MCK RhetoricInMiddleAges

Reproduced with permission of the University of Chicago Press.
For personal, noncommercial use only.
SPECU
A JOURNAL OF MEDIAEVAL STUDIES
VOL. XVII JANUARY, 1942 No. 1
RHETORIC IN THE MIDDLE AGES1

BY RICHARD McKEON
MEDIAEVAL and Renaissance rhetoricians and philosophers, following the ex-
ample of Cicero, seldom omit from their treatment of rhetoric some consideration
of the subject matter, nature, and end of the art. Long before Cicero rhetoric
had become one of the focal points of the differences of philosophic schools, and
the practice and application of the art had long wandered from field to field,
reflecting, and sometimes even affecting, the complexities of philosophic dis-
cussions. Yet rhetoric is treated as a simple verbal discipline, in histories which
touch upon it, as the art of speaking well, applied either as it was in Rome to
forensic oratory and associated with the interpretation of laws or, more fre-
quently, applied as it was in the Renaissance in the interpretation and use of
the works of orators and poets, and associated with or even indistinguishable
from poetic and literary criticism. The history of rhetoric as it has been written
since the Renaissance is therefore in part the distressing record of the obtuseness
of writers who failed to study the classics and to apply rhetoric to literature, and
in part the monotonous enumeration of doctrines, or preferably sentences, re-
peated from Cicero or commentators on Cicero. Scholarly labors have recon-
structed only a brief and equivocal history for rhetoric during the Middle Ages.
The development consists of slight and unoriginal increments of erudition in the
compendia composed from the fourth to the ninth century - which were derived
largely from the De Inventione of Cicero and the Ad Herennium - and in later
commentaries and treatises until in the twelfth century they reflect and use
doctrines from Quintilian and from the later rhetorical works of Cicero, the
Orator,the De Oratore,and the Topica. The sequence of development is fortuitous
and even implausible, for the treatment of rhetoric becomes more perfunctory
as erudition in the works of rhetoricians increases, and rhetoric disappears
abruptly when knowledge of it is at a maximum, particularly from the works
of the authors who acknowledge the influence of Cicero and Quintilian. The
1 By way of experimental departure from the customary procedure at meetings of learned societies
the following paper will be the subject of discussion at the next meeting of the Mediaeval Academy
on April 24, 1942. Rhetoric was chosen as a topic which impinges on many fields of mediaeval study,
and an effort is made in the paper to touch, at least, on as many of them as possible. The paper will
not be read at the meeting but will be considered in informal panel discussion in which it is hoped all
members present at the meeting will participate.
1
2 Rhetoricin the MiddleAges
translation of the Rhetoric of Aristotle, the pseudo-Aristotelian Rhetorica ad
Alexandrum, and the De Elocutione of Demetrius in the thirteenth century
would seem to have had, by this account, no effect comparable to that of the
other translations of the century in stimulating interest in its subject; and the
return of rhetoric to prominence during the Renaissance is explained only on
the supposition that men's minds were turned once more, after a long interval,
to literature and life.'
1 Valla, Vives, Ramus, and other Renaissance rhetoricians who treat the
history of rhetoric pass over
the intermediate period to criticize, refute, and occasionally approve of the doctrines of Aristotle,
Cicero, Quintilian, and Boethius. In early works of erudition and philology the scope of the history of
rhetoric is no broader than the scope of controversy. D. G. Morhof makes the transition from Cicero,
Quintilian and their predecessors, who are considered in the first nine of the thirty-two paragraphs
headed De ScriptoribusRhetoricis in his Polyhistor, Literarius, Philosophicus et Practicus (Lib. vi, cap.
1 [3rd ed., Lubecae, 1732], I, 941-956), to the Renaissance rhetoricians treated in the last twenty-three
paragraphs with the remark, 'Nos vero, missis nunc veteribus, ad recentiores sparsim enumerandos
progredimur.' J. Clericus carries the Historia Rhetorica down to the Church Fathers in his Ars Critica
(Pars II, Sectio I, cap. 17 [Leipzig, 1713], I, 336-352). The history of rhetoric has more recently been
extended to the Middle Ages, but it is always rhetoric in some particular sense, applied to some par-
ticular subject, and the history is usually negative or at least deprecatory. J. B. L. Crevier thus traces
the history of rhetoric in education by noting the absence of any provision for rhetoric in the regula-
tions of the University of Paris until the restoration of letters (Histoire de l'Universite de Paris [Paris,
1761], I, 299, 307,376, 479; II, 450; iv, 190, 243-44, 249,330, 349, and passim). The pattern of rhetoric
had, incidentally, not changed from the Renaissance to the eighteenth century in the important re-
spect that Crevier found little use in his own writings on rhetoric for any authors between the ancients
and his contemporaries, and the imperfections of Aristotle, Cicero, and Quintilian are his excuse for
writing: 'Aristote me paroit trop philosophe, Ciceron trop orateur, Quintilien trop scholastique'
(RhetoriqueFrangoise [Paris, 1808], I, xix). E. Norden treats rhetoric primarily in terms of style and
is able therefore to dispose of the entire period from the ninth century to the time of Petrarch briefly
in terms of the opposition of the study of authors to the study of the liberal arts, of classicism to scho-
lasticism (Die Antike Kunstprosa vom vi. Jahrhundert v. Chr. bis in die Zeit der Renaissance [4th ed.
Leipzig, 1923], II, 688-731); cf. the treatment of rhetoric and poetic (ibid., 894-898). According to C. S.
Baldwin the fate of rhetoric is determined by shifts in the interrelations of the arts of the trivium:
rhetoric was dominant until the fall of Rome, grammar during the Carolingian period, dialectic during
the Middle Ages (Medieval Rhetoric and Poetic [New York, 1928], p. 151). Rhetoric was crowded in
mediaeval education between grammar for boys and dialectic for men, and Baldwin is therefore at
pains to find reasons which explain 'why there was no medieval rhetorician who really advanced the
study' (ibid., p. 182). The history of rhetoric during the Middle Ages is consequently the acount of
its misapplications and extensions: poetic is a misapplication of rhetoric to style (ibid., pp. 185 ff.,
esp. 191-195); the dictamen is a development of rhetoric, but without need of perversion (ibid., pp.
208 ff., esp. 214-215); and preaching in the absence of political and forensic oratory makes use of
epideictic or occasional oratory, the third of Aristotle's genera (ibid., pp. 229 ff.). According to P.
Abelson (The Seven Liberal Arts, A Study in Medieval Culture [New York, 1906], pp. 5Q ff.) rhetoric
consisted of a practical training during the Roman period, then it consisted of the technical rules of
a science, and finally, when this theoretical and logical form of rhetoric fell into obsolescence, of the
practical rules for writing letters and documents. In the account of N. Valois (Guillaume d'Auvergne
[Paris, 1880], pp. 224 ff.) rhetoric was taught as a liberal art until the end of the twelfth century and
then fell into discredit except as a practical discipline applied to preaching and prayer. The judgment
of C. H. Haskins (The Renaissance of the Twelfth Century [Cambridge, Mass., 1928], p. 138) is no less
concise in statement: 'Ancient rhetoric was concerned with oratory, mediaeval rhetoric chiefly with
letter-writing,' and is illustrated with detailed evidence. More simply, if rhetoric is viewed as a form of
literary criticism and associated with poetic, the decline of rhetoric is a symptom of the eclipse of the
Rhetoricin the Middle Ages 3
There is little reflection in the histories of rhetoric of the differences concerning

the subject matter and purpose of rhetoric by which rhetoricians thought to
distinguish and oppose their doctrines, and only occasionally and opaquely do
some of the consequences of basic philosophic differences appear in the place
given to rhetoric in the enumerations and classifications of the arts and sciences.
The theoretic presuppositions which underlie the shifts and alterations of rhetori-
cal doctrines are readily made to seem verbal and arbitrary preferences, for in
the course of discussion all of the terms are altered in meanings, and the contents
and methods of each of the arts are transformed when grammar, rhetoric, poetic,
dialectic, and logic change places or are identified one with another, or are
distinguished from each other, or are subsumed one under another. Yet the con-
fident readjustments of Renaissance rhetoricians, their redistribution of techni-
cal devices among the arts, and their correction of the confusions of the ancients
seem no less whimsical and haphazard, if their reasons and criteria are ignored
in the repetition of enumerations of the disciplines and their parts. Rhetoricians
from Cicero to Ramus have in common a persistent care in defining their art,
and it seems plausible that a history of rhetoric traced in terms of its matter and
function, as successively specified, might follow the sense of altering definitions,
the differentiation of various conceptions of rhetoric itself, and the spread of the
devices of rhetoric to subject matters far from those ordinarily ascribed to it.
Such a history would not treat an art determined to a fixed subject matter (so
conceived rhetoric is usually found to have little or no history, despite much talk
about rhetoric and even more use of it, during the Middle Ages) nor on the other
hand would it treat an art determined arbitrarily and variously by its place in
classifications of the sciences (so conceived the whole scheme and philosophy of
the sciences would be uncontrolled in their alterations and therefore empty). The
history of rhetoric should have as subject an art which, although it has no special
subject matter according to most rhetoricians, nonetheless must be discussed in
application to some subject matter: rhetoric is applied to many incommensurate
subject matters; it borrows devices from other arts and its technical terms and
methods become, without trace of their origin, parts of other arts and sciences;
its own devices may be bent back upon themselves in such a way that any part
of rhetoric or any matter incidentally involved in it - words and style, char-
acter and passion, reason and imagination, the kinds of orations, civil philosophy,
practical action - may become basic to the definition of all technical terms and
distinctions. Moreover, if the succession of subject matters and functions can be
used to reduce the welter of changes in rhetoric to a significant historical se-
quence, the theories implicated in the shifts of its subject matter will emerge,
not merely as philosophic or sophistic disputes, but in concrete application, each
at least defensible and each a challenge to the conception of intellectual history
as the simple record of the development of a body of knowledge by more or less
adequate investigations of a constant subject matter.
study of ancient literature: cf. L. J. Paetow, 'The Arts Course at Mediaeval Universities with Special
Reference to Grammar and Rhetoric,' University of Illinois Studies, inI (January, 1910), esp. pp. 67 ff.,
and D. L. Clark, Rhetoric and Poetry in the Renaissance (New York, 1922), pp. 43 ff.
4 Rhetoricin the Middle Ages
I
Three distinct lines of intellectual development during the Middle Ages were
decisively determined or strongly influenced in their initial stages by rhetoric:
first, and most properly, the tradition of rhetoricians themselves who found their
problems assembled and typical answers discussed in the works of Cicero and
Quintilian; second, and less obviously, the tradition of philosophers and theo-
logians who found in Augustine a Platonism reconstructed from the Academic
and Neoplatonic philosophies (conscientiously reversing the process by which
they were derived from Plato's doctrines) and formulated in terms refurbished
and simplified from Cicero's rhetorical distinctions; and finally, the tradition of
logic which passed as 'Aristotelian' yet which followed Aristotle only in the
treatment of terms and propositions, and Cicero in the treatment of definitions
and principles. Whatever the estimate critics and historians are disposed to make
of Cicero's achievement, originality, and consistency, his choices and emphases
fixed the influence and oriented the interpretation of ancient thought, Greek as
well as Latin, at the beginning of the Middle Ages and again in the Renaissance,
and we today are far from having freed ourselves from the consequences of that
long tradition in scholarship, criticism, or taste. During the Middle Ages and
Renaissance many of the oppositions and agreements of theology and dialectic,
no less than problems internal to each, are stated in language borrowed from or
influenced by rhetoric, and reflect theories by which rhetoricians had in antiquity
opposed philosophers and logicians; surprising parallels arise in them as well as
in other arts and sciences, expressed in language familiar to the rhetorician;
innovations and discoveries are made which seem to follow the dictation of nature
if their pattern of statement is ignored; and mere equivocations are pursued into
interminable and recurrent verbal disputes.
The rhetoricians of the Middle Ages followed Cicero or suggestions found in
his works when they discussed civil philosophy as the subject matter of rhetoric,
or divided that subject matter according to the three kinds of oratory - delibera-
tive, judicial, demonstrative - or when they sought to determine it more gener-
ally by means of the distinction between propositum and causa (or thesis and hy-
pothesisas the Greek terms were Latinized), or by consideration of the characteris-
tics of controversies and the constitutions (or status) of questions.' Moreover, they
could learn, even from the De Inventione, that there had been controversy on
most of these points, and in particular the brief history of three views concerning
the matter of rhetoric - Gorgias holding that it is all things, Aristotle dividing
it into three kinds proper to the three kinds of oratory, and Hermagoras dis-
tinguishing causes, which are specific to persons, and questions, which are with-
out such specification - supplied the arguments by which to dissent from, as
well as those to support, Cicero's version of Artistotle's solution.2 Major altera-
1 Cf. Cicero, De Inventione i. 4.5.: 'Sed antequam de praeceptis oratoriis dicimus, videtur dicendum
de genere ipsius artis, de officio, de fine, de materia, de partibus.' After determining that its genus is
'civilis scientia,' its officium 'dicere adposite ad persuasionem,' and its finis 'persuadere dictione,'
Cicero defines the matter of all arts, including the art of rhetoric (ibid. 5. 7.): 'Materiam artis eam
dicimus, in qua omnis ars et ea facultas, quae conficitur ex arte, versatur.'
2 Ibid. 5. 7.-7. 9. Cf. ibid. 9. 12. for illustration of the
process by which basic terms are altered and
tions in the contents and doctrines of rhetoric follow on these differences in

