Vintage canada edition, 1998 Copyright 1996 by Alberto Manguel

areolae:
The “lucrative sciences” in the second flowerbed contained only two areolae, medicine and law. The third
flowerbed was reserved for theology.
Within the areolae, each tabula was assigned a number of letters equal to the number of books held in it,
so that one letter could be given to each of the books, and recorded on the book’s cover. To avoid the
confusion of having several books identified by the same letter, de Fournival used typographical and
colour variations for each letter: one book of grammar would be identified by a capital rose-red A, another
by an uncial A that was pansy-purple.
Even though de Fournival’s library was divided into three “flowerbeds”, the tabulae were not necessarily
allocated to subcategories in order of importance, but according to the number of volumes he had
collected. Dialectics, for instance, was allotted an entire table because there were more than a dozen
books on the subject in his library; geometry and arithmetic, represented by only six books each, shared a
single table between them.
De Fournival’s garden was modelled, at least in part, on the seven liberal arts into which the traditional
medieval education system was divided: grammar, rhetoric, logic, arithmetic, geometry, astronomy and
music. Established in the early fifth century by Martianus Capella, these seven subjects were believed to
embody the entire scope of human wisdom, apart from medicine, law and theology.
A thirteenth-century Islamic library. A group of readers is consulting one of the carefully catalogued
volumes stored flat on the small shelves in the background. (photo credit 13.3)
About a century before de Fournival proposed his system, other bookish men such as the father of canon
law, Gratian, and the theologian Peter Lombard had suggested new divisions of human knowledge based
on reconsiderations of Aristotle, whose proposed universal hierarchy of existence they found deeply
appealing, but their suggestions were not taken up for many years. By the mid-thirteenth century,
however, the number of works of Aristotle that had begun to flood Europe (translated into Latin from the
Arabic, which in turn had been translated from the Greek, by such learned men as Michael Scot and
Hermannus Alemannus) obliged scholars to reconsider the division de Fournival found so natural.
Beginning in 1251, the University of Paris officially incorporated the works of Aristotle into its
curriculum. Like the librarians of Alexandria before them, the librarians of Europe sought out Aristotle.
They found him meticulously edited and annotated by Muslim scholars such as Averroës and Avicenna, his
chief Western and Eastern exponents.
Aristotle’s adoption by the Arabs begins with a dream. One night early in the ninth century, the caliph al-
Ma’mun, son of the almost legendary Harun al-Rashid, dreamed of a conversation. The caliph’s
interlocutor was a pale, blue-eyed man with a broad forehead and frowning eyebrows, sitting regally on a
throne. The man (the caliph recognized him with the assurance we all have in dreams) was Aristotle, and
the secret words that passed between them inspired the caliph to command the scholars at the Baghdad
Academy to devote their efforts from that night onwards to the translation of the Greek philosopher.
A sixteenth-century portrait of Roger Bacon. (photo credit 13.4)
Baghdad was not alone in collecting Aristotle and the other Greek classics. In Cairo, the Fatimid library
contained, before the Sunni purges of 1175, more than 1.1 million volumes, catalogued by subject. (The
Crusaders, with the exaggeration induced by astonished envy, reported that there were more than 3
million books in the infidels’ hold.) Following the Alexandrian model, the Fatimid library also included a
museum, an archive and a laboratory. Christian scholars such as John of Gorce travelled south to make
use of these invaluable resources. In Islamic Spain too there were numerous important libraries; in
Andalusia alone there were more than seventy, of which the caliphal library of Córdoba listed 400,000
volumes in the reign of al-Hakam II (961–76).
Roger Bacon, writing in the early thirteenth century, criticized the new cataloguing systems derived from
second-hand translations of the Arabic, which in his opinion contaminated Aristotle’s texts with the
teachings of Islam. An experimental scientist who had studied mathematics, astronomy and alchemy in

