ered that I might confect in writing something pleasing and useful for Your
Holiness: in which you might have briefly annotated in selections for your
recollection [ob commemorationem] things which previously you had read
in many large codices completely in the stylish discourse of orators . . ..
While I was treating all these things carefully, it came into my mind that,
according to the custom of the ancients who put together many things
concerning natural history and etymologies of names and words, I might
myself compose for you a little work in which you would have written
down [something] concerning not only natural history and the literal
meanings of words, but also the spiritual sense of those same things, so
that you could find, continuously placed together, both the literal and
Memory and the ethics of reading
219
spiritual meaning of each single thing.’’78 And that is exactly what De
universo is, a web of interpretationes of the various matters relevant to
Scripture, arranged by key-words that are themselves organized not alpha-
betically but ‘‘logically’’ (starting with God and the angels). Much of it,
including the organizational key-words, is taken directly from Isidore’s
Etymologiae. But collated continuously (continuatim) with Isidore, as
Hrabanus promises, are interpretations or sensus, both litterales and mystici
of the various Scriptural items. These are most likely Hrabanus’s own
lecture notes here collated for Haimon; much of their content was also
used in the elementary Scriptural gloss compiled for beginning students in
the early twelfth century called the Glossa ordinaria.79
The purpose of Hrabanus’s compilation is not to substitute for the study
of original texts, but to provide cues for recollecting material read earlier.
This is a convention of the genre, florilegia being understood to be volumes
only of extracts, the notes which students took on their reading. And these
notes are memorative in both origin and purpose. Hrabanus describes
himself as a student carefully extracting (‘‘mihi sollicite tractandi’’) from
his studies, and the sources from which he gathers are much like those
William of Ockham drew from 500 years later, and recommends that his
students draw from in turn – books of sacred learning, moral philosophy,
and histories (but, as one would expect, Ockham’s memory gardens
included a great deal of canon law as well). Perhaps more concretely than
any other genre, the florilegium is the essential book of memory.
But why bother to compose them? Especially if one is expected to be
familiar as well with the full works from which these flowers are plucked?
The reason has everything to do with the difference between reading as
lecture and reading as meditation, meditation being an activity of dividing
and composing. We must remember that a trained and well-provided
memory was regarded throughout this long period not as a primitive
learning technique but as the essential foundation of prudence, sapientia,
ethical judgment. The choice to memorize or not isn’t like deciding to
trade in one’s old typewriter for a word processor. Composition was
considered to be a memorial activity too; a preacher making a sermon
must draw upon his memorial store-house. It is, after all, for the purpose of
composing that Ockham advises his student to stock his memory, so that
he may recall what he needs.
As a compositional and devotional aid, a florilegium is a promptbook
for memoria; that is clearly how Hrabanus Maurus understood it. Such
compilations provide the materials (and in some cases also suggest a
format) for the memoria of those who must lecture publicly and
220
The Book of Memory
extemporaneously, as Bishop Haimon needed to do in his sermons. Isidore
of Seville succinctly articulates this requirement by remarking aphoristi-
cally that ‘ Lectio requires the aid of memory.’’80 The various types of
florilegial books have been intensively studied recently, and it is clear
that those of the scholastic centuries (the thirteenth and fourteenth cen-
turies primarily) differ in format and in type of content from earlier ones,
and from the late medieval and Renaissance compilations of the humanists.