matter particularly when they are joined to a little erudition, such as might be
derived from study of the points of difference between the Ad Herennium and
the De Inventione, or from the information supplied by Fortunatianus, concern-
ing figures and the Greek technical terms of rhetoric, or finally from Quintilian's
orderly enumerations of divergent views and his statement and rectification of
inconsistencies attributed to Cicero.1 Even apart from the influence of theology,
and before the influence of dialectic was felt, the remnants of controversial dif-
ferences were preserved in rhetoric itself.
The influence of rhetoric on Augustine was by reaction and assimilation; he
differentiated two eloquences and two arts, much as Plato had proved rhetoric
to be a pseudo art in the Gorgias and yet had illustrated the method of the true
rhetoric based on dialectic in the Phaedrus. Augustine was first attracted to
philosophy by Cicero's Hortensius which he encountered in the course of his
rhetorical studies, and he was put off in his further attempt to combine philoso-
phy with the name of Christ by the contrast of the Scriptural and Ciceronian
styles.2 That stumbling block was finally removed in part by the aid of a rhetori-
cal device which he learned from Ambrose's preaching - the analogical method
of interpreting Scripture3-and although thereafter he refused to answer
questions concerning Cicero's Orator and De Oratore,on the grounds that it was
a task unworthy of a bishop distracted with ecclesiastical cares,4 his statement
of Christian doctrine was in the terms of Cicero sublimated to new meanings
and transformed to new uses. When he wishes to enumerate the questions basic
to all inquiry, he resorts to Cicero's three 'constitutions of causes' - whether a
thing is, what it is, and what sort; and when he enumerates the methods to be
used in treating Scriptural questions, they turn out to be two of Cicero's five parts
of rhetoric - discovery and statement; moreover, these two sets of questions
seem to him exhaustive and all problems and doctrines turn, as in the manuals
the distinctions of rhetoric are bent back on themselves: in this case the dispute is concerning
whether deliberation and demonstration are genera of causes or themselves parts of a particular genus
of cause. Isidore's list of the 'inventors' of the art reflects the influence of Cicero's history of matter
since the inventors are clearly determined by this history as is the testimony to the elusiveness of
the distinctions; cf. Etymologiae ii. B.: 'Haec autem disciplina a Graecis inventa est, a Gorgia, Aris-
totele, Hermagora, et translata in Latinum a Tullio videlicet et Quintiliano, sed ita copiose, ita varie,
ut eam lectori admirari in promptu sit, conprehendere inpossibile. Nam membranis retentis quasi
adhaerescit memoriae series dictionis, ac mox repositis recordatio omnis elabitur.'
1 Cf. Institutio Oratoriaiii. 5. 4. ff. for an excellent statement of the problems involved in rhetorical
'questions,' and the disputes concerning thesis and hypothesis, and esp. 14-15 for the development of
Cicero's doctrine. For a brief summary of some of the characteristic statements of the definition and
end of rhetoric cf. ibid. ii. 15; for disputes concerning its matter, ibid. 21. Or again, in illustration of
the bending back of rhetorical distinctions, what one man holds to be the 'parts of rhetoric' another
treats as the 'work of the orator' (ibid. iii. 3. 11. ff.); the two positions are taken respectively by Cicero,
De Inventione i. 7. 9, and Fortunatianus, Ars Rhetorica i. 1 (Halm, Rhetores Latini Minores - hence-
forth cited Halm - [Leipzig, 1863], p. 81).
2
Confessions iii. 3. 6-5. 9; Patrologia Latina (henceforth cited PL) 32, 685-686.
3 Ibid. v. 13. 23 and vi. 4. 5-6; PL 32, 717 and 721-722. Cf. also the conversion of Victorinus the
rhetorician and the effect of salus andfides on his rhetoric, ibid. viii. 2. 5; PL 32, 751.
4 Epistola CXVIII ad Dioscorum i. 2 and v. 34; PL 33, 432-433 and 448.
of rhetoric, on 'things' or on 'signs.'l This rhetorical language has, however, been

adapted to the statement of a theology: discovery has been qualified as discovery
of 'what should be understood' and statement as statement of 'what has been
understood,' with the result that the classification of signs and their uses is
dependent, as it had not been in rhetoric, on the classification of things. In
the De Doctrina Christiana the first three books are concerned with discovery, the
fourth with statement. The treatment of discovery requires, in the first book,
the distinction of things into those which as final ends are loved or enjoyed (frui)
and those which as intermediate ends are used (uti) for further ends, and, under
the former head, a theological inquiry into the attributes of God and divine
things; treatment of the nature of things is supplemented, in the second book,
by a philological inquiry into the nature of words as that problem bears on the
Scriptures and the arts and institutions of the pagans; and finally, in the third
book, the inquiry into means of removing verbal ambiguities requires appeal to
two sets of rules, grammatical rules applied to the manner of statement, and
rhetorical rules to determine the circumstances of fact.2 The treatment of state-
ment in the final book is therefore concerned not so much with the precepts of
rhetoric, although some precepts can be found from analysis of the fashion in
which the three styles of Cicero are applied to their appropriate matters by
'ecclesiastical orators,' as with an eloquence in which the words are supplied by
the things and by wisdom itself and the speaker is unlearnedly wise.3 The judg-
ment expressed by Cicero at the beginning of the De Inventione that wisdom with-
out eloquence is of little benefit to the state, and eloquence without wisdom a
great danger, is transformed, when Augustine quotes it, by a dialectical doubling
of all the terms. The wisdom and eloquence of the world are to be contrasted to
eternal wisdom and eloquence, for not only are there two kinds of things, tem-
poral and divine, but two kinds of words, the external words instituted and used
by men, which have no correspondence to things except by designation and no
1 Confessions x. 9. 16-10. 17; PL 32, 786. Cf. Cicero, Orator 14. 45: 'Nam quoniam, quicquid est
quod in controversia aut in contentione versetur, in eo aut sitne aut quid sit aut quale sit quaeritur:
sitne, signis; quid sit, definitionibus; quale sit, recti pravique partibus - quibus ut uti possit orator,
non ille volgaris sed hic excellens, a propriis personis et temporibus, si potest, avocat controversiam.'
The context and application of the questions is rhetorical in the Confessions, but cf. De Diversis
Quaestionibus LXXXIII, 18 (PL 40, 15): 'Ideoque etiam cum veritas quaeritur, plus quam tria
genera questionum esse non possunt; utrum omnino sit, utrum hoc an aliud sit, utrum approbandum
improbandumve sit.' The tendency of these questions toward generalization beyond their specifically
rhetorical meanings is assisted by some of the names attached to them: the pseudo-Augustine calls
them 'rational or logical' questions (De Rhetorica 9 [Halm 142]); Martianus Capella calls them
'principal status' (De Rhetorica 6 [Halm 455]); Clodian 'rational status' (Ars Rhetorica [Halm 590]).
A fourth question or constitution or status is added by Hermagoras, rejected by Cicero and Quin-
tilian, and mentioned by the pseudo-Augustine and Clodian. Concerning the variety and evolution
of questions (or status as he prefers to call them), cf. Quintilian iii. 6. 29-85; his own decision is pre-
sented as one prescribed by nature and coincident with the doctrine of Cicero (ibid. 80): 'Credendum
est igitur his, quorum auctoritatem secutus est Cicero, tria esse, quae in omni disputatione quaerantur,
an sit, quid sit, quale sit? quod ipsa nobis etiam natura praescribit.' For Augustine's enumeration of
scriptural methods and problems, cf. De Doctrina Christiana i. 1-2; PL 34, 19-20.
2 Ibid. iii. 4. 8, and 12. 18; PL 34, 68 and 72-73. 3 Ibid. iv. 1. 1-7. 11; PL 34, 89-94.
controllable influence on our thought except by the context of other words, and
the internal words by which a master speaking within us teaches the truth.1
Whether things be treated as signs or signs as things, only the eternal meanings
and realities are important; knowledge of temporal things and of the arts is
chiefly useful for the interpretation of the language and symbolism of Scripture,
and the sacraments are signs adapted to the mutability of human sensibilities
but immutable in their significance of the changeless things of God.2Once account
is taken of the distinction of things and words into those which are temporal and
those which are changeless, the influence of rhetoric is discernible in many
traits of the Augustinian tradition: in the analogical interpretation of Scripture
and in the numerous mediaeval encyclopedias prepared to facilitate such inter-
pretation (for words are signs which are useful less to designate things than to
express truths and persuade minds, and things therefore are useful to interpret
signs, not signs to interpret things);3 in the literal interpretation in which ap-
parently contradictory texts were reconciled in canon law and theology by use of
the rhetorician's 'circumstances' of statement, that is, by consideration of 'who'
said it, 'where, when, why, how, with what assistance';4 in the organization of
theological problems according to the distinction of things and signs; and in the
place of rhetoric after dialectic in the enumeration of the liberal arts (since it
supplies the means of stating truths once they have been discovered) instead of
before dialectic as in the enumeration of an opposed tradition (since it achieves
only probability and persuasion, but falls short of truth).5
The discussion of logic during the Middle Ages may be divided into four
periods: during the first period the elements of logic were learned from simple
1 De Magistro 3. 5-6 and 11. 36-12. 46; PL 32, 1197-98 and 1215-20. Cf. the excellent statement of
the relation of language to thought by E. Gilson, Introduction a l'Etude de Saint Augustin (Paris,
1929), pp. 87-103. Augustine's conception of rhetoric is developed most fully in the De Doctrina
Christiana, De Ordine, De Catechizandis Rudibus, and Contra Cresconium. Cf. also J. Zurek, 'De S.
Aurelii praeceptis rhetoricis,' Dissertationes Philologae Yindobonenses (Vienna, 1905), vIII, 69-109;
M. Comeau, La Rhgtoriquede Saint Augustin d'apres le Tractatus in Iohannem (Paris, 1930); G.
Combes, Saint Augustin et la Culture Classique (Paris, 1927), esp. pp. 49-56 where true eloquence is
distinguished from the oratorical art; H.-I. Marrou, Saint Augustin et la Fin de la Culture Antique
(Paris, 1938), esp. pp. 507-540 on Christian eloquence. The rhetoric of Cicero was moral and political
in its applications, and the influence of rhetoric extended to political doctrine. The differentiation of
things according to ends loved and means used had already entered Christian ethics in Ambrose's
De Officiis Ministrorum which was based on the distinctions of Cicero's De Officiis, and Cicero's rhe-
torically conceived political theory supplies, by virtue of the same distinction, the terminology for
Augustine's discussion of the city of God as well as the elements of the terrestrial city to which it is
contrasted.
2 Epistola CXXXYIII ad Marcellinum i. 7; PL 33, 527: 'Nimis autem longum est, convenienter
disputare de varietate signorum, quae cum ad res divinas pertinent, Sacramenta appellantur. Sicut
autem non ideo mutabilis homo, quia mane aliud, aliud vespere; illud hoc mense, illud alio; non hoc
isto anno quod illo: ita non ideo mutabilis Deus, quia universi saeculi priore volumine aliud, aliud
posteriore sibi iussit offerri, quo convenienter significationes ad doctrinam religionis saluberrimam
pertinentes, per mutabilia tempora sine ulla sui mutatione disponeret.'
3 Cf. Gilson, o.c., pp. 151-153. 4 De Doctrina Christiana iii. 12. 18-29. 41; PL 34, 72-81.
5 For the fashion in which rhetoric follows and supplements dialectic according to Augustine, cf.
ibid. ii. 35. 53-37. 55; PL 34, 60-61; De Ordine ii. 13. 38; PL 32, 1013.
treatises like the pseudo-Augustine's Principia Dialecticae and CategoriaeDecem
(which Alcuin recommended to Charlemagne as Augustine's translation of Aris-
totle's Categories) or the sections on dialectic in such handbooks as those of
Martianus Capella, Cassiodorus, and Isidore of Seville; during the second period,
after the curriculum instituted by Gerbert at the end of the tenth century, the
basis of instruction in dialectic was broadened to include the works and transla-
tions of Boethius, among them two of the six books of Aristotle's Organon,which
together acquired the traditional name of the Old Logic; during the third period,
the translation of the remaining four books in the twelfth century set up the New
Logic, constituted of the Introduction of Porphyry, the Organonof Aristotle, and
the Six Principles of Gilbert de la Porree, yet the authority of the Old Logic
continued strong, for the contemporaries of John of Salisbury found the Posterior
Analytics, which treats of the principles of scientific demonstration, difficult or
even unintelligible,' and indeed the first important commentary on that work
was written in the thirteenth century by Robert Grosseteste, while as late as the
fourteenth century William of Ockham prepared an Expositio aurea et admodum
utilis super Artem Veterem;and finally during the fourth period, the discussion of
logic is determined less by Aristotle's Organonthan by the Summulae written in
the thirteenth century by Petrus Hispanus, Lambert of Auxerre, and William
of Shyreswood. The extent of the influence of rhetoric on the development of
logic may be judged from the fact that - although Aristotle's logic is charac-
terized not merely by the schemata of terms, propositions, and syllogisms set
forth in the first three books of the Organon,but even more by the differentiation
of proof, in accordance with the principles on which it depends, into three kinds:
scientific or demonstrative, dialectical, and sophistical, which are expounded in
the last three books, the Posterior Analytics, the Topics, and the De Sophisticis
Elenchis - only the first three books had much influence until the thirteenth
century, while principles were treated by devices which Aristotle used in rhetoric
and dialectic, and even after the thirteenth century scientific method was in con-
stant danger of being assimilated to dialectic, the Posterior Analytics to the
Topics.
The early treatments of dialectic in the handbooks and encyclopedias run
through a familiar sequence of subjects: the predicables of Porphyry, the cate-
gories of Aristotle, a briefer treatment of propositions in which the testimony
of Aristotle's De Interpretationeis mixed in small doses with that of the treatise
by the same name attributed to Apuleius, an exposition of the categorical syl-
logism derived from the pseudo-Apuleius and of the hypothetical syllogism de-
rived from the rhetorician Marius Victorinus, and finally, in place of Aristotle's
principles of demonstration, sections on definition and on 'topics' or 'common-
1 John of Salisbury, Metalogicon iv. 6 (ed. C. C. J. Webb, Oxford, 1929), p. 171: 'Deinde hec uten-
tium raritate iam fere in desuetudinem abiit, eo quod demonstrationis usus uix apud solos mathe-
maticos est; et in his fere, apud geometras dumtaxat; sed et huius quoque discipline non est celebris
usus apud nos, nisi forte in tractu Hibero uel confinio Affrice.' In contrast to his brief and almost
flippant treatment of the Posterior Analytics, John devotes more than half of the third book (iii. 5-10,
pp. 139-164) to praise of the utility of the Topics.
places' derived from the Greek rhetoricians by way of Cicero and the lost works of
Marius Victorinus. So direct is the descent of the principles of demonstration from
rhetoric that Cassiodorus closes his consideration of the art of dialectic, having
treated of topics, with 'atechnical' arguments (which form part of the Topics of
Cicero, but figure in the Rhetoric and not the Topics of Aristotle) and memory
(which, although one of the traditional five parts of rhetoric, is common, accord-
ing to Cassiodorus, to orators, dialecticians, poets, and jurists),' while Isidore
supplements his statement of topics with a section on opposites derived from
Cicero.2The basic pattern of this logic was not crucially altered by the return in
the second period to the more extensive logical works of Boethius.' Dialectic' is
not distinct from 'logic' in the tradition of the Old Logic; rather dialectic or logic
is divided on the authority of Cicero into two parts, one (called 'analytic' by the
Greeks according to Boethius) concerned with judgment, the other (called 'topic'
by the Greeks) concerned with discovery.3 Boethius translated and wrote com-
mentaries on Aristotle's Categoriesand On Interpretation, but he also translated
and wrote two commentaries on the Isagoge or Introduction of the Neo-Platonist
Porphyry which expounds, as introduction to the Categories, the predicables
treated by Aristotle in his Topics, and this dialectical treatment of 'the five
words' appeared thereafter, even when the influence of Boethius was slight, in
mediaeval, Renaissance, and early modern treatments of Aristotle's logic and
editions of his Organon.Instead of Aristotle's treatment of syllogisms, mediaeval
philosophers had, until the twelfth century, Boethius' essays On the Categorical
Syllogism (in which the doctrine of Aristotle is modified by the doctrines of
Theophrastus, Eudemus, and Porphyry),4 On the HypotheticalSyllogism (in which
the authority of Theophrastus and Eudemus is invoked for seeking necessary
premisses in the forms of propositions rather than in the nature of things),6 and
On Division (which goes back to the 'peripatetic' tradition according to the open-
ing sentence of the essay, but cites explicitly only Andronicus, Plotinus, and
1 Institutiones ii. 3. 16-17 (ed. R. A. B. Mynors, Oxford, 1937), pp. 127-128; cf. Cicero, Topica
4. 24. and Aristotle, Rhetoric i. 15. 1375a2g-1377bl1. Mynors argues from the manuscripts that the
Institutiones went through two recensions by other hands than Cassiodorus, and in them Boethius
was substituted as authority in dialectic for Marius Victorinus (o.c., pp. xxviii and xxxvii). The
closing sections of the later versions of the treatment of dialectic included, in addition to the rhetorical
subjects of the earlier versions, a treatment of rhetorical places, discovery, and circumstances (PL 70,
1196-1202). 2 Etymologiae ii. 31.
8 De Differentiis Topicis i; PL 64, 1173; In Porphyrium Commentariai; PL 64, 73.
4 For references to Theophrastus, Eudemus and Porphyry cf. De Syllogismo Categoricoii; PL 64,
813, 814, 815, and esp. 829: 'Haec de Categoricorum Syllogismorum introductione, Aristotelem
plurimum sequens, et aliqua de Theophrasto et Porphyrio mutuatus, quantum parcitas introducendi
permisit, expressi.' The Introductio ad Syllogismos Categoricos(PL 64, 761 ff.) seems clearly another
recension of Book I of the De Syllogismo Categorico.
6 De Syllogismo Hypothetico i; PL 64, 843: 'Necessitas vero hypotheticae propositionis et ratio
earum propositionum ex quibus junguntur inter se connexiones, consequentiam quaerit, ut cum dico:
Si Socrates sedet et vivit, neque sedere eum, neque vivere necesse est; sed si sedet, necesse est vivere.
. . . Necessitas enim propositionis in consequentiae immutabilitate consistit.' Cf. De Differentiis
Topicis i (PL 64, 1176) where such propositions are called per se nota. For reference to Theophrastus
and Eudemus cf. De Syllogismo Hypothetico 831.
Porphyry, for treatment of a 'scientia dividendi' in which Aristotle himself placed

