Aristotle on the Disciplines of Argument Rhetoric, Dialectic, Analytic - J. Allen

8/11/2019 Aristotle on the Disciplines of Argument Rhetoric, Dialectic, Analytic - J. Allen

1/23

Aristotle on the Disciplines of Argument: Rhetoric, Dialectic, AnalyticAuthor(s): James AllenSource: Rhetorica: A Journal of the History of Rhetoric, Vol. 25, No. 1 (Winter 2007), pp. 87-108Published by: University of California Presson behalf of the International Society for the Historyof RhetoricStable URL: http://www.jstor.org/stable/10.1525/rh.2007.25.1.87.

Accessed: 21/09/2014 02:48

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at.http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

.

University of California PressandInternational Society for the History of Rhetoricare collaborating with

JSTOR to digitize, preserve and extend access toRhetorica: A Journal of the History of Rhetoric.

http://www.jstor.org

This content downloaded from 192.167.204.6 on Sun, 21 Sep 2014 02:48:48 AMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/action/showPublisher?publisherCode=ucalhttp://www.jstor.org/action/showPublisher?publisherCode=ishrhttp://www.jstor.org/action/showPublisher?publisherCode=ishrhttp://www.jstor.org/stable/10.1525/rh.2007.25.1.87?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/10.1525/rh.2007.25.1.87?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=ishrhttp://www.jstor.org/action/showPublisher?publisherCode=ishrhttp://www.jstor.org/action/showPublisher?publisherCode=ucal


2/23

James Allen

87

Rhetorica, Vol. XXV, Issue 1, pp. 87108, ISSN 0734-8584, electronic ISSN 1533-8541. 2007 by The International Society for the History of Rhetoric. All rights re-served. Please direct all requests for permission to photocopy or reproduce articlecontent through the University of California Presss Rights and Permissions website,at http://www.ucpressjournals.com/reprintInfo.asp. DOI: RH.2007.25.1.87.

Aristotle on the Disciplines of Argument:Rhetoric, Dialectic, Analytic

Abstract: According to an argument made by other authors, analytic

the formal logical theory of the categorical syllogism expounded

in the Prior Analyticsis a relatively late development in Aristotlesthinking about argument. As a general theory of validity, it served

as the master discipline of argument in Aristotles mature thought

about the subject. The object of this paper is to explore his early

conception of the relations between the argumentative disciplines.

Its principal thesis, based chiefly on evidence about the relation

between dialectic and rhetoric, is that before the advent of analytic

dialectic played a double role. It was both the art or discipline of one

practice of argumentation and the master discipline of argument

to which other disciplines turned for their understanding of thefundamentals of argument.

As everyone knows, Aristotle invented logic. The Prior

Analyticstackles the question when and in virtue of whatan argument is valid in an entirely new way. To be more

precise, Aristotle is concerned not with everything that might calledan argument, but with the syllogism, i.e., an argument () inwhich, certain things being laid down, something different from

I am grateful to the participants and organizers of the conference on Philosophy andRhetoric in Classical Athens for an exceptionally stimulating gathering and for manyhelpful comments. I owe a special debt to Chloe Balla for valuable written commentson an earlier version of this paper. Papers related to this one were presented at theHumboldt Universitat Berlin, McGill University, the University of Toronto, and theCentral division meeting APA in 2001. I learned a great deal from the audience on eachof these occasions, and I am especially indebted to Gisela Striker, my commentatorat the APA meeting.

This content downloaded from 192.167.204.6 on Sun, 21 Sep 2014 02:48:48 AMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp


3/23

R H E T O R I C A88

them follows of necessity by their being so (Topics 1.1.100a257;Prior Analytics 1.1.24b1820). This definition marks off a class of

valid arguments that might be of use to someone actually mak-ing a case for something, for example, by excluding argumentswhose conclusion already figures among its premises. The prin-cipal thesis of Aristotles theory is that every syllogism, i.e., ev-ery argument satisfying this definition, is a categorical syllogism,meaning an argument that belongs, or consists of parts that be-long, to one of the moods of the categorical syllogism, which weknow under their medieval names, Barbara, Celarent, Darii,and so on.

