unitary discipline with a full methodology. This study concentrated on
the essentials and was suspicious of what seemed to be shortcuts. Are there
any precepts? Charlemagne asks Alcuin. None, is the reply, only discipline,
training, and exercise. In this exercise, the importance of notae, or signa as
they are also called, was universally acknowledged, including imagines of
every kind, any of which could be used for mnemonic associations. What
makes an image a mnemonic is not its nature but how it is used. Any image
used as a mnemonic, by virtue of that use, is classified as a mnemonic
image. These can – and did – include all kinds. It is important to note that
none of these early medieval writers expresses any hostility towards using
mental images for memory work – their motives are not iconoclastic or
doctrinal, even when iconoclastic controversies were current.51 And they
are not prescriptive concerning what kinds of images to use. Their principle
seems rather to have been to use ‘‘whatever works.’’
Boncompagno da Signa, writing in 1235, discusses no systematic art of
recollection, no principles for constructing mnemonic notae. Instead he
lists virtually everything made as a potential memory aid, for somebody in
some circumstances. Every sort of sign, starting with language itself, helps
us to remember something. Indeed, for Boncompagno, any interpretative
activity is essentially a matter of recalling something to memory. So, the
cock that crowed when Peter betrayed Jesus was a memorial sign (memo-
riale signum) by which Peter recalled Jesus’ prophetic words. All books, all
pictures, images, sculptures, all cruciforms, all insignia of rank and station,
banners, alphabets, methods of calculation, notches cut in sticks to record
loans and repayments (tallies), the stories told to children which record the
events of history, even the jargon of thieves – everything has a memorative
function by which God reminds us of Himself and we remember the world
we experience. Boncompagno’s version of Neoplatonism, which is evident,
is less interesting in this context than his notably eclectic list of mnemonic
images. He has a philosophy of memory (of a sort) but he has no ars
The arts of memory
185
memorativa, no principles on which to explain how these diverse kinds of
signa work. 52
Another specialized term, in addition to nota and signum, that defines a
function of images used memoratively, is the phrase imagines rerum. All
these terms belong to the common pedagogy of memory and are not
specific to a version of an ars memorativa. They are distinguished by
function, nota being reserved for any sort of mental marker one uses to
file and cue stored material. Imagines rerum traditionally seem restricted to
the function of public speaking, declamation, and hence composition.
Though no admirer of memory arts, Quintilian does favor making imag-
ines rerum for one’s subject matters when preparing a speech. Eloquence,
he writes, depends upon an orator’s emotional state. The mind of the
speaker must be emotionally as well as intellecually engaged.
Quintilian uses the term again when he discusses in greater detail how
emotions are stirred in the orator. To do so one is much helped by vivid
images, ‘‘which the Greeks call fantasiai . . . whereby images of things
absent are again presented in our minds’’ (Inst. orat., VI, 2, 29). These
imagines can generate the very emotions in the orator which he seeks to
awaken in his audience, and cause him to re-experience what happened:
shall I not see the assassin burst suddenly from his hiding-place, the victim
tremble, cry for help, beg for mercy, or turn to run? Shall I not see the fatal
blow delivered and the stricken body fall? Will not the blood, the deathly pallor,
the groan of agony, the death-rattle, be indelibly impressed upon my mind? 53
But imagines rerum are also critical for retaining the order and flow of a
speech – a mnemonic function. Unless one can speak with order, copious-
ness, and style, one is not speaking but only ranting. To produce such
ordered pattern:
those vivid conceptions (rerum imagines) of which I spoke, and which, as I
remarked, are called fantasiai, together with everything that we intend to say,
the persons and questions involved, and the hopes and fears to which they give
rise, must be kept clearly before our eyes and admitted to our emotions. 54
It is the imagines that can be fixed by the mind’s eye and that arouse again
both one’s intent and and one’s procedure (intentio) concerning one’s subject
matters, so these are the key to holding any discourse in our memories. One
is not trying to store away entire written texts through their agency (even if
one were so unskilled an orator as to have written one’s complete speech
down in advance), for words alone are easily lost and cannot be reconsti-
tuted, if one were to stumble or need to depart suddenly from one’s
prepared theme. Martianus Capella advises the use of vivid imagines
186
The Book of Memory
rerum in the service of what Bradwardine much later called memoria
orationis, to help remember the chief subjects of one’s speech (the veiled
bride, the sword).
Bradwardine also advocates using imagines rerum, vivid images which
cue the main points one will develop extemporaneously in one’s talk. These
imagines rerum in their memory places act as focal images for the speaker’s
meditation. The places order the major stages of the speech; the imagines
rerum in each location (supplemented, perhaps, by particular words
remembered through imagines verborum) trigger the memory through
their associational power to recall various parts of one’s mental library.