little store).l The De Definitione which went under his name is by Marius Vic-
torinus, and it supplies one more channel for the influence of Cicero and rhetoric.2
Finally, instead of a treatment of the differences of demonstrative, dialectical,
and sophistical principles and proofs, Boethius left two works which had the
effect, during the Middle Ages and increasingly during the Renaissance, of trans-
lating the problem of distinguishing principles into the problem of discovering
arguments or things: his Commentaryon the Topics of Cicero and his treatise in
four books On Topical Differences,in which the topical schemes or common-places
of Themistius and Cicero are set forth and reduced to a single classification.3 With
the advent of the New Logic in the third period, during the twelfth century, how-
ever, logic was distinguished from dialectic, and rhetoric became the counterpart
of dialectic, although logic continued to be divided into judgment and discovery.
Finally, during the fourth period, in the Summulae of the thirteenth century, the
emphasis is again on the topics, as it is also in the reaction against logic during
the Renaissance, when the Topics of Cicero and of Boethius were once more used
(as John the Scot had used topics) as inspiration for a scientific method of dis-
covering, not arguments, but things, and the scholastic logic was viewed as a
verbal discipline inferior in precision and practical effectiveness to these devices
of rhetoric.
The treatment of rhetoric, in turn, showed the effects of this extension of the
devices of rhetoric to logic, since it became important to contrast rhetoric and
dialectic when both rhetoricians and dialecticians made use of 'places' for pur-
poses of discovery. Paradoxically, in this tradition in which the methods of rhe-
toric were similar to those of dialectic, rhetoric was subordinated to dialectic,
while in the tradition in which rhetoric was criticized and then transformed to
theological uses, dialectic was subordinated to rhetoric. The fourth book of
Boethius' On Topical Differences, which treats of the differences between dia-
lectical and rhetorical places, was used as a textbook of rhetoric in the twelfth
and thirteenth centuries, and two short treatises devoted to rhetorical places
passed under his name, the Speculatio de Rhetoricae Cognatione (which is more
probably a compilation derived from Book IV of De Differentiis Topicis than an
independent work by Boethius) and the Locorum Rhetoricorum Distinctio.
Boethius finds the distinction between dialectic and rhetoric in their matter, use,
and end: the matter of dialectic is 'theses,' that of rhetoric 'hypotheses,' and thesis
and hypothesis are related as two kinds of 'questions,' the one universal, the other
1 De Divisione, PL 64, 875-876; cf. Aristotle, Prior Analytics, i. 31. 46a31-46b37.
2 On the question of the authenticity of the De Definitione, cf. H. Usener, AnecdotonHolderi (Bonn,
1877), pp. 59-66. For the effect of the De Definitione in introducing rhetorical distinctions into the
mediaeval discussions of logic, cf. C. Prantl, Geschichteder Logik im Abendlande (Leipzig, 1855), I,
688-690.
3 Boethius refers to translations he has made of other books of the Organon, but no evidence has
been found in mediaeval literature of their influence prior to the twelfth century; cf. C. H. Haskins,
'Versions of Aristotle's Posterior Analytics,' Studies in the History of Mediaeval Science (Cambridge,
Mass., 1924), pp. 231 ff. For the rhetorical character and effects of the De Differentiis Topicis cf.
Prantl, o.c. I, 720-722.
particularized to circumstances; dialectic uses interrogation and response, and its

