purely oral exchange and something that has achieved the finish which
makes it worthy of the scribe’s ministrations. In short, it has the status of a
dictamen, something Hugh felt free to alter even as he wrote it out.
At the end of Book II of this work, Hugh introduces a verbal device in
the form of a tree, the arbor sapientiae, which is an avatar of the first lignum
uitae and grows in holy hearts as in an invisible Paradise, embodying
allegorically both the Tree of Knowledge in Eden and the Tree of Life in
Revelation. The figure is introduced with a set of characteristic epithets, a
series of fifteen rhyming phrases of three words each (one is actually six
words made up of two apposite phrases).66 Having given the list, Hugh
announces to his audience that he and they should rest for a little while
because his sermon has gone on to the point of tedium.
He takes the matter up again in Book III. He will expand, he says, each
item in the series announced at the end of Book II: ‘‘[h]ic ipsius incrementi
gradus, quos ibi breuiter et summatim perstrinximus, latius per singula
prosequendo explanamus.’’67 The phrase breuiter et summatim reflects also
Hugh’s advice on memory in Didascalicon (III. 11); this third book of
De archa Noe may be considered an application of what Hugh says there.
Each summary phrase is expanded as a chapter in Book III, in the same
Memory and authority
259
order announced at the end of Book II. So the first phrase of the passage in
I I . xvi, describing the tree ‘ per timorem seminatur’ (‘‘planted by fear’’) is the
first subject developed in Book III: ‘ First there it is said of wisdom that it is
planted by fear’ (alluding to Prov. 1:7, ‘ the fear of the Lord is the beginning
of wisdom’’). The second phrase, ‘‘per gratiam rigatur’ (‘‘watered by grace’’),
is the subject of the next chapter, and so on, in unviolated order. The order
of the phrases themselves follows that of the growth of the tree and its
fruit: planting, watering, the seed sometimes being sterile, sometimes root-
ing, germinating, opening, growing, strengthening, greening, leafing and
branching, flowering, fruiting, ripening, being harvested, and finally eaten.
This compositional structure is mnemonic. The orderly stages (gradus)
in the growth of a tree provide Hugh’s heuristic as he composes. To each
step, Hugh has attached a Biblical quotation about wisdom, usually a verse
or two defining the word sapientia as a virtue or quality (timor Domini,
gratia, dolor, fides, etc.). So the basic mnemonic order is the stages of tree-
growth to which the thematic texts on wisdom are linked. Each growth-
stage with its primary text is the subject of one chapter, and is stated in a
rubric at the start. Within this essential structure, a number of excursive
topics are developed from a phrase or word of the rubric; these may bring
in other linked texts. Basically the structure is that of a concordance, or
catena, in which the parts are associated by key-words, each of which pulls
other texts and sayings with it, ‘‘compounding with interest,’’ as Hugh’s
Chronicle Preface promises. With a composition planned on such complex
chains of materials, it is easy to understand why the metaphor of fishing
came to be commonly associated with memory work.
We know that Book III is extempore because Hugh tells us so at its end.
He apologizes that the whole thing has been a digression from his real
subject, which was the building of the arca sapientiae. He says that it
ballooned (or blossomed) from his discussion in the second Book, and
that he had not planned it:
But now, while we have been following out the by-ways of our exposition, we have
digressed a long way from what we had proposed. Wherefore for this also we beg
your indulgence, because, as I truly confess, most often in this treatise we have
invented many more things while writing than we have written down having
already worked them out. So likewise in this matter I blush to confess my fool-
ishness. Now, however, we will continue by returning to our plan, concerning the
making of the ark of wisdom. 68
Despite his assurance to us in the Prologue that the version he is setting
before us is, though informal, a written composition, Hugh includes this
lengthy extemporaneous digression, for which he apologizes with red face
260
The Book of Memory
as being far from his initial proposal. Are we to conclude from this that
Hugh was as prone to spontaneous digressions when he was writing as we
would expect a speaker to be? And that the heuristic structures of his
memory-library are as apt to show (or be disguised) in a written compo-
sition as in a spoken one? It is evident that we must.
A great deal has been made of what is called the oral style of medieval
sermons, and its supposed differences from written style or authorial style.
