letters of the Latin alphabet in various combinations. This latter is con-
structed from a vellum disk in which holes have been cut, its center pierced,
and a thong tied through it to the back of the leaf on which it is placed,
allowing it to move freely. Hugh de Fouilloy’s wheels imply move-
ment, most obviously those of his treatise on True and False Religious,
which depict the rising and declining patterns of the lives of a good and a
false monk.
One medieval figure who is important in the history of memory design and
arts, but about whom I have had little to say because he has been thoroughly
studied by Frances Yates and R. J. Hillgarth, is the late thirteenth-century
Spaniard, Ramoń Lull (1235–1316; the last version of his art was written
1305–1308). 91 Lull’s art was designed to be both a key to universal concepts
and a meditational memory art; his evangelical motivation, to preach
persuasively in multilingual Spain, is significant in comprehending the
importance he gave to inventing successful methods for imagines rerum.
Indeed the missionary spirit remained the significant motivator as well for
those friars and Jesuits who taught memory schemes as part of their
evangelism in Mexico, China, and elsewhere.92 Though Lull did have
some influence in the late Middle Ages, he is more important in the context
of early modern figures like Giordano Bruno and Camillo, with whom
Yates is most occupied. Lull’s is an extremely subtle, complex, and learned
system, not at all for beginners. Nor is it, as Yates points out, at all like the
system of loci and imagines described in Ad Herennium. Yet it did not come
out of nowhere. Yates relates it to the Neoplatonism of Scotus Eriugena.
But many of its characteristic uses of symbol and figure have much
commoner origins.
Frances Yates attributes two apparent innovations to Lull. First, he
‘ designates the concepts used in his art by a letter notation, which introduces
332
The Book of Memory
an almost algebraic or scientifically abstract note into Lullism.’’ And
secondly, he uses diagrams extensively, such as concentric circles, rotating
triangles within a circle, ladders (steps), and trees, to ‘‘introduc[e] move-
ment into memory.’’93 But diagrams such as these, which all imply the sort
of movement and manipulation of which Yates speaks, were a common
feature of the medieval elementary classroom, precisely for the purpose
of memory training. So was the use of letter notation to organize con-
cepts. 94 Hugh of St.Victor’s De archa Noe is a universal diagram, designed
to organize a large amount of disparate information in a readily avail-
able, mnemonically effective, way. This fact suggests that such diagram-
matic machinae universitatis were encouraged in medieval pedagogy well
before Lull.
For mnemonic purposes, diagrams, like other sorts of images in medi-
eval books, have a combination of two functions: they serve as fixes for
memory storage, and as cues to start the recollective process. The one
function is pedagogical, in which the diagram serves as an informational
schematic; the other is meditational and compositional. The functions
were more elegantly formulated in the ancient and medieval advice to
students, that memoria consists in divisio and compositio. There are a
number of diagram-like pages in liturgical and devotional books through-
out the Middle Ages, whose function is of this latter sort. In his classic
study of a much-copied sequence of such theological diagrams from the
late Middle Ages, the Speculum theologiae, Fritz Saxl commented: ‘‘A wealth
of wisdom is displayed, which slowly reveals itself to the patient reader who
does not mind the absence of a well-defined general lay-out . . . Each
picture . . . must be pondered again and again.’’95 Such diagrams are like
the seventeeth-century emblems – indeed, they may well be their medieval
predecessors. They compose a place for a meditative recollection by picturing
various theological and devotional themes. Their obscurity and partialness
are deliberate; as deliberate as is the clarity of a pedagogical diagram such as
the Genealogia.
The most easily accessible now of these late medieval diagram-encyclopedias
is the De Lisle Psalter (British Library MS. Arundel 83-II), all published in a
volume edited by L. F. Sandler. Only the drawings now exist, but in the
original book they preceded a psalter made for an English layman, Robert
de Lisle. These drawings, made in the early fourteenth century, are copies
of a group of pictures devised towards the end of the thirteenth century by a
Franciscan friar working in Paris, John of Metz. The whole group was the
Speculum theologiae or, in England, The Orchard of Consolation, names
which suggest both the encyclopedic intention of their maker and their
Memory and the book
333
contemplative purpose. (Recall Gregory the Great’s admonition that, in
our true reading, we see as in a mirror, speculum, our own vices and virtues.)