arguments are set forth in syllogisms, rhetoric uses continuous speech involving
enthymemes; the end of dialectic is to force what one wishes from an adversary,
that of rhetoric to persuade a judge.' Boethius takes over the early position of
Cicero, as expressed in the De Inventione, concerning matter, but the whole ques-
tion of end, function, and matter is raised in the context of a considerably longer
list of questions and in that context the other answers have changed. Boethius
asks no fewer than nine questions about rhetoric: its genus, species, matter, parts,
instrument, the parts of the instrument, the work and duty of the orator and his
end. The genus of rhetoric is no longer 'civil science' (as it was for Cicero) but
'faculty' (much as Aristotle had held it to be a bivaC,uLrather than a science). The
matter of the faculty is all things suited to discourse, which, as Boethius puts it,
is almost equivalent to the 'civil question'; this matter of discourse is indetermi-
nate until it is given specific form by the ends of rhetoric: the 'civil question' is
made into a judicial 'cause' when the end considered is the just; into a deliberative
'cause' when the end is the useful or the honorable; into a demonstrative 'cause'
when the end is the good. It is, as Isidore later observed, an elusive question, in
which the genus of an art can be transmuted into its matter, but that strange dif-
ference is one of the slight remnants of the difference between Aristotle's concep-
tion of rhetoric and that of Cicero and the rhetoricians, and from that remnant in
Boethius' questions, mediaeval commentators were to reconstruct, with slowly
increasing erudition, the full specifications of the old opposition.
II
These were not technical questions which were discussed by a few learned men,
but distinctions which entered into all parts of mediaeval culture and life. Chris-
tianity had grown up in the environment of a culture which was preponderantly
rhetorical: indeed the chief differences between Greek and Latin Christianity
may be derived from the difference between the Latin rhetoric of the Republic
and early Empire (in which the arts and sciences had been put to the aid of
rhetoric and civil philosophy had all but been reduced to the art of forensic plead-
ing) and the Greek rhetoric of the Empire (in which philosophy itself had been
displaced by display or epideictic rhetoric in the guise of sophistic, the rules of
oratory had become the canons of literature, and Plato's and Aristotle's compari-
son of rhetoric and medicine had been made into a scientific method which
rhetoric shared with medicine).2 Since many of the early converts who first wrote
1 De Differentiis Topicis iv; PL 64, 1205-1206. Cf. ibid. i; PL 64, 1177.
2 For Greek doctrinal
developments which led to the opposition of civil and sophistic rhetoric and
to the advancement now of one, now of the other as the preferred or unique manner of rhetoric, cf.
C. Brandstaetter, 'De Notionum 7roXLnLKSet OolITrjs Usu Rhetorico,' Leipziger Studien zur clas-
sischen Philoogie, v (1893), 128-274. For the oppositions of sophistic, rhetoric, and philosophy,
cf. H. von Arnim, Leben und Werke des Dio von Prusa (Berlin, 1898), pp. 4-114; H. M. Hubbell,
'The Rhetorica of Philodemus,' ConnecticutAcademy of Arts and Sciences, xxIII (September, 1920),
276-284; J. F. D'Alton, Roman Literary Theory and Criticism (London, 1931), pp. 153 if. For the
interpenetration of rhetoric and dialectic and the transformation of dialectic by rhetoric in Hel-
lenistic and Roman thought, cf. C. Prantl, Geschichteder Logik im Abendlande,i, Abschnitt viii, 505 ff.
Rhetoricin the Middle Ages
on Christian doctrine had been professional rhetoricians before their conversions,1
the rhetorical distinctions which they used in the statement of their problems and
the organization of their works emerged often as doctrinal differences and
empirical observations in later speculation on their statements. This emergence
of rhetoric in the materials of discussion in all fields brought new questions into
the technical disputes of the art. The numerous technical distinctions which had
entered the apparatus and discussion of rhetoric took on applications, which
echo or anticipate many of the positions of philosophers, proper to each of the
three conceptions of rhetoric distinguishable in threefold opposition in the shifting
materials to which rhetoric is applied.
Until the coming of the New Logic in the twelfth century the pattern of that
opposition is relatively simple: the rhetorician who professed to treat of subject
matters accessible to the 'common notions' of the mind without need of technical
competence, found himself opposed on the one hand by theologians who had
learned from Augustine to use the distinction between words and things both to
attack the rhetoric of the schools and to practise a rhetoric concerned with divine
eloquence and divine things, and on the other hand by rhetoricians who had
learned from Boethius to use the distinction between thesis and hypothesis to
limit rhetoric to probable reasoning concerning specifically delimited questions
subordinate to the general questions of dialectic. To the Augustinian the exces-
sive use or extension of rhetoric no less than that of dialectic was suspect; to the
peripatetic follower of Boethius limitation or criticism of dialectic, whether from
the point of view of theology or of rhetoric, was an attack on the use of reason;
and to the rhetorician as such, limitation of rhetoric by the laws of logic or theol-
Philostratus includes in his Lives of the Sophists some of the ancient philosophers who approximated
the rhetorical style of the sophists, but he distinguished philosophy from sophistic (i. 481) since
philosophers merely set snares for knowledge by their questioning, but asserted that they had no
sure knowledge, whereas sophists of the old school professed knowledge of that whereof they spoke.
Philostratus' enthusiastic account of the sophists of the Empire is vivid indication of the spread
and importance of epideictic rhetoric; its influence is likewise to be remarked in the Eastern Church,
particularly among the Cappadocian fathers: cf. E. Norden, Die Antike Kunstprosa ii, 529 ff. and 550
ff.; T. C. Burgess, 'Epideictic Literature,' University of Chicago Studies in Classical Philology inI
(1902), 89-251; L. Meridier, L'Influence de la seconde sophistique sur l'oeuvre de Gregoire de Nysse
(Paris, 1906); M. Guignet, Les Procedes Epistolaires de St. Gr4goirede Nazianze (Paris, 1911); T. E.
Ameringer, The Stylistic Influence of the Second Sophistic on the Panegyrical Sermons of St. John
Chrysostom(Washington, 1921); J. M. Campbell, The Influence of the Second Sophistic on the Style of
the Sermons of St. Basil the Great (Washington, 1922); A. Boulanger, Aelius Aristide et la sophistique
dans la provinced'Asie au ii siecle de notre ere (Paris, 1923). The crossing lines of rhetoric and medicine
are apparent in Eunapius' Lives of the Philosophers; cf. particularly his accounts of Zeno of Cyprus,
Magnus, Oribasius, and Ionicus (497-499). Magnus made a happy combination of rhetoric and
medicine by persuading the patients of other doctors that they had not been cured and then restoring
them to health, apparently also by talk and questions; Ionicus was master of philosophy and medicine
as well as the arts of rhetoric and poetry. Cf. P. H. and E. A. De Lacy, Philodemus: On Methods of
Inference (Philadelphia, 1941), pp. 130 ff., where the relations between medicine and rhetoric are
discussed in terms of an 'empirical' or 'conjectural' method.
1 Cyprian (cf. Jerome, De Yiris Illustribus 67; PL 23, 714), Arnobius (cf. Jerome, Chronicon ad
annum 329; PL 27, 675-676), Lactantius (zbzd., ad annum 319; PL 27, 669-670), Augustine (Con-
fessions iv. 2. 2; PL 32, 693-694). Most of the other early Christian writers in the West, even those
who had not been teachers cf rhetoric, had studied the art as part of their education.
ogy was unwarranted restriction of the scope of reason and visionary neglect of
the practical exigencies of the problems of law and morals. The simple lines of this
opposition appear even in the early discussions of rhetoric, and they are pre-
served after the appearance of the New Logic, beneath the surface of the more
intricate distinctions made necessary by the Aristotelian differentiation of logic
from dialectic, poetic, sophistic, and rhetoric. These three main positions, taken
throughout the Middle Ages with respect to rhetoric, may be marked off into
four historical stages sharply distinguished by the authorities on which the
discussion of the arts was successively based: a first stage extending to about
the end of the tenth century when the chief authorities were the pseudo-Augus-
tine, Martianus Capella, Cassiodorus, and Isidore; a second period extending
through the eleventh and the first half of the twelfth century dominated by
Cicero, Boethius, and the Old Logic; a third period comprising the latter part
of the twelfth century and the greater part of the thirteenth century in which
the New Logic became to some degree effective and was applied after a manner
in the interpretation of the Aristotelian corpus; and finally the fourteenth cen-
tury and the Renaissance in which Aristotle and the Greek rhetoricians, Cicero,
Quintilian, and Boethius all had increasing influence.
During the first period rhetoric was concerned - on the authority of Herma-
goras, Cicero and Boethius, Fortunatianus, Augustine and Victorinus, and all the
even more derivative authorities that depended on them - with civil philosophy.'
According to Cassiodorus, 'The art of rhetoric is, as the masters of secular letters
teach, the science of speaking well in civil questions,' and that definition is re-
peated in almost the same words by Isidore, Alcuin, and Rhabanus Maurus.2
The occasion of the dialogue with Charlemagne in which Alcuin's doctrine is
stated is a request made by the emperor for information concerning the art, since
he thinks it ridiculous for one whose daily occupation is with civil questions to be
ignorant of the precepts of the art; the dialogue, moreover, is frankly moral not
only in its traditional title, On Rhetoricand the Virtues, but in purpose, since the
transition from rhetoric to the virtues is accomplished by recognition that this
'sermocinandi ratio' which is applied to civil cases and secular business must be
1 Cf. Cicero, De Inventione i. 5. 6; Ad Herennium i. 2. 2; Boethius De Differentiis Topicis iv; PL
64, 1207; Fortunatianus, i. 1. (Halm 81); pseudo-Augustine, De Rhetorica 3 (Halm 138); Fabius
Laurentius Victorinus, Explanationes in Rhetoricam M. Tullii Ciceronis i. 5 (Halm 171). The au-
thenticity of the De Rhetorica attributed to Augustine was questioned by his Benedictine editors in
1679 (cf. PL 32, 1439) and by most authorities since that time; cf. M. Gibert, Jugemens des Savans
sur les Auteurs qui ont Traite de la Rhetorique (Paris, 1716), II, 98: 'Mais pour peu qu'on connoisse le
style du Saint, il est ais6 de voir que l'Ouvrage n'est pas de lui'; and G. Saintsbury, A History of
Criticism and Literary Taste in Europe (New York, 1900), I, 377. Its authenticity has been defended
on philological grounds by W. Crecilinus, S. Aurelii Augustini de dialectica liber (Elberfeld, 1857)
and A. Reuter, 'Zu dem Augustinischen Fragment de arte rhetorica,' Kirchenhistorischen Studien,
324-341; but the arguments adduced have been answered by J. Zurek, o.c. The pseudo-Augustine
attributes the position taken by Fortunatianus to Hermagoras.
2 Cassiodorus, Institutiones ii. i. 1, p. 97; cf. ii. Praef. 4, p. 91: 'secundo de arte rhetorica, quae
propter nitorem et copiam eloquentiae suae maxime in civilibus quaestionibus necessaria nimis et
honorabilisaestimatur'; Isidore, Etymologiaeii. 1. (Halm 507); cf. i. 2. 1. and ii. 10., where law is treated
as one of the subheads of rhetoric; Alcuin, De Rhetorica et de Virtutibus 3 (Halm 526); Rhabanus
Maurus, De Clericorum Institutione iii. 19; PL 107, 396.
supplemented by the other virtues. Yet within this broad agreement among
rhetoricians that rhetoric is concerned with civil questions, there are numerous
differences of statement, which sometimes lead to changes in the devices thought
proper to rhetoric and which seem often to entail major philosophic differences.
The chief of these is the difference between the position (which seems to go back to
Hermagoras and for which Fortunatianus is sometimes given as authority) which
treats civil philosophy in terms of the 'common notions' of mankind and therefore
undertakes to differentiate the subject matter of rhetoric in terms of the questions
treated, that is, the kinds of theses and hypotheses, and the position (which goes
back to Cicero) which finds the subject matter of rhetoric in the three genera, de-
liberative, demonstrative, and judicial. The former has the effect of emphasizing
the common bases of rhetoric in human knowledge while turning analyses to the
peculiarities of the questions that can be asked, the other the effect of centering
on the common qualities of the subject matter and directing inquiry to the
peculiarities and virtues of the orator. The problems of rhetoric arise largely in
the mixtures of the two traditions. Cassiodorus, citing Fortunatianus, defines
civil questions as those which fall within the common conception of the mind,
that is, which any one can understand when it is a question of the equitable and
the good; Sulpitius Victor as those which are proper to no art but common to the
opinion of all; Alcuin as those learned questions which can be conceived by the
natural power of the mind.' Victorinus on the other hand divides the possible
matter of rhetoric into two kinds: that with which the art operates (ubi fit),
namely deliberative, demonstrative, judicial, and that from which the art is
formed (unde fit), namely, the arguments which contribute the matter of those
three kinds, then limits the consideration of rhetoric to the former, and refutes
Hermagoras' doctrine of thesis and hypothesis in favor of the Aristotelian and
Ciceronian doctrine of the three genera.2 Martianus Capella repeats this dif-
ferentiation of two kinds of matter but goes on to the exposition of theses and
hypotheses, confining his disapproval to a remark concerning the extremely
subtle reasons of some of the sectaries of rhetoric who hold that all rhetorical
questions are general or theses.3 The difference is between a tendency to make dis-
tinctions in terms of a subject matter and arguments suited to it and a tendency
to make distinctions, often indeed the same distinctions, in terms of the orator
and his problems of discovering and stating arguments. The former emphasis
tends to intellectualize the art and change its orientation to a subject matter and
its peculiarities into problems of inquiry and understanding, as when Sulpitius
Victor, having limited rhetoric to the civil question and having divided the civil
question into two parts, thesis and hypothesis, finds three duties for the orator:
understanding, discovery, and disposition (the first of which was neglected by
Cicero, but adequately treated by the Greeks) and then three genera of causes in
the place of those long customary: the ethical, pathetic, and judicial. The latter
emphasis leads to a series of questions, which were much discussed during the
1 Institutiones ii. Q. 1. p. 97; cf. Fortunatianus, i. 1. (Halm 81) and the pseudo-Augustine (De
Rhetorica4 [Halm 139]) who supplies the Greek term KOpiP)vpotasuggestive of stoic origins. Sulpitius
Victor, Institutiones Oratoriae (Halm 314) and Alcuin, De Rhetorica et de Virtutibus 3 (Halm 526).
2 Fabius Laurentius Victorinus 5 (Halm 3 Martianus Capella 5 (Halm 454).
174-177).
Middle Ages, concerning the relation of morals and eloquence, concerning the
relation of art and wisdom, concerning the definition of rhetoric as a virtue or an
art or a discipline.l Rhetoric was to come into conflict with dialectic as a conse-
quence of this tendency, as it was to come into conflict with theology as a con-
sequence of its tendency to annex the problems of morals and the interpretation
of Scripture. Since its discipline was gradually limited by the transfer of the
commonplaces, definition, and finally proof - even in the rhetorical formulations
they had received from Cicero, Victorinus, and Boethius - to the domain of
dialectic, and since its subject matter was limited by the transfer of moral and
political questions to theology, rhetoric entered into a second period during which
it developed along three separate lines: as a part of logic, or as the art of stat-
ing truths certified by theology, or as a simple art of words.
III
The subordination of rhetoric to logic was accomplished usually in terms of the
greater particularity of its subject matter, its concern with hypotheses rather
than theses; and the terms of the discussion of the relation of rhetoric to dialectic
were borrowed from Boethius. The doctrine is expressed, however, before the
appearance of Boethius in the curriculum of the schools. According to Isidore of
Seville, logic (Isidore adds that the Greek term X6'yosmeans 'rational') has two
parts, dialectic and rhetoric.2 John the Scot omits grammar and rhetoric from his
treatise On the Division of Nature first because many philosophers think they are
parts of dialectic, second from considerations of brevity, and finally because,
unlike dialectic, grammar and rhetoric do not treat of the nature of things but
either of words significant by convention or of special causes and persons.3
1 Sulpitius Victor, Institutiones Oratoriae4 and 6 (Halm 315, 316). Cato's definition of the orator
as vir bonus dicendi peritus (Quintilian xii. 1.; Seneca, Controversiarumliber i. Praef. 9) was frequently
repeated before the Carolingian period - by Fortunatianus, Victorinus, Cassiodorus, Isidore (Halm
81, 177, 495, 507) - and one of the favorite etymologies of 'art' derived it from the Greek word for
virtue. In the twelfth century Aristotle's authority (cited from the Categories) is used to deny that
rhetoric is a virtue (cf. Abailard, Dialogus [PL 178, 1652]; Hermannus, Epitome TheologiaeChristianae
[PL 178, 1750]; Sententie Parisienses [ed. A. Landgraf, Ecrits Th6ologiques de l'Ecole d'Abelard,
Louvain, 1934, p. 52]). In the thirteenth century Aristotle's authority (cited from the Nicomachean
Ethics) could be quoted to place it, together with the other arts, among the intellectual virtues. In the
Renaissance one of the chief grounds for Ramus' violent criticism of Quintilian is found in his tend-
ency to identify rhetoric with morals (cf. P. Ramus, Rhetoricae Distinctiones in Quintilianum [Paris,
1559]). 2 De Differentiis Rerum c. 39; PL 83, 93-94.
3 De Divisione Naturae v. 4; PL 12Q, 869-870: 'Primum quidem, quia ipsae duae artes veluti quae-
dam membra Dialecticae multis philosophis non incongrue existimantur. Deinde brevitatis occasione.
Postremo, quod non de rerum natura tractare videntur, sed vel de regulis humanae vocis, quam non
secundum naturam, sed secundum consuetudinem loquentium subsistere Aristoteles cum suis sec-
tatoribus approbat, vel de causis atque personis specialibus, quod longe a natura rerum distat. Nam
dum Rhetorica de communibus locis, qui ad naturam rerum pertinent, tractare nititur, non suas, sed
Dialecticae arripit partes.' Rhetoric is limited to hypotheses or finite questions determined by the
seven circumstances, while the common conceptions of the mind have become the property of dia-
lectic; cf. ibid. i. Q7;PL 1B2,475: 'Rhetorica est finitam causam persona, materia, occasione, qualitate,
loco, tempore, facultate discutiens copiose atque ornate disciplina; breviterque definiri potest,
Rhetorica est finitae causae septem periochis sagax et copiosa disciplina. Dialectica est communium
animi conceptionum rationabilium diligens investigatrixque disciplina.'
The pseudo-Rhabanus Maurus was one of the philosophers who divided logic
into three parts: grammar, rhetoric, and dialecticl and Remigius of Auxerre
divides philosophers into four kinds: dialecticians, rhetoricians, sophists who
always come to false conclusions, and jurists who dispute concerning the status
of law.2 Gerbert, who used all the dialectical works and translations of Boethius
in his teaching at Rheims and Paris (including Cicero's Topics, which, like
Cassiodorus, he thought Cicero had translated from the Greek, and the On
Definition of Marius Victorinus), likewise considered dialectic and rhetoric
parts of logic, and taught rhetoric after dialectic.3 Fulbert, finally, who restored
studies at Chartres in the eleventh century and who knew, in addition to the
De Inventione and the Ad Herennium, Victorinus' commentary on Cicero and the
two treatises on rhetorical places attributed to Boethius, has left twenty-one
verses on the differences between rhetoric and dialectic: they are the three dif-
ferences Boethius found between the matters, uses, and ends of the arts.4
The transition to the third period in this tradition of rhetoric determined rela-
tive to dialectic, is accomplished when the increased influence, or at least the
increased repute, of the New Logic led to separation of scientific or demonstrative
proof from probable proof and to the location of rhetoric with dialectic under the
latter. It is a gradual transition, dependent on increase of erudition in logic as
much as in rhetoric. In the comprehensive collection of texts in the liberal arts
prepared by Thierry of Chartres under the title Heptateuchonabout 1141, all of
Aristotle's Organonexcept the PosteriorAnalytics and the second book of the Prior
Analytics appears, while under rhetoric are included (in addition to the De
Inventione, the Ad Herennium and Martianus Capella - the traditional sources
of rhetoric- and Cicero's Topics which with Boethius' De Differentiis Topicis is
classified under dialectic) only Cicero's De Partitione Oratoria and Julius
Severianus' Precepts on the art of rhetoric.5Yet Thierry of Chartres wrote a com-
mentary on the De Inventione in which a history of rhetoric is reconstructed to
explain the opening paragraph of that work as a refutation of Plato and Aristotle:
Plato had argued that rhetoric was no art, Aristotle that it was an art but a bad
art, while Cicero contends against both that it is a good art.6 A short Preface and
1 V. Cousin, OuvragesIn&ditsd'Ab6lard (Paris, 1836), p. 614.
2 B. Haur6au, 'Commentaire de Jean Scot Erigene sur Martianus Capella,' Notices et Extraits des
Manuscrits de la BibliothaqueImperiale, xx, 2 (186Q), 11. Cf. ibid., pp. 20-21, for his difference from
the doctrine of John concerning the natural bases of rhetoric in human nature.
3 The sequence of studies, as directed by Gerbert, were: first, dialectic, which included the Isagoge
of Porphyry (with Boethius' commentary), the Categories and On Interpretation of Aristotle, the
Topics (translated by Cicero and with Boethius' commentary), Boethius' On Topical Differences, On
CategoricalSyllogisms, On Hypothetical Syllogisms, On Definitions, On Divisions; second, as prepara-
tion for rhetoric, the poets; third, rhetoric; finally, sophistic. He includes the entire program under
the term 'logic.' Richer, Historiae, in, 44 ff. (ed. G. H. Pertz, Monumenta Germaniae Historica, t. v,
Scriptores, in, 617).
4A. Clerval, Les Ecoles de Chartres au Moyen-Age (Chartres, 1895), p. 115.
5 Clerval, o.c.
22-22;3; cf. R. McKeon, 'Aristotelianism in Western Christianity,' Environmental
Factors in Christian History (Chicago, 1939), pp. Q15-219.
6 FragmentumScholiastae Inediti ad Ciceronemde Inventione Rhetorica (W. H. D. Suringar, Historia
Critica ScholiastarumLatinorum [Leyden, 1834],,I,13-253), pp. QQ4-Q35.Thierry's reading in works of
an Introduction precede the fragment of the Commentarywhich has been pre-