Oral style, in this theory, is characterized by repetition, verbal formulas,
digressions, especially of a colloquial or informal kind, and parataxis above
all. Written style, by contrast, is hypotactic and periodic in the Latin
manner, marked by subordination and sub-divisions; it contains longer
and more unusual words, is nonrepetitive, and self-consciously artful. This
distinction has been raised in this century to the status of a truism in literary
analysis, but, unlike many truisms, this one isn’t true. It rests upon a
genuine tautology, which causatively associates the stylistic features of a
particular text whose compositional conditions are known with its method
of composition; these features are then used to demonstrate that the text
was composed in a particular way.69
What can we deduce from style alone about the methods by which a
work was composed? Nothing at all. Medieval writers extended the classical
canons of stylistic decorum by applying them not just to content and genre
but to types of audience. Thus a sermon preached to the people would
require a popular style in order to be understood, while one preached to a
learned audience would require a more evidently formal, grand style. But
medieval writers did not associate the levels of style with compositional
methods. Hugh clearly had no real objection, despite his blushes, to leaving
his meditation on arbor sapientiae as Book III of his written text; indeed, if
he did not confess it himself, there is no way from its style that one could
tell he was composing extempore, so little does the third book differ in
terms of sentence structures, vocabulary, complexity and artfulness of
expression, from the others. The composition of De archa Noe, according
to what Hugh tells us, involved at certain points all three of the composi-
tional methods described by Quintilian: it began as a premeditated collatio
with an orderly praepositum (proposal), it was written with a stylus, and it
was also at times ex tempore dicendi. Moreover, it is typical that medieval
composition involved all of these methods. 70
Eadmer’s account of Anselm tells us some interesting things about the
status or authorization
of the various products of the compositional stages.
The written product of Anselm’s lengthy and arduous cogitation (which
we have already examined) is called by Eadmer a res. This Anselm wrote
Memory and authority
261
onto tablets as soon as he had finished his completed design. Eadmer tells
us that Anselm gave these wax tablets to the brothers for safe-keeping. After
a few days, he asked for them again but they could not be found. So,
‘‘Anselm wrote another draft on the same subject [aliud de eadem materia
dictamen] on other tablets, and handed them to the same monk for more
careful keeping.’’71 But these tablets were found thrown on the floor, their
wax broken and scattered. The monks collected the pieces and brought
them to Anselm, who was able to piece the wax together and with difficulty
(vix) recover the writing. ‘‘Fearing now that by some carelessness it might
be altogether lost, he ordered it, in the name of the Lord, to be copied onto
parchment [pergamenae jubat tradi].’’72
Eadmer calls what was on the first set of tablets res and the second
dictamen. Anselm’s res, when lost, is easily recovered by him from memory;
he asks for the tablets after some time has elapsed, finds they are missing,
but is untroubled by their loss. But the second set, the dictamen, he pieces
together with difficulty, and decides to have copied onto parchment at
once to prevent their loss. This suggests that the dictamen was a much
revised, expanded, and polished version of the res, which cannot be as
readily reconstructed. It would also seem that Anselm, in the eleventh
century, was more inclined to use a stylus and wax tablets for composition
than was Thomas Aquinas in the thirteenth. I mention this as a caution
against interpreting the differences in their techniques as influenced by
something more than individual choice.
Eadmer tells these stories of Anselm’s difficulties in composing
Proslogion not to talk about his compositional methods – for to him
there was nothing noteworthy in them – but as moral tales, to indicate
that Anselm’s work was not taken seriously at first by his fellows, when it
was in its pre-exemplary state. The long time that Proslogion spent on wax
tablets before it went onto parchment is meant to convey this under-
valuing, as also is the astonishing carelessness of the monks entrusted
with it. (Only the devil – if he were responsible for the breakage – seems
to have understood its significance.) But on parchment the text will
become safe – it is worth noticing that the motive for making the exemplar
is safety.
This initiates another stage in the full composition of Proslogion.
