The Book of Memory Page 5
complete – but which the detritus of history and linguistic change have
temporarily concealed. Fundamentalist translations are considered to be
merely restatements of an inerrant truth that is clear and non-ambiguous –
they are not adaptations or interpretive readings. Fundamentalism ideally
should produce no gloss or commentary. Thus the role of scholarship is
solely to identify the accumulations of interpretive debris and to polish up
the original, simple meaning. It is reasonable, from a fundamentalist
attitude, that God must be the direct author of the Bible. This belief
holds true as well among secular fundamentalists writing about literature,
who postulate a God-like author who plans, directs, and controls the mean-
ing of his work.
But texts need not be confined to what is written down in a document.
Where literature is valued for its social functions, works (especially certain
14
The Book of Memory
ones, of course) provide the sources of a group’s memory. Societies of this
sort are ‘‘textual communities,’’ in Stock’s phrase, whether those texts exist
among them in oral or written form. The Latin word textus comes from the
verb meaning ‘‘to weave’’ and it is in the institutionalizing of a story
through memoria that textualizing occurs. Literary works become institu-
tions as they weave a community together by providing it with shared
experience and a certain kind of language, the language of stories that can
be experienced over and over again through time and as occasion suggests.
Their meaning is thought to be implicit, hidden, polysemous, and com-
plex, requiring continuing interpretation and adaptation. Taken to an
extreme, of course, textualism can bury the original work altogether in
purely solipsistic interpretation. Beryl Smalley, who spent her scholarly life
reading medieval commentaries, once remarked wryly that ‘‘choos[ing] the
most arbitrary interpreter of Biblical texts of the Middle Ages would be
rather like awarding a prize for the ugliest statue of Queen Victoria.’’27
In the process of textualizing, the original work acquires commentary
and gloss; this activity is not regarded as something other than the text, but
is the mark of textualization itself. Textus also means ‘‘texture,’’ the layers of
meaning that attach as a text is woven into and through the historical and
institutional fabric of a society. Such ‘‘socializing’’ of literature is the work
of memoria, and this is as true of a literate as of an oral society. Whether the
words come through the sensory gateways of the eyes or the ears, they must
be processed and transformed in memory – they are made our own.
Thomas Aquinas was a highly literate man in a highly literate group, yet
his contemporaries reserved their greatest praise not for his books but for
his memory, for they understood that it was memory which allowed him to
weave together his astonishing works.
Memory also marked his superior moral character; it should not go
unnoticed that the praise heaped on his memory came at his canonization
trial. In fact, prodigious memory is almost a trope of saints’ lives. One
thinks of St. Anthony, who learned the whole Bible by heart merely from
hearing it read aloud (the fact that he never saw the words written is what
astonished his contemporaries); of St. Francis of Assisi, reputed by his
followers to have a remarkably exact and copious memory. Tropes cannot
be dismissed as ‘‘mere’’ formulas, for they indicate the values of a society
and the way in which it conceives of its literature. The choice to train one’s
memory or not, for the ancients and medievals, was not a choice dictated
by convenience: it was a matter of ethics. A person without a memory, if
such a thing could be, would be a person without moral character and, in a
basic sense, without humanity. Memoria refers not to how something is
Introduction
15
communicated, but to what happens once one has received it, to the
interactive process of familiarizing – or textualizing – which occurs between
oneself and others’ words in memory.
Many historians will wonder why I have avoided assigning Neoplatonist
or Aristotelian labels in my discussion of memorial technique and practice,
especially given the role of memory in Neoplatonist philosophy. But my
decision is deliberate. A currently accepted picture of the intellectual
history of the twelfth and thirteenth centuries is one of movement from a
Neoplatonist matter/spirit dualism, influenced profoundly by Augustine
(though not identical with his thought), to an Aristotelian hylemorphism
articulated most successfully by Thomas Aquinas. But to associate an
interest in memorial practice with one of these schools more than the
other is misleading, as I discovered early in my study of the subject. While
the Neoplatonist–Aristotelian distinction is crucial in some areas of medi-
eval culture, it is not, I think, when it comes to this one. In fact, intellectual
history, as traditionally practiced, is not the best way to go about studying
the role of memory in medieval culture.