served. The Introduction is devoted to asking ten questions concerning rhetoric:
its genus, definition, matter, duty, end, parts, species, instrument, who the orator
is, and why the art is so called, and to these two specific questions are added: the
intention of Tully in this work and the utility of the work. The genus of rhetoric
is still civil science, it is not a part of logic, and its matter is hypothesis.' Nor is the
position of Thierry an anachronistic piece of conservatism, for one of the works
which was most influential in preparing the way for the new knowledge of the
thirteenth century and which was eagerly consulted as a source of information
concerning the Arabic learning, the De Divisione Philosophiae of Gundissalinus,
contains a section on rhetoric which not merely ask the same ten questions as
Thierry of Chartres but is identical, apart from slight variations, with the
Introduction to his Commentary.2Gundissalinus differs slightly from Thierry in
rhetoric was apparently more extensive than the contents of the Heptateuchon,since he quotes Quin-
tilian, ibid., p. 219.
1 Ibid., Q17: 'Genus igitur artis rhetoricae est qualitas ipsius artificii secundum ejus effectum: hoc
autem est, quod ipsum artificium est, pars civilis scientiae major. Nam civilis ratio dicitur quidquid
civitas aut rationabiliter dicit aut agit; dicimus enim: ratio est hoc vel illud facere vel dicere. Item
civilis ratio dicitur, scientia dicendi aliquid rationabiliter et faciendi. Et haec quidem ratio, scientia
civilis dicitur, cujus quidem pars integra, vel etiam major, rhetorica est. Nam sapientia i.e. rerum
conceptio secundum earum naturam, et rhetorica civilem scientiam componunt. Et enim nisi quis
sapiens et eloquens fuerit, civilem scientiam habere non dicitur. Major vero pars civilis scientiae dici-
tur rhetorica, quoniam magis operatur in civilibus causis quam sapientia, etsi sine sapientia nihil
prosit. Maximam enim virtutem habet eloquentia in civitate, si sapientiae juncta sit.' Thierry then
goes on to compare this solution with Boethius' doctrine that the genus of rhetoric is facultas, and
finds them in agreement since the same science is an art in the master who teaches its rules and a
faculty in the orator. He is explicit in excluding rhetoric from logic: 'Non est autem dicendum,
rhetoricam aut logicam esse aut ejus partem, idcirco quod logica circa thesin solam i.e. circa genera
agendi, tantummodo versatur.' Cf. ibid. 219 for materia. Cf. Adelard of Bath, De Eodem et Diverso
Beitrdgezur Geschichteder Philosophie des Mittelalters [henceforth BGPM], Band Iv, Heft 1), pp. 19 ff.
2 Dominicus Gundissalinus, De Divisione Philosophiae (ed. L. Baur, BGPM, Band Iv, Heft 2-3,
MUinster, 1903), pp. 63-69. For the strange history of scholarly inquiries into the commentary of
Thierry, cf. M. Grabmann. 'Eine lateinische Ubersetzung der pseudo-Aristotelischen Rhetorica ad
Alexandrum aus dem 13. Jahrhundert,' Sitzungsberichteder BayerischenAkademieder Wissenschaften,
Philosophisch-Historische Abteilung (1931/2, Heft 4), pp. 4-5: In spite of the fact that it was pub-
lished by Suringar in 1834, the fragmentary Commentarywas discussed as an unpublished document
by Rohde in 1881, BiUchelerin 1883, and Thomas in 1884; its author was supposed to have been a
contemporary of Theodoric the Great until Thomas suggested that the document was mediaeval
(and consequently of very little interest); finally Manitius identified it as the work of Thierry or
Theodoric of Chartres, and Klibansky pointed out its identity with the work published by Suringar.
Grabmann does not notice that there is one further coincidence, viz., the identity of one of the three
sections - pp. 216-923 in Suringar - with the section on rhetoric in Gundissalinus. Short of ex-
amination of the manuscripts - unfortunately impossible at this time - the question of priority
is difficult to decide: some of the sections contained in Thierry but omitted by Gundissalinus seem
rather in the nature of additions to than omissions from an original text, and the references seem better
suited to the Commentarythan to the De Divisione Philosophiae (as, e.g., where Thierry says [p. 220]:
'Sed quid sit circumstantia, in sequentibus melius dicetur,' Gundissalinus says [p. 66]: 'Set quid sit
circumstancia in Tullio dicetur,' although no further reference is made to Cicero on this point); on
the other hand, the supposition that the work of Thierry was prior runs into the grave difficulty that
all of the sciences in the De Divisione Philosophiae are treated by means of the same ten questions
here applied to rhetoric.
the classification of rhetoric, for whereas Thierry would have it a part of civil
science and not a part of logic, Gundissalinus classifies both rhetoric and poetic
among the eight parts of logic, but he also classifies rhetoric and poetic as parts of
civil science.'
Hugh of St Victor, who was contemporary with Thierry of Chartres, follows
the suggestion of the Aristotelian division of the sciences into theoretic, practical,
and mechanical (which seems to be Hugh's substitute for Aristotle's productive
science): logic is a fourth branch and not a part of politics, which falls under the
practical sciences. Moreover, his classification of logic makes an excellent transi-
tion from the customary classification according to the trivium of grammar,
rhetoric, and dialectic to the 'Aristotelian' classification as parts of logic and ac-
cording to the kinds of proof. Following Isidore of Seville, Hugh points out the
double etymology of X6yos,i.e., sermoand ratio, and argues that logic can be called
either a verbal or a rational science (sermocinalis sive rationalis scientia); rational
logic (which Hugh also calls dissertiva) is divided into dialectic and rhetoric,
while verbal logic is the genus of grammar, dialectic, and rhetoric, and therefore
rational logic is contained under it.2 This treatment of the traditional trivium is
supplemented, however, by another division of logic into grammar and ratio dis-
serendi or 'dissertive' logic, which is concerned with words as understood (de
vocibus secundum intellectus). 'Dissertive' or rational logic is in turn divided into
integral parts, i.e., parts shared by its kinds, which turn out to be the Ciceronian
distinction into discovery and judgment, and divisive parts: demonstrative,
probable, and sophistic; the two parts of probable proof are dialectic and rhetoric.3
John of Salisbury, one of the pupils of Thierry of Chartres, who had studied the
whole of Aristotle's Organonand who was widely read in Cicero and Quintilian,
attributes to Plato the division of logic into dialectic and rhetoric, but prefers,
as more philosophic, the division into demonstrative, probable, and sophistic,
with the further division of probable into dialectic and rhetoric.4 William of
Conches, on the other hand, whom John calls the finest grammarian after Bernard
1 The section on the genus of rhetoric in Gundissalinus (p. 64) is the same as the statement quoted
above (p. 17, n. 1) from Thierry, but stops short before the discussion of Boethius and the statement
that rhetoric is not a part of logic. In the section on logic, Gundissalinus cites Alfarabi for the eight
parts of logic (ibid., 71):'Secundum Alfarabium octo sunt partes logice: cathegorie, perihermenias,
analetica priora, analetica posteriora, thopica, sophistica, rethorica, poetica.' He need not have gone
to the Arabs for this doctrine, for the equivalent of the six books of Aristotle's Organonplus rhetoric
and poetic constituted the logic taught by Gerbert (cf. above, p. 16, n. 3). Gundissalinus gives as the
genus of logic that it is a part and instrument of philosophy (ibid., 69) and denies that its matter is
'thesis,' arguing that it is the second intention of the understanding (ibid., 70-71). The genus of
poetic is the same as rhetoric (ibid., 54): 'Genus huius artis [sc. poeticae] est, quod ipsa est pars ciuilis
sciencie, que est pars eloquencie. Non enim parum operatur in ciuilibus, quod delectat uel edificat in
sciencia uel in moribus.'
2 Didascalicon i. 11
(ed. C. H. Buttimer [Washington, 1939], pp. 20-21 [or i. 12; PL 176, 749-750]).
Cf. Isidore of Seville, Etymologiae ii.24. 7.
3
Didascalicon ii. S8-30. pp. 44-47 (or ii. 29-31; PL 176, 763-766).
4 Metalogicon ii. 3. (ed. C. C. J. Webb, Oxford, 1929), pp. 64-65. Baldwin complains (o.c., p. 157)
that rhetoric is barely mentioned in the Metalogicon and seems 'to have no distinctive composing
function'; the few references which he finds indicate that he was looking for rhetoric before the treat-
ment of logic, whereas John treats it under the Topics. Cf. Metalogicon iii. 5. p. 139 and esp. 10,
pp. 154-155: 'Quia ergo exercitatio dialectice ad alterum est; pares, quos producit et quos rationibus
of Chartres,' divides eloquence, which the ancients called logic, into grammar,
ratio disserendi, and rhetoric.2
IV
The translations of Aristotle affected the discussions of theology no less than
those of philosophy, and the changes in rhetoric, and in the relations of rhetoric
to dialectic, are reflected in the methods of theology: 'Aristotelian' conceptions
of the organization of logic with rhetoric as one of its parts were not, how-
ever, intruded into theology, since the opposition was between the Augus-
tinian conception of a single body of theological and philosophic truth possessed
of a single method, and the conception of a philosophy independent in method
and subject matter from theology; and therefore the simple organization of the
trivium as three rational or verbal sciences continued in theology and even in
philosophy under the influence of Augustine long after it became obsolete in the
philosophy influenced by Aristotle. Even as early as the sixth century when Cas-
siodorus wrote his Expositio in Psalterium he could appeal, in his introductory
chapters, On the Eloquence of the Whole Divine Law and On the Proper Eloquence
- Augustine, Jerome,
of the Psalter, to an impressive list of learned Fathers
Ambrose, Hilary- who had studied both the figures which are common to
sacred and secular letters and the proper modes of divine speech which are not
touched by grammarians or rhetoricians.3 Divine eloquence is not formed of
human words or involved in human ambiguities, but since its purpose is to spread
divine law to all the corners of the world, it makes many uses of modes of speech,
and it is 'succinct with definitions, adorned with figures, marked by the propriety
of words, expedited by the constructions of syllogisms'; and while these devices
are certain and clear in the Scriptures, they stand in need of the liberal arts when
they come into contact with the opinions and disputes of men.4 His commentary
consists largely of such aids to understanding, dotted with identifications of kinds
muniuit et locis, sua docet arma tractare et sermones potius conserere quam dexteras, et tanta cautela
imbuit, ut totius eloquentie precepta hinc tracta principaliter, uelut a primitiuo fonte originis sue,
manare perspicuum sit. Indubitanter enim uerum est, quod fatentur Cicero et Quintilianus, quia
hinc non modo rethoricorum adiumentum, sed et principium rethores et scriptores artium assump-
serunt; postmodum tamen propriis dilatata est institutis.' The matter of dialectic is still the 'question'
as distinguished from the 'hypothesis,' which is the matter of rhetoric (ibid., ii. 12, pp. 83-84).
1 Ibid. i. 5, pp. 16-17: 'Willelmus de Conchis, grammaticus post Bernardum Carnotensem opu-
lentissimus.'
2 C. Ottaviano, Un Brano Inedito della 'Philosophia' di Guglielmodi Conches (Naples, 1935), p. Q8.
3 Expositio in Psalterium, Praefatio xv; PL 70, 21.
4 Ibid. 19:
'Eloquentia legis divinae humanis non est formata sermonibus, neque confusis incerta
fertur ambagibus, ut aut a rebus praeteritis oblivione discedat, aut praesentium confusione turbetur,
aut futurorum dubiis casibus eludatur; sed cordi, non corporalibus auribus loquens, magna veritate,
magna praescientiae firmitate cuncta dijudicans, auctoris sui veritate consistit.... Eloquentia
siquidem est ad unamquamque rem competens et decora locutio.' Cf. ibid. 20: 'Haec mundanarum
artium periti, quos tamen multo posterius ab exordio divinorum librorum extitisse manifestum est,
ad collectiones argumentorum, quae Graeci topica dicunt, et ad artem dialecticam et rhetoricam
transtulerunt; ut cunctis evidenter appareat, prius ad exprimendam veritatem justis mentibus datum,
sacris
quod postea gentiles humanae sapientiae aptandam esse putaverunt. Haec in lectionibus
tanquam clarissima sidera relucent, et significantias rerum utilissimis compendiis decenter illumi-
nant.'
of definition, figures of speech, forms of arguments.1 The evolution of this use of
rhetoric consists primarily in the increasing formalization of the methods of
interpreting Scripture and the rules of divine eloquence, and secondarily in the
recurrent application of the secular art to Scripture and the recurrent expressions
of concern at the excesses of the liberal arts in such application. In the one line
of development, Augustine's simple suggestion that things as well as words are
signs was elaborated until the spiritual sense, which balanced the literal sense,
was divided into three kinds, the allegorical, the moral, and the anagogic; and
this theological development of rhetoric eventually in turn influenced mundane
or poetic rhetoric.2 In the other line of development, more suspect of error and
more frequently condemned in one form by conservative theologians who
practised it in another form, defending it as indispensable to the understanding
of Scripture, rhetoric supplied devices to clarify the meanings and remove the
ambiguities, of Scriptural statements. Abailard begins his Commentary on the
Epistle of St Paul to the Romans with the statement: 'The intention of all divine
Scripture is to teach or to move in the manner of a rhetorical speech,' and derives
his triple division of the Old and New Testaments from these two purposes.3The
1 Cf. ibzd. i (PL 70, 27) for identification of two kinds of definition according to the technical terms
of Victorinus; (PL 70, 33) where the figure is explained by means of the mathematical disciplines;
vi. 1 (PL 70, 61) where the fashion in which the divine eloquence has been enriched by the various
arts and disciplines is illustrated by discussion of rhetorical status; xliii. 15 (PL 70, 314) where the
figure of anaphora is identified, and so passim.
2 Cf. Augustine, De Utilitate Credendi3, 5; PL
4P, 68 (historical, aetiological, analogical, allegorical
senses); Gregory the Great, Moralia, Epistola Missoria; PL 75, 510-515 (historical, allegorical, moral);
Peter Abailard, Expositio in Hexaemeron; PL 178, 731 (historical, moral, and mystic); Hugh of St
Victor, De Sacramentis, Prologus 4; PL 176, 184 (historical, allegorical, tropological); and Peter of
Poitier, Allegoriae super TabernaculumMoysi, Prologus (ed. P. S. Moore and J. A. Corbett, Notre
Dame, 1938), p. 1 (historical, allegorical, moral, anagogic); cf. P. S. Moore, The Works of Peter of
Poitiers (Notre Dame, 1936), pp. 65-77. Cf. Thomas Aquinas, Summa Theologica i. q. 1, a. 10:
'Respondeo dicendum quod auctor sacrae Scripturae est Deus, in cujus potestate est ut non solum
voces ad significandum accommodet (quod etiam homo facere potest) sed etiam res ipsas. Et ideo,
cum in omnibus scientiis voces significent, hoc habet proprium ista scientia quod ipsae res significatae
per voces, etiam significant aliquid.' The first of these significations is historical or literal, the second
(in which things signify other things) spiritual, and the spiritual interpretation is further divided into
allegorical, moral, and anagogic. Dante follows the division of Aquinas; cf. Epistola X Domino Cani
Grandi de Scala vii. 98-116; Convivio,ii. 1 (cf. ibid. 14 for rhetoric). The 'four senses' are also used to
explain the'form of wisdom' (cf. Bonaventura, InHexaemeron, Collatioii [OperaOmnza,ed. Quaracchi,
1891], v, 336-342, i.e., uniform, multiform [allegorical, anagogic, tropological, each of which has two
forms], omniform, and nulliform) and to classify the sciences (cf. M. Grabmann, Die Geschichteder
Scholastischen Methode [Freiburg i/B, 1911] ii, 43, n. 1, where a quotation is given from an unpub-
lished manuscript, dated broadly as posterior to Hugh of St Victor, in which the sciences are divided
into theoretic, practical, and logical; practical science in turn is divided into actual [ethics, economics,
and politics] and inspective, which is divided into historia and sptritualis intelligentia; history simply
states the order of things without any hidden meaning apparent from that conveyed by the words;
the spiritual understanding is divided into the tropological, allegorical, and anagogic. Rational logic
is divided into dialectic, apodictic [or demonstrative], and sophistic). Bonaventura also uses them as
the fourfold division in the 'light of sacred Scripture' (De ReductioneArtium ad Theologiam, 5 [ed.
Quaracchi, 1891], v, 321).
3 Commentariasuper S. Pauli Epistolam ad Romanos, Prologus, PL 178, 783-784.
divine pages cannot be read and appreciated without grammar and rhetoric.l An
anonymous commentary on Romans repeats Abailard's statement of the twofold
rhetorical purpose of the Old and New Testament after having specified that all
the arts are servants to divinity: grammar which teaches constructions, dialectic
which expounds by arguments, and rhetoric which consists in persuasion.2 Even
theologians who, like Robert of Melun, opposed the excessive use of rhetoric in
secular as well as in divine letters, repeated the same judgment of the rhetorical
purposes of Scripture.3
The method of rhetoric was, moreover, put to another and even more charac-
teristic use in the interpretation of theological doctrine. The scholastic method,
as it came to be called, grew out of the assemblage of 'sentences' which derived
their name and their initial methods of treatment from rhetoric.4 The early
collections of canon law were collections of authorities - statements from Scrip-
ture, decisions of councils, decretals, opinions of the Fathers - which because
of the practical problems involved in direction of action presented urgently the
problem of bringing discordant or apparently discordant canons into concordance.
When Peter Abailard assembled apparently contradictory texts in his Sic et Non,
the rules for interpreting them which he set forth in the Prologue are develop-
ments of the rules elaborated by a long line of canon lawyers - notably Hincmar
of Rheims, Bernold of Constance, Ivo of Chartres - and involve such directions