Imagines rerum can act as compositional sites and cues that can ‘‘gather in’’
(re-colligere) much related material laid down elsewhere in memory,
because they invite the orator’s mental eyes to stay and contemplate.
This function is also emphasized in medieval advice about the work of
memory. It accounts for the particularly medieval use of the verb remember
to describe what one was doing when one meditated in vivid picture-form
on hell and heaven, two places one could never have visited oneself and
thus could not actually remember, in our sense. Yet meditation, as we will
see in the next two chapters, was thought to be a particular activity of
memory. The use of imagines rerum as sites for memorial composition
(compositio) was analyzed as a kind of remembering. Medieval diagrams as
images of subject matters require one to stay and ponder, to fill in missing
connections, to add to the material which they present. They are the
instruments and machines of thought. 55
T E A C H I N G A N A R T O F M E M O R Y I N U N I V E R S I T I E S
Memory became a ‘‘speculative’’ art, given serious attention in universities,
about the year 1250, while continuing as always to be an elementary skill to
be mastered by all who claimed to read. This distinction is important to
keep in mind when considering the history of any ‘�
��art of memory.’’ Of the
three writers whose mnemotechnique I discussed earlier in this chapter,
only John of Garland was a teacher of rhetoric. Albertus Magnus was a
professor of theology and logic; so was Thomas Bradwardine. Yet it was
John of Garland who produced an old-fashioned pragmatic of mnemonic
advice, despite using some of the terms of the Ad Herennium. These he
would have learned from masters who used the ‘‘Alanus’’ gloss, which
attempts only to explain the ancient treatise’s hard words and to paraphrase
some of its sentences. Ten or fifteen years after John of Garland, in the
same university setting though in a different faculty, Albertus Magnus
The arts of memory
187
wrote his sympathetic and philosophically systematic commentary on the
Ad Herennium’s precepts for an art of memory based upon images in
places, giving it a seriousness which it did not lose again until well after
the Renaissance. It seems, thus, that university rhetoric teachers had little
role in the academic development of medieval arts of memory.
Commenting on the intellectual situation of memoria at the end of the
twelfth century, G. R. Evans remarked that ‘‘[t]he elements of a formal art
of memory and of a formal study of the faculty of memory, based on older
authorities, are all present in the twelfth century schools . . . What was
missing was a stimulus – perhaps in the form of an individual scholar with a
special interest in the topic – to bring the elements together.’’56 We may
safely conclude that this individual scholar was Albertus Magnus. Though
we know very little about his early life, it seems probable that he studied at
Padua during the first decade of its foundation as a scholium around 1222.
Padua was founded by scholars from Bologna – indeed the influence of
Boncompagno da Signa on its rhetoric curriculum was strong from the
start. But Padua was also noted as an early center for the study of Aristotle’s
libri naturales, including the brief essays of the Parva naturalia, and if
Albertus was there, it was the natural history books that he studied, rather
than rhetoric. 57
Three commentaries on the Herennian memory art were made before
that of Albertus Magnus (though it should always be noted that Albertus’s
comments on the Ad Herennium advice are made in the context of a treatise
on ethics and a commentary on Aristotle). The earliest is that in a gloss
known as Etsi cum Tullius, likely the work of William of Champeaux,
made in Paris early in the twelfth century. It was followed a decade or so
later by a commentary of Thierry of Chartres. Towards the end of the
century, perhaps in the 1180s, a third commentary was made, the work of a
‘‘magister Alanus,’’ who may or may not have been the famous teacher and
Latin poet, Alan of Lille. This became the standard text for the thirteenth
century and later. All of these commentaries were Parisian in provenance,
and reflect the teaching of rhetorical memoria in Paris. The first commen-
tary specifically on the Ad Herennium made after the twelfth century is an
early fourteenth-century Italian one, possibly from Bologna, called, from
its introductory words, Plena et perfecta.58
The only one of the twelfth-century commentaries to be published in a
modern edition is that of Thierry of Chartres, composed in the 1130s, a full
century and more before Albertus Magnus’s commentary on Aristotle’s De
memoria. Thierry was a teacher at Chartres, and later chancellor, in the first
half of the twelfth century, a contemporary of Hugh of St. Victor and of
188
The Book of Memory
John of Salisbury’s teacher, Bernard of Chartres. His interest in rhetoric is
primarily in its organization of topics; he is best known for his studies in the
Old Logic of Aristotle and in Plato. His commentary on the Rhetorica ad
Herennium is preserved only in a single manuscript of the twelfth century.