Memoria is better considered, in the context of my study, as praxis rather
than as doxis. Practices are sometimes influenced by ideas (and vice versa) –
Chapter 4 describes one major instance of this in later medieval mnemo-
technique – but they are distinct, and follow different patterns and tempos
of change. Historians of rhetoric have sometimes described Memory as
one of the two ‘‘technical’’ parts of their subject, along with Delivery,
distinguishing it thereby from the ‘‘philosophical’’ areas of Invention,
Arrangement, and Style. This classification may well have contributed to
the impression that memoria, being merely technical, was limited in its
applicability to the conditions of oral debate, as was Delivery.28 But as the
practical technique of reading and meditation, memoria is fundamental in
medieval paideia, having even greater importance in that context than it
does as a ‘‘part’’ of rhetoric. If my study achieves nothing else, I hope it will
prevent students from ever again dismissing mnemonics and mnemotech-
nique with the adjective ‘‘mere,’’ or from assuming that memory technique
had no serious consequence just because it was useful and practical.
The historian Lawrence Stone has wisely remarked that all historical
change is relative. Within any given period, we may stress differences or
continuities. Most historians of the Middle Ages are now engaged in
detailing the differences that existed among Western peoples during that
immensely long stretch of time, geography, and linguistic and institutional
developments that we hide within the blanket designation ‘‘the Middle
Ages.’’ In this study, I stress the continuities, though I am aware of the
16
The Book of Memory
differing circumstances that separate the various scholars and poets whose
work I discuss. I am concerned with elementary assumptions and the
commonplaces which underlie the practices that are the
subject of my
study.
My method is, I hope, made legitimate by the nature of the topic I am
studying: memoria, in the rich complex of practices and values that word
acquired. It is also justified by the elementary nature of my subject, whose
training began in one’s earliest education and was basic to both reading and
composition. And while this study lays some foundations, it is in fact the
first of three. It must be complemented by a full study of how mnemo-
technique changed over the medieval centuries (I glance at this aspect of my
subject in Chapter 4). The third study would consider memoria not as a
technique but as a cultural value or ‘‘modality’’ (in the sense developed by
A. G. Greimas) of literature, and this aspect of it is touched upon especially
in Chapters 5 and 6. 29
I begin this book with an examination of two of the governing models
for the operation of memory in respect to knowledge, expressed as two
families of related metaphors: memory as a set of waxed tablets upon which
material is inscribed; and memory as a storehouse or inventory. These
models are complementary; they are also archetypal Western commonpla-
ces. The next two chapters examine the workings of memory itself.
Chapter 2 begins by considering memory’s psychosomatic nature in clas-
sical and medieval psychology, its intermediary relationship between sen-
sory information and intellectual abstraction, and its identification with
habit in the ethical realm. Chapter 2 next considers the ethical imperative
attached to memory training, and ends by examining a parallel between the
ancient memory system based upon placing images mentally in architec-
tural places and the case history of a ‘‘memory artist’’ described by the
Soviet psychologist A. R. Luria.
Chapter 3 describes several elementary schemes taught for designing a
trained memory, which utilize the principle of a rigid order into which
short pieces of material are placed and consciously supplied with a network
of associations, the aim being to provide a securely accessible ‘‘library’’ (as it
was often called) known by heart. In Chapter 4, I examine the circum-
stances in which the ancient mnemotechnique described in the Rhetorica
ad Herennium, attributed to Cicero, was revived in the scholastic setting of
the universities and by the early humanists, and examine carefully three
scholastic arts of memory that seem to show how an essentially medieval
mnemotechnique was married (somewhat awkwardly) to principles of the
ancient architectural scheme. I have provided, in appendices, English
Introduction
17
translations of three medieval texts that are not easily available now, but
that are important descriptions of various memory techniques. 30
The last part of this book turns from the theory and practice of
mnemotechnique itself to examine why it was held in such esteem. Here
I define in detail the important institutional role of memoria, first in
relation to reading and then in the context of the activity of composition.
These related discussions in Chapters 5 and 6 clarify how literature was thought to contribute to the ethical life of the individual and to the public
memory of society. Finally, in Chapter 7, I examine how the memorial
needs of readers and the memorial nature of literature affect the presenta-
tion and layout of the text in the physical book itself.
I would like to acknowledge at this point certain works whose influence
on my opinions came as I was working out the earliest parts of this study,
and is consequently more profound than may be entirely evident from my
notes: Paul Ricoeur, The Rule of Metaphor; Jacques Derrida, ‘‘White
Mythology’’; Wesley Trimpi, Muses of One Mind; Gerald Bruns,
Inventions; Richard Rorty, Philosophy and the Mirror of Nature; Clifford
Geertz, The Interpretation of Cultures; A. J. Greimas, On Meaning; Jean
Leclercq, The Love of Learning and the Desire for God; the studies of ancient
and early medieval education by H.-I. Marrou and Pierre Riche´; and Brian
Stock, The Implications of Literacy. I have raided the footnotes of many
scholarly studies, but none more fruitfully than those of Richard and Mary
Rouse. Finally, and most importantly of all, any work on artificial memory
systems must begin with the studies of Frances Yates and Paolo Rossi;
though mine range far from theirs, I could not have done without them.