as careful consideration of context, comparison of texts, specification of time,
place, and person, determination of original cause of statement, differentiation
of general measures from particular.5 Although this method led to a further step
in the dialectical resolution of the contradictions, the method at this stage is
rhetorical rather than dialectical. The rules of interpretation of the Prologue of
the Sic et Non, thus, approximate the performance of Abailard's Commentary
1 Introductio ad Theologiam ii. i, PL 178, 1044: 'At jam profecto nec grammaticam a Christiano
legi convenit, sine documentis cujus nec divina intelligi pagina, nec scriptura aliqua. Sic nec rhetori-
cam, quae omnis eloquentiae tradit ornamenta, quibus maxime sacra Scriptura est referta, nec ejus
decor nisi his diligenter assignatis elucere poterit.'
2 Commentarius Cantabrigiensis in Epistolas Pauli e Schola Petri Abaelardi: In Epistolam ad
Romanos (ed. A. Landgraf, Notre Dame, 1937), pp. 1-2.
3 M. Grabmann, Die Geschichteder Scholastischen Methode, II, 349, n. 2. H. Denifle, Die Abend-
ldndischen Schriftauslegerbis Luther iiber Justitia (Rom. 1. 17) und Justificatio (Mainz, 1905), p. 76:
'Ad erudicionem autem ipsius omnes scripture facte sunt, quarum partes sunt tam sacre scripture,
quam ethnice. In ethnicis enim, id est gentilibus, scripturis et sermonum compositio et rerum pro-
prietas docet. Sermonum composicio in trivio, rerum proprietas in mathematicis disciplinis secundum
extrinseca et intrinseca. ... Intencionem vero more rethorice oracionis docere et monere.... '
4 Cf. G. Par6, A. Burnet, P. Tremblay, La Renaissance du XIIP Siecle: Les Ecoles et l'Enseignement
(Ottawa, 1933), pp. 267 ff., for an excellent statement of the rhetorical beginnings of the sententiae.
6Ibid. 286 ff., where, however, the'method is stated as dialectical. Cf. M. Grabmann, Die Geschichte
der Scholastischen Methode, I, 234 ff. and P. Fournier and G. Le Bras, Histoire des Collections Cano-
niques en Occident (Paris, 1932), ii, 334 ff. In the more orthodox tradition theology derived its cus-
tomary organization, indirectly from rhetoric, in Augustine's division of all doctrine into problems
of things and problems of signs; cf. P. Lombard, Sententiarum Liber I, dist. 1, cap. 1. The other dis-
tinction which Augustine makes at the beginning of the De Doctrina Christiana, of all treatment of
the Scriptures into the mode of discovery and the mode of statement, served as basis of organiza-
tion of treatises on preaching (cf. Bonaventura, Ars Concionandi [ed. Quaracchi, 1891], ix, 8.
2~2 in the Middle Ages
R~Rhetoric
on Romans, which is grammatical and rhetorical; but the texts such as those as-
sembled there serve him as a store house of quotations for his systematic works,
the Theologia 'Summi Boni,' the Theologia Christiana, and the Introductio ad
Theologiam,in which the method which Abailard calls dialectical is used to resolve
their differences, not by consideration of contexts and circumstances, but by
reduction to an orderly body of true propositions. The difference, far from being
slight, was to grow into one of the marks of differentiation between the line of
Christian theology which adapted itself to the Aristotelian philosophy and made
use of logic and dialectic and the line of Christian theology and philosophy which
continued the distinction of the trivium and subordinated dialectic to rhetoric.
One of the numerous admirers of Abailard who tried to remove the taint of un-
orthodoxy from his doctrines made that readjustment by shifting the functions
of the arts, assigning to grammar a concern with meanings, to dialectic the pro-
duction of conviction, and to rhetoric finally the motivation of the will.' This is a
doctrine, moreover, which need suffer no opprobrium because of its connection
with Abailard, since the same domination of the trivium by rhetoric is expressed,
partly in the same words, by Bernard Sylvester, the friend of Thierry of Chartres,
in his commentary on Virgil's Aeneid, a context which seems safe from the danger
of heresy.2
1 [Anonymous] Ysagoge in Theologzam(ed. A. Landgraf, Ecrits Theologiques de l'Ecole d'Abelard
[Louvain, 1934]), p. 72: 'Eloquentia vero est scientia ad congruam agnitorum prolationem suum
formans artificem. Que, quia triplicem habet efficatiam, tres habet partes, respondentque efficatie
partibus ut effectus causis. Est enim prima grammatica, que pertinet ad intellectum; secunda dia-
lectica, que ad fidem; tertia rethorica, que ad persuasionem. Quod enim prima vocum attendit acci-
dentia, ideo fit, ut secundum ea competens fiat earum contextus ad manifestandum conceptum
loquentis vel ad constituendum consimilem in auditore. Sed quia, si pulsetur de veritate, intellectus,
quem indicat et constituit, nequit fidem facere, succedit dialectica, que acceptis orationibus a prima
componit ex eis argumentationem, qua fidem confert. Sed quia possumus intelligere et intellectum
credere et tamen illud nolle, consummationem dat rethorica. Hec enim accipiens argumentaciones a
logica, ut illa orationes a grammatica, ex eis per orationem [I. perorationem] facit et, quod prima in-
telligere, secunda credere, ipsa facit velle.'
2 CommentumBernardi Silvestris super sex libros Eneidos Virgilii, lib. vi (ed. G. Riedel [Gryphis-
waldae, 1924], p. 31): 'Eloquentia est scientia formans suum lectorem ad congruam cognitorum
prolationem. Haec autem Trivia dicitur quia [a] tribus artibus quasi tribus viis ad eam incedimus.
Ut autem perfecte habeatur eloquentia, primo oportet scire loqui absque soloecismo et barbarismo
quod per grammaticam habetur. Deinde sic loquendo oportet scire aliquid probare vel improbare
quod fit per dialecticam. Adhuc necessarium [oportet] persuadere vel dissuadere: possunt enim
auditores grammatica oratione aliquid intelligere, dialectica probatione de eodem certi esse et tamen
illud nolle: ideo necessaria rethorica persuasio. Itaque est grammatica initium eloquentiae, dialectica
dicitur provectus, rethorica perfectio. Atque adeo dicitur eloquentia Trivia.' Cf. ibid., pp. 36, 38,
87-88. It would easily be possible to attach too much significance to the order in which the arts of the
trivium are enumerated; yet many of the enumerations of the twelfth and thirteenth centuries under-
line the importance of the order, and even before that time authors tend to a consistency in their enu-
merations which suggests that some degree of importance was attached to the enumeration. Dialectic
appears third in the lists of Cassiodorus (Institutiones, ii Praefatio, p. 91), Isidore (Etymologiae i. 9)
Alcuin (Grammatica, PL 101, 853: 'Sunt igitur gradus, quos quaeritis, et utinam tam ardentes sitis
semper ad ascendendum quam curiosi modo estis ad videndum: grammatica, rhetorica, dialec-
tica . . . ), Rhabanus Maurus (De ClericorumInstitutione iii. 18, PL 107, 395: 'Prima ergo liberalium
artium est grammatica, secunda rhetorica, tertia dialectica ...'). Rhetoric is third in Augustine
The two general tendencies which came to their culmination in the thirteenth
century, that by which rhetoric was made part of logic and that by which rhetoric
became an instrument of theology, are determined by the important methodo-
logical differences which separate the Aristotelians and the Augustinians. For
Thomas Aquinas rhetoric is one of the parts of logic concerned with probable
argumentation; for Bonaventura rhetoric is the culmination of the trivium.
Thomas wrote a commentary on two books of Aristotle's Organon, and since he
separated the method and subject of the philosophic from those of the theological
truth, he could use the devices of Aristotle in the a posteriori proofs of his sys-
tematic theology and those of Augustine in his commentaries on Scripture;
Bonaventura wrote no work on logic but did compose an excellent Art of Preach-
ing, which is useful for the interpretation of his theological treatises and com-
mentaries as well as his sermons.
The translation of the whole of the Nicomachean Ethics (which was called the
New Ethics in contrast to the truncated earlier translation) brought to further
refinement the solution of questions concerning the relation of rhetoric to civil
questions: according to Aquinas the matter with which rhetoric is concerned is
civil,1 but rhetoric must not be confused with politics.2 In much the same fashion
the terminology and conclusions of the earlier rhetorical discussion enter into
Thomas' classification of the parts of logic. The parts of logic or rational science
or rational philosophy are determined by the diversity of the acts of reason: they
are three of which the first is an act of immediate understanding and the last
two are acts of reason. The first is the operation of the mind called (by Averroes)
information of understanding or imagination through understanding; the doctrine
which Aristotle treats in the Categories is ordered to this act of reason. The
second is the operation of composition and division which results in truth or
falsity; the doctrine which Aristotle treats in the De Interpretationeis concerned
with this act of reason. Finally, the third act, which is the proper function of
reason, is discursive movement from one thing to another, from something known
to the unknown: the remaining four books of the Organonare concerned with this
operation of reason. It may take any of three forms in conformity to a threefold
diversity in nature: in some cases nature acts from necessity without the pos-
sibility of divergence, in some cases it operates for the most part in a certain way
but with the possibility of deviation from its proper act, and there are therefore
in addition to necessary operations two additional kinds of natural acts, those
(De Ordineii. 13, PL 32, 1013), Martianus Capella, Gerbert. After the eleventh century the tendency
is either to place the trivium, together with demonstration and sophistic, under logic, or to list the
three with rhetoric in the dominant position. The rule is far from being universal; e.g., Adelard of
Bath, who was strongly influenced by the Platonism of Chartres, places dialectic third in his allegory
(De Eodem et Diuerso [BGPM, Band Iv, Heft 1, p. 21]).
1 In Decem Libros Ethicorum Aristotelis ad Nicomachum Expositio i. Lectio 3 (ed. A. M. Pirotta
[Turin, 1934], n. 36, p. 12). Infallible proof is impossible in human affairs, and therefore the con-
jectural probability of the rhetorician is adequate; cf. Summa Theologica,i a, ii ae, q. 105, a. 2, ad 8:
'Ad octavum dicendum, quod in negotiis humanis non potest haberi demonstrativa probatio et
infallibilis, sed sufficit aliqua conjecturalis probabilitas secundum quam rhetor persuadet.'
2 Ibid. x. Lectio
16, n. 2173, p. 689.
which occur for the most part and those in which nature deviates from what is
proper to it. Corresponding to these there are three processes of reason: those by
which scientific certitude is acquired and in which no deviation from truth is
possible; those which come to conclusions true for the most part but not neces-
sary; and those in which reason deviates from the true because of some defect of
principle. The part of logic which treats the first of these processes is called
Judicative, since its judgment is made with the certitude of science, and this part
is treated in the Analytics: the Prior Analytics is concerned with the certitude of
judgment which is based only on theform of the syllogism, the Posterior Analytics
with the demonstrative syllogism in which the certitude depends on matter or on
the necessary propositions of which the syllogism is composed. The part of logic
which is subject to the second process of reason is called Inventive,for discovery is
not always with certitude. Topic or Dialectic treats of this process when it leads
to conviction or opinion (fides vel opinio); Rhetorictreats of it when it leads only
to a kind of suspicion without total exclusion of the contrary possibility; Poetic
treats of it when estimation inclines to one of the two parts of a contradiction
only because of the manner of its representation. Finally the third process of
reason is called Sophistic and is treated by Aristotle in the De Sophisticis Elenchis.1
Bonaventura's conception of rhetoric and logic, on the other hand, is quite
unaffected by the Aristotelian philosophy: they are ordered in the trivium, domi-
nated by rhetoric, and they are treated, with the other arts, by reduction to
theology, or as parts of the first vision of God which is by natural intelligence,
or as part of the gift of science which is one of the seven gifts of the Holy Spirit.
There are four lights by which we are illuminated in knowledge: the exterior light
of the mechanical arts, the inferior light of sensitive knowledge, the interior light
of philosophic knowledge, and the superior light of grace and sacred Scripture.
The interior light by which we are illuminated to intelligible truths is of three
kinds, rational, natural, and moral, corresponding to the traditional division of
the philosophic sciences into logic, physics, and ethics. Rational truth or the
truth of words is of three kinds, the expression of concepts (treated by grammar),
the movement to belief (treated by logic), and the movement to love or hate
(treated by rhetoric).2 The actual reduction of rational philosophy to theology is
accomplished by consideration of the speaker (his expression of the conception
of his mind is dependent on the eternal Word), his statement (in its congruity,
truth, and adornment, is seen the order of living, for actions by virtue of these
have measure [modus], beauty [species], and order [ordo]), and the hearer (in
whom the ends of speech are expressing, teaching, and moving, which are ac-
complished, as Augustine shows, only by the one true doctor who can impress
1 In Libros Posteriorum
Analyticorum Expositio i. Lectio 1 (ed. Rome, 1882), I, 138-140.
2 De Reductione Artium ad
Theologiam 4 (Opera Omnia [Quaracchi, 1891] v, 321): 'Et quoniam
tripliciter potest aliquis per sermonemexprimere quod habet apud se, ut scilicet notum faciat mentis
suae conceptum, vel ut amplius moveat ad credendum, vel ut moveat ad amorem, vel odium: ideo
sermocinalis sive rationalis philosophia triplicatur, scilicet in grammaticam, logicam et rhetoricam;
quarum prima est ad exprimendum, secunda ad docendum, tertia ad movendum. Prima respicit
rationem ut apprehensivam;secunda, ut iudicativam; tertia, ut motivam. Et quia ratio apprehendit
per sermonem congruum, iudicat per verum, movet per sermonem ornatum: hinc est, quod haec
triplex scientia has tres passiones circa sermonem considerat.'
species, infuse light, and give virtue to the heart of the hearer).' Or again, the
first vision of God, which is by natural intelligence, is divided into three rays,
since the light which is the truth of the soul illuminates the truth of things, of
signs, and of morals: the second irradiation of truth is divided into three parts:
grammar, logic, and rhetoric.2 The consideration of general and special forms of
argument in necessary matter as well as the consideration of 'topical places' (in
which induction proceeds by probable rather than necessary arguments) and
sophistical places falls within logic, while rhetoric is concerned once more with
civil utility and is divided into demonstrative, deliberative, and judicial.3 Or
again, the fifth gift of the Holy Spirit is science, comprising the three philosophic
sciences (rational, natural, moral), in all of which, including rational philosophy
or verbal science, Solomon was adept.4
It is in the platonizing Augustinian tradition, moreover, that music and poetry
assume a broad sense and dominant importance: Roger Bacon assigns to music
the function which Bonaventura ascribed to rhetoric, and then distinguishes both
rhetoric and poetic into two kinds, a theoretic rhetoric and poetic (or rhetorica
docens and poetica docens) which are parts of logic, and an applied rhetoric and
poetic (or rhetoricautens and poetica utens) which are parts of moral philosophy.'
The opposed tendencies which led to the dominance of rhetoric in the Augustinian
tradition and to the importance of logical demonstration in the Thomist tradition
are integral with the total complexions of the two theologies as evidenced in the
conclusion of Bonaventura that theology is neither theoretic nor practical but an
affective habit mid-way between theory and practice as opposed to the argument
of Thomas that theology subsumes both theoretic and practical sciences and is
itself more theoretic than practical.6 It is a distinction which later historians have
treated crudely by trying to differentiate 'voluntarism' from 'rationalism.'
V
Separate both from the tradition of the rhetoric assimilated to dialectic and
proof and that of the rhetoric assimilated to theology and edification - and the
1 Ibid. 15-18, pp. 323-324. 2 In Hexaemeron, Collatio iv, 18-25; v, 352-353.
3 Ibid. 20-21, pp. 352-353.
4 De
Septem Donis Spiritus Sancti, Collatio iv. 5-12; v, 474-475; esp. 8: 'Impossibile est, quod
sapientia fiat doctrina nisi per sermonem. Sermo autem non est sufficiens ad docendum, nisi sit
sententiosus. Et non loquitur homo sententiose, nisi sermo eius discussivus, inquisitivus, et persuasi-
vus, scilicet quod habeat sermonem potentem ad loquendum omne illud, quod potest apprehendi vel
nosci, vel ad quod affectus potest inclinari. Congrue autem exprimit quod dicit per grammaticam,
rationabiliter investigat per scientiam logicam et efficaciter persuadet per rhetoricam. Ista igitur est
pars philosophiae, scilicet scientia sermocinalis, quae triplex est, ut patet, quam adeptus est Salomon.'
6 Opus Tertium, cap. 75 (Opera Inedita, ed. J. S. Brewer [London, 1859]), pp. 303-308, esp. 306-
307: 'Nam moralis philosophus scit uti sermone suavi, et gestibus convenientibus orationi delectabili
conformandis. Similiter logicus et grammaticus.... Grammaticus igitur utitur his pueriliter; sed
logicus quantum ad formam arguendi quam constituit, in his procedit viriliter, et causas et rationes
assignat. Sed quantum ad decorem et ornatum et suavitatem argumenti, certe non potest logicus,
sicut nec grammaticus, causas et rationes assignare, sed musicus; sicut geometer causas linearum,
et angulorum, et figurarum, quibus utitur carpentator, habet dare.' Cf. Opus Majus iii and iv. 2 (ed.
J. H. Bridges [Oxford, 1897]), pp. 71 and 99-102.
6 Bonaventura, Proemium in Librum Primum Sententiarum q. 3 concl.; I, 13; Thomas Aquinas,
Summa Theologicai. q. i, a. 4.
object of suspicion and attack by both - a third tradition of rhetoric seems to