To employ language that Aristotle does not use himself, hisanswer to the question is that syllogisms are valid if and only ifthey are formally valid, and they are formally valid if and only ifthey can be shown by analysis to be categorical syllogisms. It is fromthe operation of analyzing arguments into categorical form that the

Analyticstake their name (cf.Prior Analytics1.32.46b3847a5).Aristotle has notoriously little to say about where this logical

theory belongs in his classification of the sciences. At the beginningof thePrior Analytics, he emphasizes that the study of the syllogism

on which he is about to embark is an essential preparation for thestudy of demonstrative syllogisms and the kind of knowledge orunderstanding that one has by grasping them, which will occupyhim in the Posterior Analytics (1.1.24a12; cf. 4.25b2631).1 In Prior

Analytics2.23, however, he asserts that:

Not only are dialectical and demonstrative syllogisms brought about bymeans of the figures [of the categorical syllogism] but also rhetoricalsyllogisms and, quite generally, any attempt to produce conviction() of any kind whatever.

68b914

And this is only one of a number of passages in which Aristotle insistson the universal application of the categorical theory to argumentsof any and every kind wherever they may be found (Prior Analytics1.23.40b20 ff., 41b15; 25, 42a301; 28.44b68; 29.45b3646a2; 30.46a34). Let us call the discipline to which the theory of the categoricalsyllogism belongs analytic, even though this term has, at best, only

1J. Brunschwig, Lobjet et la structure des seconds analytiquesdapres Aristote,in E. Berti, ed., Aristotle on Science: the Posterior Analytics (Padua: Antenore, 1981),6196.



4/23

Aristotle on the Disciplines of Argument 89

a slender basis in Aristotles own usage.2 If, as Aristotle maintains,every syllogism is a categorical syllogism, then every discipline in

which argument plays a part relies on principles that it is the businessof analytic to study; and, to the extent that a discipline requires anexplicit understanding of the fundamentals of argument, it must turnto analytic. Aristotle has in effect made analytic the master disciplineof argument.

As such, it can be contrasted with two other kinds of disciplinethat concern themselves with argument. On the one hand, there arearts ofargumentlikedialecticandrhetoric,whoseaimistofurnishthecorrespondingpractices of dialectical and rhetorical argument with

a system or method. The Topics contains a method for the practitionerof dialectic; theRhetoric, a method for the orator. On the other hand,there are the special sciences. The material discussed in thePosterior

Analytics, where the conditions that a syllogism must satisfy if it is toqualify as a demonstration proper to one of these sciences are tackled,is either an appendix to analytic, construed narrowly as the generallogical theory of the syllogism, or a second part of analytic, which,however, applies not to all syllogisms, but only to demonstrations.

The picture of a system of argumentative disciplines dependent

on analytic that emerges in this way seems to receive confirmationfrom the structure of the Organon, where Aristotles ancient editorsbrought together the works they took to be about logic.3 Accordingto tradition, the first two works of theOrganon, theCategoriesand theDe interpretatione, prepare the way for the categorical theory of thesyllogism tackled in the Prior Analyticsby studying terms and propo-sitions respectively. ThePosterior Analyticsand theTopicsthen applythe theory to the domains of demonstrative and dialectical argumentin turn, and theSophistical Refutationsbrings the study of argument

to a close by examining fallacious argument. At the very end of theSophistical Refutations, and therefore of the Organon itself, Aristotleobserves that, because he had no predecessors in the study of syl-logizing, it was necessary for him both to found the discipline and tobring it to the level already attained by other disciplines (34.183b34ff., 184b1 ff.). He compares the more typical case of rhetoric, whichhad reached its then present condition gradually as one rhetoricianafter another built on, and added to, the contributions of his prede-

2It is found only atRhet. 1.4.1359b10.Metaphysics.2.1005b25, which finds faultwith those who are ignorant of analytics, may furnish a parallel.

3J. Brunschwig, LOrganon: Tradition grecque, in R. Goulet, ed.,Dictionnairedes philosophes grecques(Paris: CNRS, 1989) I.485502.



5/23

R H E T O R I C A90

cessors (183b26 ff.). For a very long time, it was assumed that thesecelebrated remarks were about the categorical theory of the syllo-

gism expounded in thePrior Analyticsand were meant to emphasizethe place of central importance that analytic occupies among thedisciplines of argument.