My subject is multifaceted. I have tried to keep my analysis adequate to it,
though I know I have simplified some things, perhaps overly so. I must ask
for some patience from my readers, as I follow out various strands of what is,
actually, a skein. If I seem to be digressing unconscionably, I hope that they
will bear with me until we come back to the main subject, enriched in
understanding. (And perhaps some of the memory techniques described in
my early chapters will help in remembering the parts of this discussion.) For
this book can be read in at least two ways: as a history of a basic and greatly
influential practice of medieval pedagogy, and as a reflection on the psycho-
logical and social value of the institution of memoria itself, which is in many
ways the same as the institution of literature.
C H A P T E R 1
Models for the memory
T A B U L A M E M O R I A E
Readers who are familiar with a current opinion that there are radical
differences between oral culture (based upon memory) and literate culture
(based upon writing) may be puzzled by the very title of this book, and even
consider it self-contradictory. My source, however, is Dante,1 who was
newly articulating a very old observation. Even the earliest writers I discuss
did not operate within a culture that could be described as fully oral. Yet for
all these writers, memory is a central feature of knowledge – its very basis in
fact – whether through recollection (as for Plato) or as the agent building
experience (as for Aristotle). This emphasis upon memory persists, shared
by societies varying widely in the availability of books to readers: that is, in
their literacy. (I am adopting here Eric Havelock’s useful definition of
‘‘literacy’’ as coterminous with ‘‘book-acquiring public.’’)2
In none of the evidence is the act of writing itself regarded as a supplanter
of memory, not even in Plato’s Phaedrus. Rather books are themselves
memorial cues and aids, and memory is most like a book, a written page
or a wax tablet upon which something is written. Cicero writes about the
relationship of writing to memory in his elementary work, Partitiones
oratoriae:
[M]emory . . . is in a manner the twin sister of written speech [litteratura] and is
completely similar to it [persimilis], [though] in a dissimilar medium. For just as
script consists of marks indicating letters and of the material on which those marks
are imprinted, so the structure of memory, like a wax tablet, employs places [loci]
and in these gathers together [collocat] images like letters. 3
This metaphor is so ancient and so persistent in all Western cultures that it
must be
seen as a governing model or ‘‘cognitive archetype,’’ in Max Black’s
phrase. 4 In the passage most familiar to the later Middle Ages, the image is
used by Aristotle in his treatise De memoria et reminiscentia. A memory is a
mental picture (phantasm; Latin simulacrum or imago) of a sort which
18
Models for the memory
19
Aristotle defines clearly in De anima, an ‘‘appearance’’ which is inscribed in
a physical way upon that part of the body which constitutes memory. This
phantasm is the final product of the entire process of sense perception,
whether its origin be visual or auditory, tactile or olfactory. Every sort of
sense perception ends up in the form of a phantasm in memory. And how is
this ‘‘mental picture’’ produced? ‘‘The change that occurs marks [the body]
in a sort of imprint, as it were, of the sense-image, as people do who seal
things with signet-rings’ (my emphasis). 5 In this particular passage, Aristotle
uses, in addition to his usual word phantasm, the word eiko¯n or ‘‘copy,’’
which he qualifies by calling it ‘‘a sort of eiko¯n.’’ His language here derives
in turn from Plato, who uses several of the same words in his own
descriptions of what constitutes the physiological process of memory. As
Richard Sorabji notes, for Plato, too, recollection involved ‘‘the seeing of
internal pictures’’ which are imprinted upon the memory as if with signet
rings.6
The idea that the memory stores, sorts, and retrieves material through
the use of some kind of mental image was not attacked until the eighteenth
century. 7 It has recently been vigorously revived by certain cognitive
psychologists, some of whose experimental work provides startling appa-
rent corroboration of ancient observations concerning what is useful for
recollection. 8 According to the early writers, retention and retrieval are
stimulated best by visual means, and the visual form of sense perception is
what gives stability and permanence to memory storage. They do not talk
of ‘‘auditory memory’’ or ‘‘tactile memory’’ as distinct from ‘‘visual mem-
ory,’’ the way some modern psychologists do. 9 The sources of what is in
memory are diverse, but what happens to an impression or an idea once it