have flourished, at least during the second and third periods of the other two
traditions, indifferent alike to the logical differentiation of necessary and probable
arguments and the theological limitation of persuasion to profound or salubrious
truths. Since the three traditions engaged in a three-cornered dispute there is no
single statement of the issue, for to logicians the practitioners of this new art
seemed sophists, while theologians lumped them with the heretical dialecticians
and garrulous ratiocinators; from the point of view of the new art, which professed
an exclusive concern with practical issues and effective applications, that is, with
actions or with words, the rules of logic were themselves open to question, and
visionary theory and inapplicable generalization were devoid of moral attraction.
For the most part we know about the early members of this tradition from the
violence of the attacks upon them and the bitterness of the satire in which they
were portrayed, but gradually in the course of the twelfth and thirteenth cen-
turies they limited their statements to figures and forms of words, accomplishing
their practical objectives by that device in a fashion which met with little effective
opposition from logicians or theologians: and since they were unhampered by the
need to consider things or thoughts, they were prolific in production of the 'new'
methods - they were fond of calling themselves moderni - which constituted
one of the important guises in which rhetoric entered the fourteenth century and
the Renaissance.
A few fragments of the works of Anselm the Peripatetic are the only remains
of the 'Drogonic' sect - followers of the philosopher Drogo - which Anselm
would have us believe was numerous and influential. He calls his art rhetoric; he
professes allegiance to that art along with Hermagoras, Tully, Servius, Quintilian,
Victorinus, Grillius, and Boethius, and thought to illustrate it in his treatise De
Materia Artis (now lost) and in the examples of rhetoric set forth in his Rhetori-
machia; he specifies that rhetoric demonstrates, not truths, but verisimilitudes
disguised as truths.' The Rhetorimachiais divided into three parts, one devoted
to each of the genera, demonstrative, deliberative, and judicial, and each example
takes the form of an attempt to turn the arguments of an opponent against him.
The bearing of Anselm's performance on logic is not far to seek, since his approach
permits him to deny the principle of excluded middle,2 while its relation to theol-
ogy is no less apparent from the allegory of the dream, related as part of his
treatment of deliberative rhetoric, in which the embraces and arguments of three
virgins named Dialectic, Rhetoric, and Grammar turn him from communion
with the saints in Heaven.3 A 'sophist' named John seems in like fashion to have
had a numerous following of whom Roscelin of Compiegne alone is easily identi-
fiable.4 In the twelfth century John of Salisbury attacks the doctrine of a teacher
1 Epistola ad DrogonemPhylosophum (Anselm der Peripatetiker, nebstandern Beitragen zur Literatur-
geschichteItaliens Eilften Jahrhundert, ed. E. Diimmler [Halle, 1872], pp. 19-20).
2 Rhetorimachia i, p. 34; cf.
Epistola ad Drogonem Magistrum et Condiscipulos de Logica Dis-
putatione in Gallia Habita, pp. 56-58. 3 Rhetorimachiaii, p. 42.
4 Cf. Historia Francica
(quoted by J. Reiners, Der Nominalismus in der Friihscholastik [BGPM,
Band viII, Heft 5], p. 33, n. 9): 'In dialectica quoque hi potentes extiterunt sophistae: Joannes, qui
eandem artem sophisticam vocalem esse disseruit, Rotbertus Parisiacensis, Roscelinus Compendi-
ensis, Arnulfus Laudenensis. Hi Joannis fuerunt sectatores, qui etiam quamplures habuerunt audi-
whom he disguises under the name of Cornificius (allying with himself in the at-
tack the most illustrious masters of the age, among others Gilbert de la Porree,
Thierry of Chartres, William of Conches, Peter Abailard) who broke that union
of wisdom and eloquence which is the foundation of philosophy, of society, and
of morals, and who made everything new in his teaching, innovating in grammar,
modifying dialectic, despising rhetoric; his exclusive reliance on the precepts of
eloquence apparently leads Cornificius to exploit the traditional puzzles of the
sophists which turn on the confusion of word and thing or the application of a
word or statement to itself.'
This tradition of rhetoric took form, for the most part, not in controversy or
theory but in a vast number of textbooks which grew in three distinct groups dif-
ferentiated according to the subject matters once treated by rhetoric but now
concerned with verbal forms employed in those three fields in lieu of direct treat-
ment of subject matter. First, rhetoric had contributed to the method of studying
law, but the substantive consideration of law had moved into theology and had
taken with it most of the appurtenances which might have made the law a learned
profession, leaving only the verbal rhetoric of the dictamen.2Second, the art of
preaching which had assumed in the Christian tradition an exhortative function
approximating that of ancient deliberative oratory - once due allowance is
tores.' Cf. J. A. Endres, Forschungen zur Geschichteder friihmittelalterlichen Philosophie (BGPM,
Band xvII, Heft 2-3), and Petrus Damiani und die weltlicheWissenschaft (BGPM, Band vmII,Heft 3);
also J. de Ghellinck, 'Dialectique et dogme aux Xe-XIIe siecles' (BGPM, Supplementband I, pp.
79-99).
1 John of Salisbury, Metalogicon i. 1-6, pp. 5-Q1, esp. 21: 'Plane eloquentie precepta sapientiam non
conferunt; sed nec amorem eius et sepissime quidem ei obtinende non conferunt. Res enim philosophia
(aut finis eius, que est sapientia) querit, non uerba. Ex his itaque liquet quia precepta eloquentie ab
operis suis philosophia eliminat.' Cf. the confusion of the arts of the trivium reported by Hugh of St
Victor, Didascalicon iii. 5, pp. 55-57. Cf. P. Mandonnet, Siger de Brabant et l'Averro'sme Latin au
XIIIme Siecle (2nd ed., Louvain, 1911), I, 122-123.
2 For the voluminous literature on the Ars Dictaminis and Ars Notaria, cf. L. J. Paetow, A Guide
to the Study of Medieval History (2nd ed., New York, 1931), pp. 448-452; for the relation of these arts
to rhetoric, cf. N. Valois, De Arte Scribendi Epistolas apud Gallicos Medii Aevi Scriptores Rhetoresve
(Bibliothequede l'Ecole de Chartres,22 [1880], 161, 257); for the relation of rhetoric to the teaching of
law, cf. P. Abelson, The Seven Liberal Arts, pp. 60-66. The manner of the change, no less than the
pride in the novelty of it, may be judged from the contents of Boncompagni's two works, the Rhetorica
Antiqua (arranged in six books according to the character of the letter to be written) and the Rhetorica
novissima (arranged in thirteen books: 'Primus est de origine iuris. Secundus est de rhetoricae partibus
et causarum generibus. Tertius est de diffinitionibus. Quartus est de naturis et consuetudinibus
oratorum. Quintus de causarum exordiis. Sextus de principiis conuentorum. Septimus de rhetoricis
argumentis. Octavus de memoria. Nonus de adornationibus. Decimus de invectivis. Undecimus de
consiliis. Duodecimus de colloquiis. Tertius decimus de conditionibus'). Boncompagni professes in the
prologue to the former work not to remember ever having read Cicero, but he adds that he never
dissuaded anyone who wanted to read him; and in the latter work he gives three reasons why he
undertook to find a new rhetoric after Cicero had compiled a rhetoric from the infinite precepts of
rhetoricians: (1) according to Boethius the rhetoric edited by the ancients consists solely of precepts,
without doctrine or utility, (2) students in civil and canon law would not get a solid foundation in the
liberal arts, (3) Cicero's rhetoric is rendered void according to students of law because it is never read
in 'ordinary' courses, but is run through and taught like a mechanical art by stealth; to these he adds
a fourth: that Cicero was mistaken about the origin of the law. (Cf. L. Rockinger, 'Uber die Ars
Dictandi und die Summae dictaminum in Italien,' Sitzungsberchte der Konigl. Bayerischen Akademie
der Wissenschaften zu Miinchen, hist. Kl., I [1861], 135-145.) For the closely related art of pleading,
made for differences between the terrestrial and celestial city - gradually moved
to a formalism in which doctrine was left to theology and attention was centered
on three problems: propriety of division of the subject stated in the theme of the
sermon, brevity of distinction, and utility of expansion.1Finally, the art of poetry
came to be considered after the twelfth century, not a branch of grammar, but
alternately a kind of argumentation or persuasion (and as such subordinate to
logic or morals) and a form of composition (and as such to be treated in terms of
style, organization, and figures borrowed from rhetoric).2 In common, these three
tendencies continue the terms and some points of the organization of the Ad Heren-
cf. A. Wilmart, 'L"'Ars arengandi" de Jacques de Dinant avec un Appendice sur ses Ouvrages "De
Dictamine",' Analecta Reginensia (Vatican City, 1933), pp. 113-151. The relations of rhetoric to
law and logic are reflected satirically in the 'battle of the seven arts'; cf. 'La Bataille des vii Ars of
Henri d'Audeli,' ed. L. J. Paetow, Two Medieval Satires on the University of Paris (Berkeley, Calif.,
1927), pp. 43 and 51. Cf. H. Kantorowicz, Studies in the Glossatorsof the Roman Law (Cambridge, 1938).
1 Cf. the anonymous Art of Preaching, portions of which are edited in the Opera Omnia S. Bon-
venturae(ix, 6-7), in which four modes of preaching are distinguished: (1) that which concords really
and verbally with the words of Scripture - used by 'modern' doctors and expounded in this treatise
- (2) that which employs only real concordance with Scripture - appropriate to those newly learned
in theology - (3) that limited to verbal concordance, and (4) contrasted to the modern method, the
ancient mode 'quod observant antiqui Sancti, sicut Augustinus et Bernardus et multi alii, quorum
sermones in Ecclesia recitantur, in quibus non proponitur aliquod thema, quod sit materia praedi-
candi, nec solent divisiones vel distinctiones fieri, quae postmodum concordentur, sed quasi narrative
procedit.' The modern doctors advise against following this mode for the curious reason that these
Fathers were, in a manner, founders of the Church (quasi Ecclesiae fundatores), and therefore they
avoided all curiosity concerning distinctions of themes and subdivisions of members and concordances
of both. Bonaventura likewise divides the problems of preaching into three parts, divisiones, dis-
tinctiones, dilatationes (cf. Ars concionandi, Prooemium, ix, 8). For an excellent exposition of the
technique of the mediaeval sermon and review of the methods expounded in most important medi-
aeval handbooks, cf. E. Gilson, 'Michel Menot et la Technique du Sermon Medieval,' Les Idees et les
Lettres (Paris, 1932), pp. 93-154. Cf. H. Caplan, 'Classical Rhetoric and the Mediaeval Theory of
Preaching,' Classical Philology, xxvII (1933), 73-96, 'Rhetorical Invention in Some Mediaeval
Tractates on Preaching,' SPECULUM,II (1927), 284-295), 'Henry of Hesse on the Art of Preaching,'
PMLA XLVIII(1933), 340-361. The treatises of Robert of Basevorn and Thomas of Wales are pub-
lished in T.-M. Charland, Artes Praedicandi, Contributiona l'Histoire de la Rhetoriqueau Moyen Age
(Ottawa, 1936), preceded by a survey of writers of Arts and the customary form of theory. Cf. the
differentiation of the two modes most used by moderns, the French and the English (Robert of
Basevorn, Forma Praedicandi 7, p. 244). Cf. also M. M. Davy, Les Sermons Universitaires Parisiens
de 1230-1231 (Paris, 1931), G. R. Owst, Preaching in Medieval England, an Introduction to Sermon
Manuscripts of the Period c. 1350-1450 (Cambridge, 1926), and C. H. Haskins, Studies in Mediaeval
Culture(Oxford, 1929), pp. 36-71.
2 In early treatments poetry, considered as metric, was a part of grammar, while as a form of
argument it was a part of topic or dialectic. Thus Cassiodorus defines grammar (Institutiones ii. 1. 1.
p. 94): 'grammatica vero est peritia pulchre loquendi ex poetis illustribus auctoribusque collecta;
officium eius est sine vitio dictionem prosalem metricamque componere,' but he includes the poets
among the artists to whom topical arguments are supplied by memory (ibid. 3. 17, p. 127; cf. ibid.
P. 2. p. 98, for the function of memory in discovery). Cf. Isidore of Seville, Etymologiae i. 39; the
Venerable Bede, De Arte Metrica (PL 90, 149). John of Salisbury notes the tendency to make poetic
an art by itself or to assimilate it to rhetoric rather than to grammar, but he is explicit in his own
resolution of the problem; cf. Metalogicon i. 17, p. 43: 'Profecto aut poeticam grammatica obtinebit,
aut poetica a numero liberalium disciplinarum eliminabitur.' Cf. C. Fierville, Une GrammaireLatine
In6dite du XIII Siecle (Paris, 1886), pp. 94-119. The transition is gradual from a consideration
merely of the words, their character, and position to the consideration of the general conditions or
nium and of Cicero's De Inventione, but the commonplaces which have been
put to so many uses are no longer devices for discovering arguments of things
and their traits, but devices for remembering, for amplifying, for describing,
and for constructing figures.'
VI
Two translations of Aristotle's Rhetoric were produced during the thirteenth
century, and there were also translations of the Rhetorica ad Alexandrum, Aver-
roes' commentary on the Rhetoric,and Demetrius' De Elocutione. The effect of the
Aristotelian rhetoric and its variant interpretations (both Demetrius and Aver-
roes passed as 'Aristotelian') on philosophy may be judged from the fact that
these works on rhetoric are frequently found in manuscripts which contain
works on morals, politics, or economics, and indeed, specific marks of the Rhetoric
can be seen in Aquinas' analysis of the passions.2 Yet there are relatively few
early commentaries on the work itself: Aegidius Romanus in the thirteenth
century, and John of Jandun and John Buridan in the fourteenth century
are the only outstanding scholastics to have left such commentaries.3 The old
problem of the genus of rhetoric, whether it is a part of civil philosophy or
logic, is resolved by Aegidius into the difference between Aristotle (who placed
it under dialectic) and Cicero (who made it a part of politics).4 The position
of this pupil of St Thomas is indeed almost a parody of Bonaventura's doc-
trine that theology is midway between the practical and speculative sciences,
for he locates rhetoric midway between the moral and the rational sciences.6
The readjustment is striking illustration of the fashion in which unchanged
analyses may in the context of altered philosophies take on contrary signifi-
cances, for the effort of rhetoricians from Quintilian through the early Middle
Ages was to claim consideration of general or indefinite questions or theses
and to resist efforts to restrict rhetoric to determinate questions or hypotheses
lest rhetoric yield its place and importance as a science to philosophy: the dif-
ference between politics and ethics on the one hand and rhetoric on the other,