Monastic compilations are primarily meditational in nature, designed for
an audience of varying sophistication, whereas the scholastic florilegia, like
the Manipulus florum or Handful of Flowers studied by the Rouses, are
often highly technical, even jargonish textbooks for clerics composing
sermons. These arose to fill a particular market demand to which I have
already alluded: the greatly increased population of university scholars and
preachers. Many of these scholastic florilegia are organized so technically
that they would be hard to memorize as such, but their usefulness assumes
an already well-trained memory. They are thus still memorial prompt-
books, made ob commemorationem, but a derivative sub-genre for special-
ists. Indeed, as Christina von Nolcken has shown, the specialization of
preachers’ florilegia into collections of Scriptural distinctiones (or scholastic
definitions) and sermon exempla was reorganized in the fourteenth century
in a more compendious form of various sorts of material from an array of
sources. Their entry-headings were moral topics. 81 The resulting composi-
tion was one that a late twelfth-century cleric approvingly characterized as
‘‘adorned with flowers of words and sentences and supported by a copious
array of authorities. It ran backwards and forwards on its path from its
starting-point back to the same starting-point’ (my emphasis). The memorial
principle of such organization is evident, and as von Nolcken comments,
these features remain impressively constant. 82
In a careful study of pre-scholastic florilegia, Birger Munk Olsen notes
that monastic collections follow a number of different principles of order,
sometimes even in different sections of the same compilation.83 These
range from clear ordering schemes by moral topic, some even alphabeti-
cally arranged by topic or incipit, to those in which extracts are arranged by
author, to ones showing no discernible order at all, including those ‘‘mini-
florilèges’’ which are nothing but short pieces copied onto margins or the
blank pages of manuscripts. Extracts themselves range in length from single
lines to lengthy passages. The audience intended is sometimes beginning
students; several address pueri in their prefaces, but (as we saw with
Hrabanus) adult scholars are mainly addressed. This fact again confirms
their nature as meditational prompts or sources for a memory already
Memory and the ethics of reading
221
basically trained, rather than as a primary-school text. The compiler of
Florilegium Duacense says that, because there are so many books and human
memory is so yielding (‘‘madida memoria’’), ‘‘from a multitude of books I
have plucked a few [matters] among a great many, which seemed to me
most useful and necessary’’; these one can read and meditate and
learn by
heart, for they are kept ‘‘non scriniis aut armariis, sed archa pectoris,’’ ‘‘not
in treasuries nor in bookcases but in the chest of memory.’’84 By these
extracts, the needy soul is cured, the healthy preserved, the weary recreated,
the hungry fed, the simple soul may be instructed, the scholarly one
aroused, the impoverished one gain something worth noting down.
It is interesting that rigidly structured compilations exist side by side
with those that seem totally nonstructured, for it confirms again that the
form itself of a work like this cannot be trusted as an indication of whether
or not its contents were to be memorized, nor in what manner. An
unorganized compilation could hardly be used unless it were to cue an
already-formed memoria, readers slipping the material into their own
heuristic schemes, as they had been taught to do. But collections that
come with their own organization were also designed to stock or cue the
memories of their users, as is apparent from such a late, vernacular,
Dominican product as Bartolomeo’s Ammaestramenti. Moreover, it is
not clear always that the compilers themselves distinguished their audien-
ces carefully; the author of Florilegium Duacense clearly thought his readers
would include both adults and beginners.
Florilegia have retained their immense popularity until the present. The
justification for their use goes back to ancient pedagogy, and is well
described by Quintilian. Once a child learning to read progressed from
syllables to whole words, his writing exercises (which kept pace with his
reading) consisted of writing out words, especially hard words, on the
ancient principle that one should always concentrate on what is difficult,
for then the mastery of what is easier is also accomplished. He was also set
to copy lines, which he memorized for reciting during this task, repeating
them in a murmur while he wrote. 85 Because memory was so intimately
and constantly involved, Quintilian urges that ‘‘the lines set for copying
should not be meaningless sentences, but convey some moral lesson. The
memory of such things stays with us till we are old, and the impression thus
made on the unformed mind will be good for the character also. The child
may also be allowed to learn, as a game, the sayings of famous men and
especially selected passages from the poets (which children particularly like
to know). Memory is very necessary to the orator.’’86 Such an exercise, in
other words, is not useful except as these bits of reading are memorized, or
222
The Book of Memory
as they recall to mind matter previously stored away. In providing the
maxims for discourse, they provided the materials of character too.
Texts such as Cato’s Distichs provided such elementary material, as did
the Psalms.87 Quintilian’s advice is echoed by Julius Victor, and we see the
same practice centuries later in the teaching of Bernard of Chartres,
described by John of Salisbury:
Bernard used also to admonish his students that stories and poems should be read
thoroughly, and not as though the reader were put to flight like a spurred horse.
Wherefore he always insistently demanded from each one, as a daily debt, some-
thing committed to memory. 88
Cassiodorus provides one of the best medieval statements of the virtue of
acquiring such material, the topica as he calls them, stored in memory.
‘‘Clearly [memoria is] a remarkable sort of work – that in one place could
be gathered together whatever the mobility and variety of the human mind
was able to learn by inquiring about the sensible world through various
postulates – it contains the free and willful mind; for wherever it turns itself,
whatever thoughts it enters into, of necessity the human mind falls into some
one of those common places earlier mentioned.’ 89 The commonplaces are
understood here to be habits of thought, habits of character as well, the hexis
or firma facilitas, complete mastery of subject and self, that Quintilian
understood hexis to mean. One cannot think at all, at least about the
world of process and matter (‘‘in sensibus . . . per diversas causas’ ), except
in commonplaces, which are, as it were, concentrated ‘‘rich’’ schemata of the
memory, to be used for making judgments and forming opinions and ideas.