The word ‘‘orchard’’ was used in the title of other medieval English
vernacular works of devotion, such as The Orchard of Syon; it is a variant
on the ancient ‘‘flowers of reading’’ (florilegium) trope, these being the
fruits of meditational memoria presented as materials for yet further
meditation. Three of the diagrams in the Speculum theologiae are known
to have originated from texts, as pictures, imagines, of their res: a ‘‘Tree of
Life,’’ a ‘‘Tree of Vices and Virtues,’’ and a ‘‘Cherub,’’ the diagram-picture
which is the content of a sermon-treatise attributed in the Patrologia Latina
to Alan of Lille: On the Six Wings (of the Cherub [or Seraph]). 96
These theological pictures, textually derived, are realizations on parch-
ment of the kind of meditative imagining for compositional purposes that
we encounter as well in Hugh of St. Victor’s De archa Noe. They often have
more in common with mnemonic rebuses than with the kind of schematic
to which we now restrict the word ‘‘diagram.’’ But they are rebuses of a very
elaborate type. An ordinary rebus is an image-for-the-word, like the rebus
of ‘‘Morton,’’ which I described earlier in this section. These, however, are
true imagines rerum, designed to call to mind the framework and contents
of a composition, which each individual should ponder and elaborate
further. They provide places for memorial gathering, collatio, in the
manner which Hugh de Fouilloy provided for his treatise On the Dove
and the Hawk.
The spectacular pages of some of the earliest books of the Middle Ages,
the Insular gospel books designed for lectionary use and study, can be
considered in this way. The carpet-pages of interlace that grace such
Gospels as the Book of Durrow (late seventh century) or the Book of
Kells (around about 800?; figure 30) are most like diagrams in the way in
which they treat space. For example, Jacques Guilmain has demonstrated
the simple, orderly geometry upon which the forms of ornament in these
carpet pages
rests. As he comments, the making of these pages ‘‘is an art that
cannot be described simply as a catalogue of its component parts, for just as
significant is the syntax of those parts . . . all details relate to the wholes as
completed fabrics.’’97 Moreover, they are not pages which one can easily
digest; like the texts they introduce they must be looked at and looked at
again, ruminated, absorbed and made one’s own. The figures that peek
through the interlace are not apparent until one looks long enough to begin
putting together what seems at first fragmentary. It is a process such as the
one Hugh of St. Victor describes as the journey from ignorance to con-
templation; one first sees only an overwhelming jumble of fragmentary
334
The Book of Memory
30. Dublin, Trinity College MS.A.1 (formerly MS.58 ‘‘The Books of Kells’’; Ireland or
North Britain (?Iona), about 800?), fo. 34r. The first page of the Gospel of Matthew (the
text is Mt. 1:18: ‘‘Christi hic generatio’’), showing ‘‘Chi-rho-i’’ in large letters. Two
sedentary cats and several mice, two of which are nibbling a communion wafer, can
be made out at the right of the bottom of ‘‘Chi.’’ To the right of them, at the base
of the stem of ‘‘rho,’’ is a black otter eating a salmon.
Memory and the book
335
detail, then as one meditates one begins to collect the pieces, and then in
contemplation forges a meaningful pattern.
In a famous description, which some scholars think may be that of the
Book of Kells itself, the twelfth-century historian, Gerald of Wales,
recounts his encounter with such a book. It is remarkable not so much
because of its apparently realistic depiction of the artistry of one of these
Insular manuscripts as because of its articulation of a process of seeing,
reading, and meditation that, Gerald says, fulfilled his ordinary expect-
ations to an extraordinary degree:
Look at [the forms in this book] superficially with an ordinary casual glance, and
you would think it is an erasure, and not tracery. Fine craftsmanship is all about
you, but you might not notice it. Look more keenly at it and you will penetrate to
the very shrine of art. You will make out intricacies so delicate and subtle, so exact
and compact, so full of knots and links, with colors so fresh and vivid, that you
might say that all this was the work of an angel and not of a man. For my part, the
oftener I see the book, and the more carefully I study it, the more I am lost in ever
fresh amazement, and I see more and more wonders in the book.98
What Gerald describes is not an act of picturing or illustration, but an
act of reading, in the monastic understanding of what reading is. It is
basically a rhetorical practice, an act of memoria in which the figures
grouped in the picture are designed both to recall and to stimulate further
mental image-making in the reader. Pages that consist of a framework
alone, without super-imposed images, are especially suggestive. An exam-
ple is the carpet-pages of the Book of Durrow. Here empty rectangular
panels are set in a framework of interlace, some panels having a plain-
colored background, some a simple interlace design, like the empty places
into which particular imagines rerum may be projected, and then erased
and used again for other occasions and other texts – and by a variety of
people, as is fitting for a book made to be pondered.99
The ornamentation of a medieval page does not consist of images to be
memorized precisely. Instead, they are presented as examples and invita-
tions to the further making of such images. Nothing could prevent a lazy
user from employing such cues only to trigger rote memory. This was
recognized as praying only with the lips, and excoriated as a form of sloth.