according to Aegidius, consists in the fact that a science is determined by its
subject matter and that, whereas politics and ethics have a determinate genus,
rhetoric is indeterminate, being concerned with knowledge of certain common
notions which bear on moral questions. John Buridan divides all sciences into
two kinds, the 'principal' science which deals with the proper things of the
places relevant to the choice and disposition of words; a further step is needed to carry it, during the
Renaissance, from the figures of speech and the figures of doctrine to the rhetorical consideration of
the thoughts of the author and the effects on the audience.
1 Cf. E. Faral, Les Arts Pogtiques du XIIe et du XIIIe Siecle (Paris, 1924), esp. pp. 52 ff. and 77 ff.
2 M.
Grabmann, 'Eine lateinische tYbersetzung der pseudo-aristotelischen Rhetorica ad Alex-
andrum,' pp. 6 if.; G. Lacombe, A. Birkenmajer, M. Dulong, A. Franceschini, Aristoteles Latinus
(Rome, 1939), I, 77-79; B. V. Wall, A Medieval Latin Version of Demetrius' De Elocutione (Wash-
ington, 1937).
3 The commentaries of Jandun and Buridan are unedited; that of Aegidius was published in 1515
in Venice, but I have been unable to consult a copy.
4 De
Differentia Rhetoricae, Ethicae, et Politicae (ed. G. Bruni, The New Scholasticism, vi [1932]),
pp. 5-8.
6 Ibid., p. 2. Cf. Expositio in Artem Veterem (Venice, 1507), 2v-3r, where speculative science is
divided into principalis (concerned with things) and adminiculativa (the three arts of the trivium).
science, and the 'instrumental' science which is concerned with the mode of
statement and teaching. The instrument of the theoretic sciences is logic or
dialectic, but in moral science the problem involved concerns not only the doubt-
ful and the true, but also the need to stir desire as it bears on understanding, and
a special moral logic or dialectic is required which is divided into two parts,
rhetoric and poetic.l John of Jandun divides philosophy into non-organic (practi-
cal and speculative) and organic, which includes grammar and logic, rhetoric
being a subdivision of logic.2
The three main lines in which rhetoric developed during the Middle Ages - as
they had grown out of philosophic oppositions in antiquity and as they had been
continued by mediaeval writers under the compulsion of the circumstances and
nature of the problems they treated - are extended through the discussions of
the Renaissance, notwithstanding revolt against the scholasticism of the Middle
Ages, alike by the weight of tradition and by the exigencies of the problems them-
selves. The tradition of rhetoric as a part of rational philosophy subordinate to
logic had a long and honorable continuation which included Zabarella, Campa-
nella, Varchi, Robertelli, and many others.3 The tradition in which rhetoric
dominated the arts continued into the Renaissance not only in the methods and
1 Questiones super Decem Libros Ethicorum Aristotelis ad Nicomachum, Prooemium (Paris, 1518),
fol. 4r.
2 Quaestiones Subtilissimae super Tres Libros de Anima, Prohemium (Venice, 1519), fol. v.
3 According to Zabarella (De Natura Logicae ii. 13-23 [Opera Logica, Cologne, 1597, pp. 78-100])
rhetoric and poetic are instruments of civil discipline and parts of logic (the arts of demonstration,
dialectic, and sophistic are also parts of logic); like logic they are rational faculties, not verbal like
grammar. Logic is divided into two principal kinds, universal and particular; rhetoric and poetic are
instances of particular logic. Campanella divided his Philosophia Rationalis into four parts: Grammar,
Dialectic, Rhetoric, and Poetic. Rhetoric is a part of rational philosophy deriving its arguments from
dialectic and its matter from morals; it does not treat of all questions, but is limited to persuasion
and dissuasion of good and evil; poetic has the same function, but it differs from rhetoric in its uni-
versality, since it presents all goods and all truths to all audiences (Philosophia Rationalis, Pars Tertia,
Rhetorica 1.1. [Paris, 1538], pp. 1-7; cf. Pars Quarta, Poetica 1.1. pp. 89-93). B. Varchi follows the
traditional division of philosophy into real, active, and rational; rhetoric and poetic are subdivisions
of rational philosophy, although strictly speaking poetic is neither an art nor a science, but a faculty;
dialectic, rhetoric, and poetic are essentially the same thing, differing only accidentally, and the
dialectician, rhetorician, and poet can be put on the same level of nobility and honor; cf. 'Della
Poetica in Generale,' Operedi BenedettoYarchi (Trieste, 1859), II, 684: 'La filosofia razionale, la quale
favellando di parole e non di cose, non e veramente parte della filosofia, ma strumento, comprende
sotto se non solo la loica (intendendo per loica la giudiziale) e la dialettica (intendendo per dialettica
non tanto la topica, quanto eziandio la sofistica e la tentativa) ma ancora la rettorica, la, poetica, la
storica e la grammatica.' Robertelli raises the question, not in terms of the form of the art but in
terms of its matter and end: poetic shares its matter, oratio, with four other disciplines: demonstra-
tion, dialectic, rhetoric, and sophistic; grammar is excluded from the list since it does not involve the
intellectual content of what is said. The five are easily and rapidly distinguished (In Librum Aristotelis
de Arte Poetica [Basel, 1555], p. 1): 'Ex his quaelibet facultas unum arripit genus. Demonstratoria
verum. Dialectice probabile. Rhetorica suasorium. Sophistice id, quod probabilis, sed verisimilis habet
speciem. Poetice falsum, seu fabulosum.' Its end (ibid. 2) is the 'imitating word,' as the end of rhetoric
is the 'persuading word'; it is (borrowing from Cicero) the imitation of life, the mirror of custom, the
image of truth. Cf. H. Cornacchinus, Indagatio Yerae et Perfectae Definitionis Logicae, Pars V, cap. 21
(Padua, 1606), p. 247: poetic and rhetoric are parts or offshoots of logic, or rather aggregates com-
posed from logic, grammar, and civil philosophy, and (ibid. Pars iv, cap. 10, pp. 220-221) dialectic,
sophistic, and rhetoric are midway between grammar and logic.
doctrines of theology but in a secular tradition which took one of two forms: either
all philosophy and all subjects are assimilated to rhetoric, as in the doctrines of
Majoragius and Nizolius,l or the method of discovery is refurbished and trans-
ferred from rhetoric to revitalize and revolutionize dialectic, as in the doctrines of
Rudolph Agricola and Petrus Ramus.2 The tradition in which rhetoric had be-
come a discipline of words, independent alike of philosophy and dialectic, finally,
established verbal distinctions which grew into doctrines of things: the long and
subtle speculations of fourteenth-century philosophers on insolubilia, obligatoria,
and sophisms laid the foundations for many of the early theories in physics and
mathematics, and symbolic logic, though unconcerned with its past, still repeats
the elements of this heritage; the analysis of the figures of the poet was made,
without undue or violent alteration, into a theory of poetry which dealt with
imagination, passion, truth, and virtue; and political philosophy has never en-
tirely lost the rhetorical turn from which its theories derived their modern con-
creteness and practicality.
Once the general movements in the arts, of which the variegated history of
rhetoric is a symptom, have been set into some intelligible schema, the startling
and revolutionary shifts of doctrines and of problems are more easily understood.
1 J. L. Vives (De Causis CorruptarumArtium, Liber iv [Lugduni Batavorum, 1586], p. 239) reports
the doctrine of philosophers who distinguish two rhetorics, one universal and applicable to all things,
the other particular and suited to civil use; Vives interprets the position as being in opposition to the
tendency to make rhetoric part of logic. The position is defended by M. A. Majoragius on the au-
thority of Cicero (De Finibus ii. 6. 17) against Aristotle (Aristotelis Stagyritae De Arte RhetoricaLibri
Tres cum M. Antonii Maioragii Commentariis, Liber I [Venice, 1591], p. 2). M. Nizolius holds, again
on the authority of Cicero, that philosophy and oratory are not two separate faculties but one and the
same art, composed of two arts which are imperfect when separated (De Yeris Principiis et Vera
Ratione Philosophandi contra Pseudophilosophos, Liber II, cap. 3 [Parma, 1553], p. 911); he quotes
Laurentius Valla, with approval, when he argues that dialectic is a part of rhetoric, since it consists
of only one of the five parts of rhetoric, namely discovery (ibid., cap. 5, p. 240); and finally he holds
that rhetoric is a general art and science under which are subsumed all other arts and sciences (ibid.
III, cap. 8, p. 268). The distinction of the two rhetorics, the rhetoric of precepts and the rhetoric in
use, is preserved by Riccoboni, who also adds 'ecclesiastical' as a fourth genus to the traditional
three, 'deliberative,' 'demonstrative,' and 'judicial' (A. Riccobonus, De Usu Artis Rhetoricae Aris-
totelis Commentarii Vigintiquinque, Quibus Duplex, Rhetorica Strictim Explicatus, Altera, Quae
Praecepta Tradit Persuadendi, Altera, Quae re ipsa persuadet, etc. [Frankfurt, 1595]). The use of
rhetoric in refurbishing Scriptural interpretation is well illustrated in the Heptaplus of Pico della
Mirandola (in which Moses emerges as the 'Idea' of the writer, the exemplar of the prophet) and John
Colet's Enarrationes in Epistolas S. Pauli.
2
Rudolph Agricola undertook to reinstate in dialectic the processes of discovery which had become
part of rhetoric because civil philosophy came into prominence in Greece before the maturity of the
other arts (De Inventione Dialectica, Liber II, cap. 18 [Cologne, 1588], pp. 538 ff.), and to correct the
errors which Aristotle, Cicero, and Boethius had committed in treating and classifying the places. The
function of rhetoric was limited to ratio dicendi. According to Petrus Ramus logic or dialectic is a
general art, the whole art of reason (Scholae in Liberales Artes [Basel, 1569], Scholae Dialecticae, Liber
ii, cap. i, pp. 35-37). The parts of dialectic are discovery and judgment (ibid., cap. 8, p. 53); the
parts of rhetoric are elocution and action (Scholae Rhetoricae,Liber I, p. Q38). The logic of Aristotle
abounded in errors, confusions, vain precepts, and altercations: Ramus professed to have supplied
the missing virtues, removed the errors, and to have made the art usable. The error of Cicero consisted
in transferring all the Aristotelian devices of dialectic to rhetoric and of having made one art of two;
and Quintilian mixed rhetoric with all the other arts; Ramus undertook to correct both errors (Rhe-
toricae Distinctiones in Quintilianum [Paris, 1559], pp. 3-8).
Rhetoricin the Middle Ages
Since the problems of the sciences and the arts are closely related and are often
stated in almost identical language, a slight shift of theory or terminology may
at a point bring an unsuspected richness from one art into the threadbare termi-
nology of another. The three customary questions of rhetoric, whetherit is, what it
is, and what sort, merged readily with the questions of logic and influenced early
modern attempts to formulate the scientific method. The customary rhetorical
inquiry into the duty of the artist, and the matter and end of the art, took on
metaphysical generality when it was merged, in the thirteenth century, with the
Aristotelian causes by the simple addition of questions of form to what were
already questions concerning the efficient, material, and final causes; and meta-
physics apart, the four questions contributed to the foundations of philology in
the inquiries into the four causes of books with which Aquinas and Bonaventura
and other mediaeval writers opened their commentaries. The controversy concern-
ing thesis and hypothesis merged with Plato's dialectical use of hypothesis and
Aristotle's differentiation of thesis, hypothesis, and definition, and contributed un-
suspected commitments and implications in modern discussions of scientific
method. Rhetoric is at most an unusually clear example among the arts and sci-
ences of a tendency which is possible in the history of rhetoric only because it is
universal in intellectual disciplines. In application, the art of rhetoric contributed
during the period from the fourth to the fourteenth century not only to the meth-
ods of speaking and writing well, of composing letters and petitions, sermons and
prayers, legal documents and briefs, poetry and prose, but to the canons of in-
terpreting laws and scripture, to the dialectical devices of discovery and proof,
to the establishment of the scholastic method which was to come into universal
use in philosophy and theology, and finally to the formulation of scientific in-
quiry which was to separate philosophy from theology. In manner of application,
the art of rhetoric was the source both of doctrines which have long since become
the property of other sciences (such as the passions, which were considered in
handbooks of rhetoric until Descartes proposed a 'scientific' treatment of them
different only in details) and of particular devices which have been applied to a
variety of subjects (such as to the 'common-places,' which were sometimes
techniques for inventing arguments, sometimes means for dilating statements,
sometimes methods for discovering things, or to 'definition' or 'order' which may
be determined entirely by consideration of the verbal conditions of expresssion,
the psychological requirements of persuasion, or the circumstantial probabilities
of fact). In theory of application, the art of rhetoric was now identified with, now
distinguished from, the whole or part not only of grammar, logic, and dialectic
(which were in turn distinguished from or identified with each other), but also
of sophistic and science, of 'civil philosophy,' psychology, law, and literature,
and finally of philosophy as such. Yet if rhetoric is defined in terms of a single
subject matter - such as style, or literature, or discourse - it has no history dur-
ing the Middle Ages; the many innovations which are recorded during that period
in the arts with which it is related suggest that their histories might profitably
be considered without unique attachment to the field in which their advances
are celebrated.
THE UNIVERSITY OF CHICAGO.

MCK RhetoricInMiddleAges

Uploaded by

Copyright:

Available Formats

MCK RhetoricInMiddleAges

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MCK RhetoricInMiddleAges

Uploaded by

Copyright:

Available Formats

Reproduced with permission of the University of Chicago Press.

For personal, noncommercial use only.

RHETORIC IN THE MIDDLE AGES1

There is little reflection in the histories of rhetoric of the differences concerning

tions in the contents and doctrines of rhetoric follow on these differences in

of rhetoric, on 'things' or on 'signs.'l This rhetorical language has, however, been

Porphyry, for treatment of a 'scientia dividendi' in which Aristotle himself placed

particularized to circumstances; dialectic uses interrogation and response, and its

an Introduction precede the fragment of the Commentarywhich has been pre-

object of suspicion and attack by both - a third tradition of rhetoric seems to

You might also like