But scholarship has gradually made it plain that Aristotle wasnot talking about analytic at the end of Sophistical Refutations. Themost important piece of evidence is the curious failure of both theTopicsand theSophistical Refutationsto take account of the categoricaltheory of the syllogism or its technical vocabulary, even though thedeclared object of the former is to expound a method of syllogizing

for use in dialectic (1.1.100a1 ff.). What is more, the syllogisms that theTopics tells us how to form by and large do not conform to the rules ofthe categorical syllogistic. Surely, the argument runs, the TopicsandtheSophistical Refutationswould have betrayed the influence of thePrior Analytics, or its theory, if the Prior Analytics, or its theory, hadbeen there to influence them.

The Topics and the Sophistical Refutations form a unity4from thispoint, when I speak of the Topics I mean to include the Sophistical Refu-tationsand on closer inspection it transpires that Aristotles proud

claim to have been the first student of the syllogism concludes a reca-pitulation of the inquiry that corresponds to the program of the Topics(SE34.183a37184b8). It is this inquiry that he meant to describe asthe first investigation of syllogizing.5 The curious omissions of theTopics and the fact that the arguments it tells us how to form aretypically not categorical syllogisms are explained by the fact that, al-though Aristotle had the idea of the syllogism when he composed theTopics, the categorical theory of the syllogism was still in the future.

This is an old story. I have rehearsed it here not because of its

intrinsic interest, but in order to set the stage for the question withwhich I shall be chiefly concerned. Suppose that Aristotles concep-tion of the relations between analytic and argumentative disciplineslike dialectic and rhetoric was not, as long assumed, constant, butthat analytic and the categorical syllogistic were relative latecomersto the scene. What consequences does this have for the picture of theargumentative disciplines that emerged above? More precisely how

4T. Waitz, Organon Graece, (Leipzig: Hahn, 184446), II.5289; L.-A. Dorion,Aristote:Les refutations sophistiques(Paris: Vrin, 1995), 245.

5C. Thurot, Etudes sur Aristote: Politique, Dialectique, Rhetorique (Paris: Durand,1860), 1957; L.-A. Dorion, Aristote et linvention de la dialectique in M. Canto-Sperber and P. Pellegrin, eds.,Le Style de la pensee: Recueil de textes en hommage aJacquesBrunschwig(Paris: Les Belles Lettres, 2002), 182220.



6/23


did Aristotle conceive of these disciplines and their relations beforeanalytic?

It will help to sharpen this question if we consider an objection.The Topics and Prior Analytics have very different aims. The Topics is amanual whose purpose is to furnish a certain practice of argument,dialectic, with a method. The Prior Analytics expounds the worldsfirst formal logic. Why then, runs the objection, should the Topicsenter into the details of a logical theory that it is the business of adifferent kind of work to study? And it concludes that the Topics couldhave been written in much the same way whether it was composedbefore or after the discovery of the categorical syllogism.6

The strongest form of this objection would undermine the casefor a development in Aristotles thinking of the kind I have just de-scribed. If it is on the right lines, the Topics silence about analyticand the categorical theory of the syllogism do not speak as loudlyas proponents of the developmental account suppose. I believe thatthis form of the objection can be answered, so that the case for therelative chronology of the TopicsandPrior Analytics and a develop-ment in Aristotles thinking about argument between them stands.Quite apart from other considerations, there is abundant evidence

that Aristotle thought the theory of the categorical syllogism didhave a contribution to make to argumentative disciplines like dialec-tic and rhetoric. After the passage about the universal applicabilityof the figures of the categorical syllogism from Prior Analytics 2.23that I cited above, Aristotle goes on to analyze forms of argumentespecially prominent in rhetoric in the light of the categorical the-ory. This analysis seems to be cited and repeated in less technicalform in the Rhetoric in a pair of passages that appear to be late in-sertions in otherwise earlier, Topics-oriented surroundings (Rhetoric

1.2.1357a2258a2; 2.25.1400b131403a6).7

What is more, in thePriorAnalytics Aristotle explains how the method of invention based on

6R. Smith, Dialectic and the Syllogism, Ancient Philosophy14 (1994): 13351 (p.140).