It has been edited by K. M. Fredborg, probably its first publication since
then.
What Thierry has to say about Tullius’ mnemonic techniques ‘‘hardly
goes beyond mere paraphrase,’’ as Fredborg notes, and only one of his
remarks about the memory section (an explanation of the characters
Domitius and the Martii Reges which figure in the example given by
Tullius for memoria verborum) is picked up by any of the writers who
immediately followed him and used his comments in their own work. 59
Like his contemporaries, then, Thierry seems to have been little interested
in the Herennian mnemonic, or in contributing to its revival as a practical
tool for oratory, nor does he appear to have understood significantly better
than they did what it was all about. He comments on it because it is part of
Tullius’ text, paraphrasing almost exactly what it already says, glossing
words, explicating grammar, explaining the mythological allusions.
Nonetheless, Thierry’s commentary is not without interest, because of
the intelligence of his comments and his generally positive stance towards
his subject. He exhibits none of the outright scorn for the Herennian
memory scheme that we find in John of Salisbury and Geoffrey of
Vinsauf, though they both wrote a good deal later in the twelfth century
than he did. He expands the skepticism, apparent in the Ad Herennium
itself, concerning the utility of memory for words – this sort of thing, he
says, is useful as an exercise only, and to remember verses from the poets. 60
He then rehearses in paraphrase the example in the Ad Herennium of how
to remember the verse ‘‘Iam domi ultionem [sic] reges Atridae parant.’’ And
he adds an example of his own: if you should want to remember the first
two lines of Book II of the Aeneid (‘‘Conticuere omnes intentique ora
tenebant / cum pater Aeneas lecto sic orsus ab alto est [sic]’’), you might
imagine someone sitting on a couch (‘‘sedentem in lecto’’) and reciting
something, while surrounded by many intent listeners (‘‘multos intentos ad
audiendum’’).61 This is a perfectly sensible image – rather more sensible
than those given as examples in the Ad Herennium – but it does not, in fact,
exemplify the advice of Tullius to construct visual homophonies that will
cue a line for you word-by-word. It is much more like an image ad res than
an image ad verbum in the manner of the Ad Herennium.
Yet the vocabulary of the ancient mnemonic advice does not seem
wholly foreign to Thierry. He gives a sympathetic and comprehensive
The arts of memory
189
explanation of imagines rerum and their value to orators, of the reasons
why memory craft can benefit natural talent, and of the need for practice
and disciplined exercise in making sets of numbered locations. He is clear
about what the locations are and how the images are fitted into them,
why the locations remain stable while one changes the images to suit the
needs of each particular orational cause. He understands the reason for
having a fixed order among the locations, so that one can go backwards as
well as forwards, or in whatev
er way one wishes. He knows that intervalla
refer to locorum distancias, the distance in the mind’s eye from which one
views the locations, for one’s view of the content will be confounded if
one’s view-point is situated too close or too far away. 62 He understands that
grotesque, unusual, wonderful images excite the memory, and need to
be used instead of ordinary or routine ones. And he approves the advice
in Tullius against relying entirely on ready-made images instead of
finding one’s own. It is only in regard to memoria verborum that he
seems dubious, and in this he is in the company of Quintilian and the
anti-Sophist Roman tradition.
Yet Thierry’s commentary seems to have had virtually no influence on
either his contemporaries or later writers – certainly Albertus Magnus was
not influenced by Thierry’s clear explanation of that wretchedly difficult
crux concerning Aesopus, Cimber, Iphigenia, and the sons of Atreus. Had
he known Thierry, Albertus might not have given such a muddled gloss of
this passage. Albertus and the other later memory writers got their under-
standing of memory locations, images for things, and images for words
in other ways, most likely from the commentary of ‘‘magister Alanus,’’ who
was muddled in just the places that Albertus was. The fact that Thierry is so
clear about most of the Ad Herennium’s memory terms and descriptions
of technique is of greater significance than the few moments of hesitation
he displays, for it supports the conclusion that many of the basic concepts
of memory advice from late antiquity remained current in the twelfth
century – current enough for an exceptionally intelligent commentator,
like Thierry of Chartres, to explain the gist of the Ad Herennium on
memory with understanding, if without enthusiasm.
M E M O R Y A N D D I A L E C T I C I N T H E V E R B A L A R T S
The reassessment of memorial art was given a crucial intellectual impetus
by the translation into Latin of Aristotle’s De anima and its related treatises,
including the one on memory and recollection. This material was first
translated by James of Venice (the Old Translation), and had begun to
190
The Book of Memory
circulate widely by about 1200.63 Albertus commented on it during the
The Book of Memory Page 34