Parchment support makes the text not only safe but public. This decision
by Anselm was made, Eadmer says, ‘‘livore carens’’, ungrudgingly, invoking
a venerable commonplace that attributes the publication of a work to the
author’s generosity, humble spirit, and freedom from envy. It is interesting
that the sin resisted is envy. We moderns with our firmly held assumptions
262
The Book of Memory
about intellectual ownership and the consequent possibility of intellectual
theft, might be inclined to attribute non-publication to avarice, miser-
liness. Attributing envy, however, suggests that the motive was thought to
be pure malice, an act against society itself. So an author who does not
share his work and launch it, as it were, into the stream of literature, is
thought to be guilty of a sin against community. The last step of authorship
is to overcome such envious feelings and submit one’s work to the com-
munal process of authorization through public comment and readerly
response.73
The end of Eadmer’s story of the finishing of Proslogion comes with
Gaunilo’s criticism of its arguments, and Anselm’s further response, both
of which he incorporated into the text itself:
This work came into the hands of someone who found fault with one of the
arguments in it, judging it to be unsound. In an attempt to refute it he wrote a
treatise against it and attached this to the end of Anselm’s work. A friend sent this
to Anselm who read it with pleasure, expressed his thanks to his critic and wrote
his reply to the criticism. He had this reply attached to the treatise which had been
sent to him, and returned it to the friend from whom it had come, desiring him
and others who might deign to have his little book to write out at the end of it the
criticism of his argument and his own reply to the criticism. 74
Anselm’s humility is exemplary, but, unlike Eadmer, that is not what I find
interesting in this account. In the first place, we note that an unknown
reader, Gaunilo, composed a criticism of Anselm’s text which he simply
wrote out as though it were a continuation of it. Moreover, he is not the
one to send it to Anselm; a friend of Anselm’s does this, sending a copy of
the text with Gaunilo’s addition which has come into his possession. This
behavior would get a modern reader a stiff fine, if not jail, but Eadmer finds
it unremarkable, not reproving Gaunilo’s manner of publication in the
least. And indeed scholars familiar with medieval readers’ habits of simply
adding material as they choose to texts by someone else will not be
surprised either. It is a mark that one’s work has been truly read, and
made his own, by someone else, and this in turn is another way of
indicating that it is gaining authority – in the word’s basic meaning of
‘‘growth’’ – as it generates further texts.
In this history of the Proslogion, we can observe the initial stages of its
socialization, as it were, as it enters a public memory-bank and becomes
literature. In order for the authority of the original work to come fully into
being, it is necessary that there be a Gaunilo, as it also requires Anselm’s
reply to him – and the many comments, pro and con, that it has generated
since. For a text demonstrates its authority not by closing down debate but
Memory and authority
263
by accumulating it – that is what we learn from Eadmer’s tale about what
constituted the originality of a text for a medieval public. Notice that
Anselm, welcoming Gaunilo’s response (there is less of humility in this
than joy at being finally taken seriously in a way his own careless brothers
had not done), adds his own comments to Gaunilo’s, and instructs that
both sets of comments be incorporated as the end of the text in all
subsequent copying. His behavior suggests that once his work was made
public (in an exemplar) Anselm saw himself as a co-equal reader along with
other readers like Gaunilo, adding to the readerly flow that keeps the work
alive and original in its proper sense. A modern author, responding to
critics, will relegate them to the footnotes or to selective quotation; and we
expect this because, to us, his authority attaches to him personally, he is
author. But the way Anselm treats it, authority is a property of the work,
proven by its ability to generate other work. Once the original exists, he
places himself with Gaunilo as equally its reader, and what he thinks of the
merits of Gaunilo’s comments is given the status simply of more commen-
tary, further dialogue.
This same observation concerning Anselm’s attitude towards his written
work is made by Brian Stock in his fine analysis of Anselm’s description of
the way he came to compose the Monologion, which began and was carried
out in a way like that of Hugh of St. Victor’s De archa Noe. 75 Stock
emphasizes two matters: first, that Anselm regards his written version as
the result of a process that is reductive and not complete; secondly, that the
text both results from and remains the focal point of a dialogue Anselm has
in his own mind (in cogitatio), with his clerical brethren (Monologion began
as colloquia which his brethren requested him to write down), and finally
with ‘‘a putative reading public.’’76 In the case of Gaunilo’s response to
Proslogion, that public was putative no longer, and it is both significant and
as typical as anything can be of the Middle Ages that Anselm immediately
invited it, in writing, into his text. If his first written exemplum is ‘‘reduc-
tive’’ (in Stock’s terms), it is so only as it offers the essentials for subsequent
dialogue – indeed, it requires dialogue in order to achieve its proper textual
function. Anselm says that Monologion began ‘‘in familiar conversation’’
(‘‘sermones colloquendo’’) with his brother monks, thus, like Hugh,
emphasizing the communal talk that started and will sustain his work.
This is not just a monkish peculiarity, though the ideal of such a dialogue-
text would be well realized in such a setting. The principle behind it is
fundamental in the rhetorical tradition of text-making. As a composition,
the written exemplum is expansive; it offers a ‘‘common place’’ which
collects subsequent comments, glosses, references, as readers apply, adapt,
264
The Book of Memory
restate, meditate upon it. Truly it is commentary and imitation which make
a text an auctor – not the activities of its writer but of its readers. This has
little to do, I think, with literacy, but everything to do with the institutional
nature of literature in a memorial culture.
The author’s dictamen, whether scribally transcribed or not, was thus a
The Book of Memory Page 47