7So F. Solmsen,Die Entwicklung der aristotelischen Logik und Rhetorik.Neue philol-ogische Untersuchungen 4(Berlin: Weidmann, 1929), 1331; followed by M. F. Burnyeat,Enthymeme: Aristotle on the logic of persuasion in D. J. Furley, A. Nehamas, eds.,Essays on Aristotles Rhetoric (Princeton: Princeton University Press, 1994), 355 (pp.315); J. Allen,Inference from Signs: Ancient Debates about the Nature of Evidence(Oxford:Oxford University Press, 2001), 2022; opposed by J. Barnes, Proof and the Syllo-gism in Berti,Aristotle on Science, cited in n. 1 above, 1759 (p. 52) (in the context of

broader agreement with Solmsens developmental thesis); C. Rapp (trans.),Aristoteles:Rhetorik(Berlin: Akademie Verlag, 2002), II.2024.



7/23

R H E T O R I C A92

the categorical theory of the syllogism will be of use to the dialecti-cian in his quest for syllogisms (1.27.43b1130; 30.46a210), and he

sometimes pauses to give tactical advice based on the theory to thedialectician.8

This shows that Aristotle was concerned not merely to demon-strate the universal application of his theory to every domain ofargument, but that he also believed that understanding it, or parts ofit, or at a minimum some of its results will help the practitioner of ev-ery kind of argument (cf. Prior Analytics1.27.43a204; 29.45b3646a2;30.46a3 ff.; 31.46a29). Had the theory been available when Aristotlecomposed his manual of dialectical argument, then some of its results

would have been included there just as some of them were includedin his revised manual of rhetoric. The fact that they were not stands aspowerful evidence that the Topics and its view of syllogistic argumentantedate thePrior Analyticsand its theory.

But this evidence is about how the new discipline of analyticadded to the stock of dialectics techniques. The objection can beraised in a different form, which retains considerable force. In view ofthe very different aims of the two disciplines, why should Aristotlesconception of the discipline of dialectic and the place it occupies in

the system of argumentative disciplines have been different beforethe emergence of analytic? And it is chiefly this question, and notthe question of what changes in dialectical or rhetorical methodsanalytic may have required, that I shall pursue here. The view thatanalytic replaced dialectic is sometimes attributed to proponents ofthe developmental thesis that I sketched a moment ago, and thisis clearly much too strong.9 The evidence I have cited showingthat Aristotle thought that analytic had something to contribute tothe practice of dialectical argument does not show that he thought

that analytic abolished the discipline of dialectic anymore than itabolished that of rhetoric. But obviously the view that analytic leftdialectic entirely unaffected and the view that it replaced dialectic donot exhaust the field.

It is hardly surprising that generations of readers took Aristo-tles claim at the end of the Sophistical Refutations to have been the

8On this point, see H. Maier, Die Syllogistik des Aristoteles, (Tubingen: Laupp18961900), II.b.78 n. 3. An especially striking example is furnished by the counsel to

be on the lookout for recurring terms in an opponents argument because the middleterm on which his syllogism will depend must occur twice (2.19.66a25).

9Solmsen is sometimes said to have held this view, but though an occasionalincautious formulation of his may suggest this (Entwicklung, 26, 195), he did not holdit in anything like this unqualified form.



8/23


first to study syllogizing as a reference to analytic and the Prior An-alytics, for, as we have seen, in his mature view the study of the

syllogism in general is the function of analytic. This is clearly im-plied at Prior Analytics 1.4, where Aristotle characterizes the sub-ject of the inquiry underway as the syllogism in general by contrastwith a species of syllogism, demonstration, that is to be tackled later(25b2631). And in a couple of passages in the Posterior Analytics,Aristotle refers back to thePrior Analyticsas the work about the syl-logism (1.3.73a14, 1.11.77a33). So it is puzzling to find a manual ofdialectic, the Topics, presented as the fruit of an inquiry into syllo-gizing, rather than a kind of syllogizing, viz. the dialectical kind. To

add to our puzzlement, theSophistical Refutationselsewhere uses theterm syllogistic to characterize dialectical arguments in a contrastwith rhetorical arguments (5.167b21 ff.).10 And the Rhetoric constantlytreats the syllogism as the special concern of the discipline of dialec-tic by contrast with the enthymeme, the form of argument properto rhetoric.11

