www.boys.bolton.sch.uk/chainlib.htm (This source has since been removed. See: http://www.boltonschool.org/SeniorBoys/Facilities/ChainedLibrary/)
5 F. Wormald, CE. Wright. The English Library Before 1700. London: Athlone. 1958. p19.
6 RH. Rouse. The Early Library of the Sorbonne. Scriptorium. Vol. 21. 1967. p55.
7 Blades: 18.
8 Wormald: 21.
9 W. Loschburg. Historic Libraries of Europe. Leipzig: Edition Leipzig. 1974. p10
10 A. Predeek. A History of Libraries in Great Britain and North America. Chicago: ALA. 1947. p20.
11 Blades: 18.
12 Johnson: 114.
13 Rouse: 42.
14 EA. Savage. Old English Libraries. London: Gale. 1912. p109.
15 Johnson: 127-128.
16 Blades: 49.
17 Loschburg: 18. (includes illustration)
18 Johnson: 114.
19 Savage: 113. (includes photo)
20 BH. Streeter. The Chained Library: A Survey of Four Centuries in the Evolution of the English Library. New York: Burt Franklin. 1970. p1.
21 Loschburg: 34 and Streeter, 11. (both include photos)
22 Streeter: 17. (includes illustration)
23 Loschburg: 38. (includes photo)
24 Streeter: 25. (includes photo)
25 Blades: 20.
26 Streeter: 59, 65, 71 respectively. (all include photo or illustration) Hereford is also shown in Loschburg: 50.
27 Streeter: 73. (includes illustration)
28 Blades: 53.
30 Streeter: 51.
31 Blades: 53.
32 Wormald: 215.
33 Wormald: 236.
34 Rouse: 60.
35 Johnson: 125.
36 Johnson: 171.
37 Blades: 20.
38 Wormald: 216-7.
39 Wormald: 217, 237.
40 Loschburg: 11.
41 Blades: 52.
42 Blades: 25.
43 Blades: 9, 10.
44 Wormald: 238.
45 Blades: 24. (mistakes are from the original)
46 Wormald: 237-8.
47 Wormald: 216-7.
48 Wormald: 247.
49 Blades: 10, 53, 55, 59.
Bibliography and Availability
TU = Main Library, University of Tsukuba
SLIS = Library of the School of Library and Information Science, University of Tsukuba
D. Byrne. Chained Libraries. History Today, Vol. 37 1987.
JMJ. Fletcher. Chained Books in Dorset and Elsewhere. Dorset Natural History and Archaeological Society, Vol. 35 p. 8-26.
W. Blades. Books in Chains. London: Gale. 1892.
[SLIS: 289 B52]
H. Hixson. Chains of Knowledge 1686-1986: Wimborne Minster Chained Library. Wimborne, Dorset, Wimborne Minster.
ED. Johnson. Libraries in the Western World. New York: Scarecrow. 1965.
[SLIS: 010.23 J64]
W. Loschburg. Historic Libraries of Europe. Leipzig: Edition Leipzig. 1974.
[SLIS: 010.23 L89]
M. Perkin. The Parochial Libraries of the Church of England: A Revised Directory. Library History. Vol. 14. May 1998.
A. Predeek. A History of Libraries in Great Britain and North America. Chicago: ALA. 1947.
[SLIS: 010.23 P91]
RH. Rouse. The Early Library of the Sorbonne. Scriptorium. Vol. 21. 1967.
EA. Savage. Old English Libraries. London: Gale. 1912.
[SLIS 010.23 Sa92]
BH. Streeter. The Chained Library: A Survey of Four Centuries in the Evolution of the English Library. New York: Burt Franklin. 1970.
[SLIS: 010.23 St8]
J. Williams. Old Books in New Buildings: The New Library Building at Hereford Cathedral. Library History. Vol. 14. May 1998.
F. Wormald, CE. Wright. The English Library Before 1700. London: Athlone. 1958.
[SLIS: 010.23 W88]
A reader of this article wrote in July 2009 to say that there is a chained library extant (but, like the one at Hereford, treated as a museum) in Zutphen, Netherlands. He says, “It’s in the church of St Walburgis. Zutphen is a very attractive town on the river Ijsel (we visited there earlier this week), so recommended!”