This is really no more astonishing than saying that one cannot think
without opinions, or without categories. But it is true that the nature of
those opinions for medieval students was more consciously literary in
origin than is the nature of most of our received opinions, which we like
to think are factual instead (some people I know even claim to have factual
opinions). And this memorized chorus of voices, this everpresent florile-
gium built up plank by plank continuously through one’s lifetime, formed
not only one’s opinions but one’s moral character as well. Character indeed
results from one’s experience, but that includes the experiences of others,
often epitomized in ethical commonplaces, and made one’s own by con-
stant recollection.
Abelard relates the following famous and instructive anecdote about
Heloise:
I admit that it was shame and confusion in my remorse and misery rather than any
devout wish for conversion which brought me to seek shelter in a monastery
Memory and the ethics of reading
223
cloister. Heloise had already agreed to take the veil in obedience to my wishes
and entered a convent. So we both put on the religious habit, I in the abbey of
St. Denis and she in the convent of Argenteuil which I spoke of before. There were
many people, I remember, who in pity for her youth tried to dissuade her from
submitting to the yoke of monastic rule as a penance too hard to bear, but all in
vain; she broke out as best she could through her tears and sobs into Cornelia’s
famous lament:
O noble husband,
Too great for me to wed, was it my fate
To bend that lofty head? What prompted me
To marry you and bring about your fall?
Now claim your due, and see me gladly pay . . .
So saying she hurried to the altar, quickly took up the veil blessed by the bishop
and publicly bound herself to the religious life.90 (My emphasis)
Heloise quotes from Lucan’s poem, Pharsalia, the verses with which
Pompey’s wife, Cornelia, greets her husband after his shameful defeat in
battle, offering to kill herself in sacrifice to placate the gods. In his study of
the episode, R. W. Southern states the parallel: as Cornelia offers herself to
death to save her husband at the moment of his greatest shame, so Heloise
sacrifices her life that Abelard might overcome his shame, for ‘‘[l]ong
before Abelard had seen himself as the modern Jerome, Heloise had seen
herself as the modern Cornelia.’’91 It was a natural thing for her to do,
because these lines from Lucan were in her memory, they helped to make
up her experience. Since Heloise read in the medieval way Hugh of
St. Victor describes, she did not see herself as Cornelia in the sense of
assuming or acti
ng a role. Rather, Cornelia’s experience, given voice by
Lucan, had been made hers as well – so much her own that she can use it,
even, perhaps, with irony, in such an extreme personal situation.
Literary fragments of florilegial length, dicta et facta memorabilia, are a
frequent (though to our eyes peculiar and somewhat embarrassing) accom-
paniment to medieval moral decisions. Augustine, at the very moment of
his greatest personal anguish, thinks immediately of the hermit Anthony,
as he hears the mysterious command Tolle, lege, and picks up the Book to
read whatever his eye chances on. 92 One sometimes gets the impression
that a medieval person, like Chaucer’s Dorigen, could do nothing (espe-
cially in duress) without rehearsing a whole series of exemplary stories, the
material of their experience built up board by board in memory, and, as
Gregory says, transformed into their very selves, so that even in moments of
stress the counsel of experience will constrain a turbulent and willful mind.
If the task is properly done, as Cassiodorus says, the mind of habitual
224
The Book of Memory
necessity (a meaning of hexis) will have a ‘‘place’’ for anything it may
encounter, whatever the circumstances.
A modern woman would be very uncomfortable to think that she was
facing the world with a self constructed out of bits and pieces of great
authors of the past, yet I think in large part that is exactly what a medieval
self or character was. Saying this does not, I think, exclude a conception of
individuality, for every person had domesticated and familiarized these
communes loci, these pieces of the public memory. It does underscore the
profound degree to which memory was considered to be the prerequisite
for character itself. The link is suggested in the fact that Greek charakteˆr
means literally ‘‘the mark engraved or stamped’’ on a coin or seal; by
transference, the word came to mean ‘‘distinctive mark’’ and hence the
‘‘distinctive quality’’ of a person or thing, and ultimately also ‘‘type’’ (of
person) or ‘‘style.’’93 The word’s literal meaning continued to be recog-
nized, however; in Orator, Cicero translates it with the Latin word forma,
which had a similar range of meanings.94 One basic conception of a
The Book of Memory Page 40