My thesis is that, in a way, analytic did replace dialectic. It re-placed dialectic, however, not by abolishing it or, necessarily, byrendering all of its counsels obsolete, but by supplanting it as the

master discipline of argument whose responsibility it is to treat ofthe syllogism in general. As a result, dialectic was demoted to therank of a subordinate discipline of argument, oriented toward onepractice of argument and dependent on analytic for what it needsto know about the fundamentals of argument. But analytic was adiscipline of a new kind, not a method oriented to one practice ofargument or a special science that makes use of argument, but alogical theory. If you will, in the course of becoming the master disci-pline of argument, it changed what it was to be a master discipline

of argument.It is relatively easy to see how analytic was able to be the masterdiscipline of argument. As the general theory of syllogistic validity,it is the home of principles to which every valid syllogism mustconform.12 To be sure, thePrior Analyticscontains a good deal morethan a logical theory, and to the extent that it is intended as an exposi-

10Cf. Rapp,Aristoteles: Rhetorik, cited in n. 7 above, II.97 ad Rhet.1.1.1355a30.11Rhet. 1.1.1355a30, b1517; 2.1356a36-b2, b1013, 1358a26, 15; 2.22.1395b224.12This claim has to be qualified, and is qualified by Aristotle in Prior Analytics

1.44, where he allows that so-called syllogisms on the basis of a hypothesis cannotbe analyzed as categorical syllogisms. It is nonetheless surprising just how minor aqualification Aristotle takes this to be. Cf. G. Striker, Aristoteles uber Syllogismenaufgrund einer Hypothese,Hermes107 (1979): 3350.



9/23

R H E T O R I C A94

tion of the discipline, so does analytic.13 But most of the additionalmatter consists of applications of the theory to issues that arise in

connection with argument and different forms of argument. As wehave already observed, Aristotle applies the categorical theory toforms of argument that are prominent in rhetoric. He also uses theformal machinery of the theory to tackle fallacies (2.1618), to provethat it is possible to syllogize true conclusions from false premises(2.24), and to construct an extremely simple method of inventionthat can be applied mechanically to any desired categorical conclu-sion to yield syllogisms for it (1.278). It remains to be seen how thediscipline of dialectic, which unlike analytic is a method oriented

towards one practice of argument, could also have been the masterdiscipline of argument.For help we can turn to the relations between the disciplines

of dialectic and rhetoric as Aristotle describes them in the Rhetoric.The work begins with the famous declaration that rhetoric is thecounterpart of dialectic (1.1.1354a1). Two disciplines are counter-parts when they stand in the same relation to different objects. In theGorgiasPlato had set up a system of correspondences between artsthat care for the body and those that care for the soul (463a-466a).

Justice, for instance, is the counterpart of medicine because it standsin the samecorrectiverelation to the soul that medicine standsin to the body. Rhetoric is the counterpart of cookery (),or so Socrates maintains, because it is a false likeness of justice inthe way cookery is of medicine, aiming at pleasure rather than thegood of its subject and relying on experience and conjecture ratherthan knowledge. Aristotle defends rhetoric by making it an art ofargument like dialectic. The features that it shares with dialectic, andwhich they both owe to their status as arts of argument, are therefore

no more to be held against rhetoric than they are against dialectic. Atthe same time, he rebukes contemporary rhetoricians who neglectedargument in favor of appeals to the emotions and discussion of theparts of an oration (Rhetoric1.1.1354a1116, b1622).

As arts of argument, rhetoric and dialectic have no special subjectmatters of their own of the kind that distinguish the ordinary runof arts and sciences, but can in principle be applied to any andevery subject (Rhetoric 1.1. 1354a13, 1355b79; 2.1356a314, 1358a2125; 4.1359b1116). The objects to which they are related are instead

different practices of argument, and the relation in which they stand

13G. Striker, Aristotle and the Uses of Logic, in J. Gentzler, ed., Method inAncient Philosophy (Oxford: Oxford University Press, 1998), 20926.



10/23


to them is that of supplying them with an art or method.14 In thepassage that we are considering, these practices are described as

activities common to a certain extent to all human beings, namelyexamining and upholding an argument in the case of dialectic, andaccusing and defending in that of rhetoric. Later in the Rhetoric,Aristotle distinguishes other varieties of rhetoric with reference totheir characteristic activities and the concerns of the audiences towhich they are addressed. To accusation and defense, which belongto forensic oratory, he adds advocating or opposing a course of action,which belong to deliberative oratory, and praise or blame, whichbelong to epideictic oratory (1.3.1358a36-b13).