---

Paris, Bacon was the first European to describe in detail the manufacturing of gunpowder (which would
not be used in guns until the next century) and to suggest that, thanks to the energy of the sun, it would
one day be possible to have boats without rowers, coaches without horses and machines that could fly. He
accused scholars such as Albert the Great and Saint Thomas Aquinas of pretending to read Aristotle in
spite of their ignorance of Greek, and while he acknowledged that “something” could be learned from the
Arabic commentators (he approved, for instance, of Avicenna and, as we have seen, he assiduously
studied the works of al-Haytham), he considered it essential that readers base their opinions on the
original text.
In Bacon’s time, the seven liberal arts were allegorically placed under the protection of the Virgin Mary,
as depicted in the tympanum over the western portal of Chartres Cathedral. In order to achieve this
theological reduction, a true scholar — according to Bacon — required a thorough familiarity with science
and language; for the former the study of mathematics was indispensable, for the latter the study of
grammar. In Bacon’s cataloguing system of knowledge (which he intended to detail in a huge, never-
completed and encyclopedic Opus principale), the science of nature was a subcategory of the science of
God. With this conviction, Bacon fought for years to have the teaching of science fully recognized as part
of the university curriculum, but in 1268 the death of Pope Clement IV, who had been sympathetic to his
ideas, put an end to the plan. For the rest of his life Bacon remained unpopular with his fellow
intellectuals; several of his scientific theories were included in the Paris condemnation of 1277, and he
was imprisoned until 1292. It is believed that he died shortly afterwards, unaware that future historians
would give him the title “Doctor Mirabilis”, the Wonderful Teacher, for whom every book had a place that
was also its definition, and every possible aspect of human knowledge belonged to a scholarly category
that aptly circumscribed it.
A scribe busy at his craft, sculpted in the thirteenth century on the Western Portal of Chartres Cathedral.
(photo credit 13.5)
The categories that a reader brings to a reading, and the categories in which that reading itself is placed
— the learned social and political categories, and the physical categories into which a library is divided —
constantly modify one another in ways that appear, over the years, more or less arbitrary or more or less
imaginative. Every library is a library of preferences, and every chosen category implies an exclusion.
After the Jesuit order was dissolved in 1773, the books stored in its Brussels house were sent to the
Belgian Royal Library, which, however, had no room to accommodate them. The books were therefore
kept in a vacant Jesuit church. As the church was infested with mice, the librarians had to devise a plan to
protect the books. The secretary of the Belgian Literary Society was commissioned to select the best and
most useful books; these were placed on shelves in the centre of the nave, while all the others were left
on the floor. It was thought that the mice would gnaw their way around the edges, leaving the central
core intact.
There are even libraries whose categories do not accord with reality. The French writer Paul Masson, who
had worked as a magistrate in the French colonies, noticed that the Bibliothèque Nationale in Paris was
deficient in Latin and Italian books of the fifteenth century, and decided to remedy this by compiling a list
of appropriate books under a new category that “would save the prestige of the catalogue” — a category
that included only books whose titles he had made up. When Colette, a long-time friend, asked what was
the use of books that did not exist, Masson’s answer was an indignant “Well, I can’t be expected to think
of everything!”
A room determined by artificial categories, such as a library, suggests a logical universe, a nursery
universe in which everything has its place and is defined by it. In a celebrated story, Borges took Bacon’s
reasoning to its uttermost reach, imagining a library as vast as the universe itself. In this library (which in
actual fact multiplies to infinity the architecture of the old Buenos Aires National Library on Calle Méjico,
where Borges was the blind director) no two books are identical. Since the shelves contain all possible
combinations of the alphabet, and thus rows and rows of indecipherable gibberish, every real or
imaginable book is represented: “the detailed history of the future, the autobiographies of the archangels,
the faithful catalogue of the Library, thousands and thousands of false catalogues, the demonstration of
the falsity of these catalogues, the demonstration of the falsity of the real catalogue, the gnostic Gospel of
Basilides, the commentary on that gospel, the commentary on the commentary on that gospel, the true
account of your death, a version of every book in every language, the interpolation of every book in all the
other books, the treatise the Venerable Bede might have written (and never wrote) on Saxon mythology,
the lost books of Tacitus.” In the end, Borges’s narrator (who is also a librarian), wandering through the
exhausting corridors, imagines that the Library itself is part of another overwhelming category of
libraries, and that the almost infinite collection of books is periodically repeated throughout a bookish
eternity. “My loneliness,” he concludes, “is cheered by this elegant hope.”

---

Rooms, corridors, bookcases, shelves, filing cards and computerized catalogues assume that the subjects
on which our thoughts dwell are actual entities, and through this assumption a certain book may be lent a