But for a serious reader, one willing to become mindful of reading and
ponder it, like Mary, in her heart, all memory advice is clear that one
should not rely on ready-made images, one should learn to fashion one’s
own, for only this exercise will concentrate the mind enough to ensure a
safe investigation of one’s memory. A true memory-image is a mental
creation, and it has the elaboration and flexibility, the ability to store and
336
The Book of Memory
sort large amounts of information, that no pictured diagram can possibly
approach.
There is a built-in indeterminacy of meaning, and even of relationship of
parts, to medieval diagrams, for they follow the logic of recollection –
which is associative and determined by individual habit – and not the
universal logic of mathematics. Like the tituli, the rubrics, and the punc-
tuation, the picture-diagrams are a part of the apparatus of a text – aiding
its mnemonic divisio, surely, but deliberately inviting meditation, compo-
sitio, as well, the recollective process by means of which a particular reader
engages a particular text (with all that includes) on a particular occasion.
The idea that manuscript decoration had a practical use is now broadly
adopted by codicologists and art historians; as Christopher de Hamel has
observed, ‘‘Decoration is a device to help a reader use a manuscript.’’100
But, as I hope this chapter has shown, if we take a wholly utilitarian
approach to decoration, especially if we identify it with some pre-conceived
notion of literate as opposed to oral culture, we will misunderstand its full
function as much as we did when we thought of it as only decoration, or as
a help for those who couldn’t read. Like reading, of which it is a part,
decoration is practical in the medieval understanding of that word, having
a basic role to play in every reader’s moral life and character because of its
role in the requirements of memory practices.
Every medieval diagram is an open-ended one; in the manner of exam-
ples, it is an invitation to elaborate and recompose, not a prescriptive
schematic. Hugh of St. Victor did not think that he had produced a
model of what the Ark was really like, whether at the literal, allegorical,
or moral levels; this is apparent from the way in which he freely adds to the
Genesis account of its lineaments and freely contradicts his own previous
interpretations. Once again, we can see how his whole attitude towards a
text differs from ours; it isn’t, to him, a definitive statement of fact or
experience but an occasion for rumination and meditation, for the engage-
ment of memoria. Invitations to meditate further are found throughout his
Ark pictura. After describing his elaborately painted ladders, for example,
Hugh remarks, ‘‘There are many other things that could have been said
about these [figures] that we must skip over here out of necessity.’’101 But
the reader may stay to meditate and contemplate and learn from the
examples which Hugh has provided.
The rhetorical indeterminacy of a medieval diagram extends as well to all
the elements that ‘‘distinguish’’ a medieval page. Iconography, in art as well
as literary criticism, treats images as direct signs of some thing, as having an
inherent meaning that will be
universal for all readers. But, like the painted
Memory and the book
337
letters, all the other decorative elements are signs that act directly only
upon memory. In the memory of a particular reader they will become
meaningful as I make them mine. And memories, as we have seen, are
differently stored, having different tracks and associative paths. The one
thing that a manuscript image must produce in order to stimulate memory
is an emotion. It must be aesthetic in the ancient sense of this word. It must
create a strong response – of what sort is less important – in order to
impress the user’s memory and start off a recollective chain. That combi-
nation of image and response makes up the memory-image, and only then,
when the fully formed image is in memory, can it become a matter of
thought.
Let me take as a last example of this distinction the famous black otter
eating a salmon that is tucked into the bottom border of the rho on the Chi-
rho page at the beginning of Matthew in the Book of Kells (figure 30). It is
very hard to find amid the myriad and apparently fragmentary forms on
this page. But as I suggested earlier, the page is designed to make one
meditate upon it, to look and look again, and remake its patterns oneself;
the process of seeing this page models the process of meditative reading
which the text it introduces will require. The letters on this page are
virtually hidden away in the welter of its other forms; indeed, thinking of
this aspect of its design, Franc¸oise Henry calls it ‘‘a sort of rebus’’ (182). Yet
it is not a true rebus, as are the visual puns suggested for remembering
proper names by the author of Dialexeis. Nor is the Chi-rho page by any
means a diagram-picture for the gospel of Matthew, even to the extent that
De formatione arche is for De archa Noe. It helps instead to initiate the
divisional and compositional process that is required to read Matthew, and
it is this that makes it valuable to memory. In that process, the discovery of
the successful otter, and – next to it – the two cats surrounded by mice they
are too lazy to catch, surprises us with a shock of delight, whether or not we
know the proverbs about lazy cats, or care to make a meditational link
between industrious fishermen and cats who won’t work for their food. 102
The Book of Memory Page 57