The Topics likewise distinguishes varieties of discussion withthe difference that Aristotle seems to single out one of these asdialectic in the strictest sense (Topics 8.5.159a2536 cf. 11.161a25;Sophistic Refutations 2.165a38 ff.). This is discussion for the sake of(exercise or training) and(putting to the test), whichhealsocharacterizesasforthesakeof and (investigation)(159a33). These he contrasts with discussions between teachers andlearners, where the aim is to impart knowledge, and competitiveencounters, where there is no common task and each party strives

for victory by any available means. Common to all of them arethe presence of two parties, a questioner and an answerer, andthe requirement that the argument advance by stages in which thequestioner secures the assent of the answerer to premises put forwardin the form of questions.

Aristotle does not explain what is put to the test in dialecticalargument. It is tempting to connect what he says here with ,which is introduced in theSophistical Refutationsas either a branch ofdialectic or a closely related sister discipline, and whose function is

to put an interlocutors claim to knowledge to the test by arguing inthe Socratic manner from histhe interlocutorsadmitted opinions(Sophistical Refutations 2, 164b4; 8, 169b25; 11, 171b4; 172a217; 34,183b1).15 But it does not appear that the dialectical answerer Aristotleenvisages in the Topics lays claim to knowledge. The defense oftheses to which he is personally attached is only one possibilityamong others (8.5.159b1, 257). He will also, and perhaps more

14J. Brunschwig, Rhetorique et Dialectique,RhetoriqueetTopiques in Furley andNehamas,Aristotles Rhetoriccited in n. 7 above, pp. 5796 (p. 59); C. Rapp,Aristoteles:Rhetorik,cited in n. 7 above, II.201.

15P. Moraux, La joute dialectique dapres le huitieme livre des Topiques inG. E. L. Owen, ed.,Aristotle on Dialectic: The Topics (Oxford: Oxford University Press,1968), 271311 (2889).



11/23

R H E T O R I C A96

typically, uphold theses without regard for his own opinions; inthis case, his task is to admit only premises which are reputable

or more reputable than the thesis (8.5.159a38-b25).16

In one form ofdialectical discussion, the answerer undertakes to defend a thesisof a famous philosopher, and his task then becomes to admit onlypremises that would be acceptable to that philosopher (8.5.159b2735; cf. 1.11.10461928). Though the answerers ability to upholdtheses will undoubtedly be put to the test in dialectical discussionsof these kinds, if he performs his task correctly, it will be chiefly thethesis itself that is tested and made the object of investigation (cf.8.4.159a204). And this seems to be what Aristotle has in mind in the

Metaphysicswhen he observes that dialectic ispeirastic, that is, that itprobes or examines or puts to the test, where philosophy is gnoristic,i.e., has knowledge ( 2.1004b226). I shall tackle the question ofwhat Aristotle means by exercise or training shortly.

The aimof the discipline of dialectic, and therefore Aristotles aimin theTopics, is to equip these practices of argument, and especiallydialectical argument in the strictest sense, with a method. The methodhad to unite those elements that are necessary to this practice, whichserves, therefore, as the methods organizing principle.

In treating rhetoric and dialectic as counterparts, each with itsown sphere of operation, however, Aristotle is not, as it might at firstappear, treating them as completely autonomous and coequal disci-plines. It is plain from the Rhetoric that the discipline of rhetoric relieson dialectic for its understanding of the fundamentals of argument. Inthe first, introductory chapter of theRhetoric,Aristotle explains thatthe enthymeme is a syllogism of a kind (1355a810),17 and that theorator who combines a grasp of the syllogism with an understandingof the effects on argument of rhetorical subject matters will be best

equipped to argue in rhetorical contexts. But for his understandingof how and from what a syllogism arises, Aristotle insists, the or-ator is reliant on dialectic since the consideration of every syllogismwithout distinction () is the business of dialectic, either dialec-tic as a whole or a part of it (1355a810).18 The specifically rhetorical

16M. Wlodarczyk, Aristotelian Dialectic and the Discovery of Truth, OxfordStudies in Ancient Philosophy18 (2000): 153210.

17Translated a syllogism of a kind in accordance with Burnyeats recom-mendation in order to leave open the possibility that an argument may qualify asan enthymeme without meeting the standards strictly necessary to be a syllogism(Burnyeat, Enthymeme, cited in n. 7 above, pp. 1315)..

18Without distinction is the Oxford translator, Rhys Roberts, rendering. Withthe use ofhere compareDe gen. et corr. 1.1, 314a2, where Aristotle proposes to



12/23


knowledge that the orator must add to the grasp of the syllogismthat he takes from dialectic is an understanding of the sort of matters

with which enthymemes are concerned and how enthymemes differfromlogikoi syllogismoi(a1014).The termlogikosis an important clue about the nature of dialecti-

cal argument proper. To tackle an issuelogically, in this sense, is todo so on the basis oflogoi, arguments or reasonings, viewed in oneway or another, and to one degree or another, in abstraction fromthe special features of the subject matter at issue. These features areaccessible only to those with a specialists substantive understandingof the subject matter, and in the treatises proceedinglogicallyis typi-

cally opposed to doing so on the basis of such an understanding.19

Dialectic proceeds logically because when it tackles a question, it ison the basis of abilities available to the master of an art of argumentwithout substantive specialized understanding. To the extent thatone draws on such an understanding, one leaves dialectic behind(Rhetoric1.2.1358a236; 4.1359b1216).

The orators understanding of the enthymeme is in part the resultof adding to the grasp of the fundamentals of argument that he takesfrom dialectic knowledge of the kinds of matters with which rhetoric

deals, but also of the kinds of occasions and audiences to whichrhetorical argument is suited (Rhetoric 1.2.1357b26; 2.20.1394a247).One effect is a relaxation in the rigor or stringency permitted in en-thymemes by comparison with syllogisms proper. This is becausethe matters about which one argues in rhetoric, like those whichare the concern of practical reason, rarely lend themselves to reso-lution by conclusive arguments (Rhetoric1. 2.1357a17, 1315, 227;2.25.1402b324). Typically it is possible only to present considera-tions of a certain weight, and orators arguments, though able to

make a conclusion a reasonable thing to believe, cannot exclude thepossibility that it is false. Often the best decision that jurors attend-ing conscientiously to the arguments on both sides of the case canreach is one that new evidence may show to be mistaken. And as-semblies that arrive at the decision that does the most justice to the

treat generation and corruption , i.e., without entering into thespecial features or peculiarities belonging to the generation and corruption of thedifferent kinds of natural substance (cf. 2.9.335a258).

19On the term see Waitz, Organon Graece, cited in n. 4 above, II.353 ff.(adAn.post. 82b35); A. Schwegler, ed.,Die Metaphysik des Aristoteles(Tubingen: Fues,1847), IV.4851 (ad Metaph. Z 4.1029b13); H. Maier, Syllogistik, cited in n. 8 above,II.a.11 n. 3; M. F. Burnyeat,A Map of Metaphysics Z(Pittsburgh: Mathesis Publications,2001), 1922 et passim.



13/23


14/23


these, , practice, in the sense of exercise or training, to be thepractice, in the sense of activity or pursuit, of which the discipline

of dialectic elaborated in the Topics is the art.21

If this is right, thenthe second and third uses of the method, though to one degree oranother dialectical, will not, strictly speaking, be forms or varieties ofdialectic.

But if the practice for which the discipline of dialectic set out intheTopicsis practice in the sense of training, what is it training for?Aristotles idea seems to have been that dialectical argument was aform of intellectual practice or training, which gives rise to intellec-tual and argumentative facility in the way that a program of physical

exercise performed under the direction of a gymnastic instructorproduces physical strength and coordination (cf. Topics 1.11.105a9;8.5.159a25; 8.11.161a25). Aristotle seems also to have believed that di-alectic stood in an especially close relation to philosophy (Metaphysics4.2.1004b226;Topics 1.2.101a34-b4; 8.1.155b716, 14.163b916). It isnot merely that the sample dialectical problems that are mentioned intheTopicstend to be philosophical.22 The method it expounds seemsto have in view philosophical discussions of the kind that we are ledto believe took place in the Academy and which are mocked in the

famous fragment of Epicrates in which Platos followers are picturedsolemnly defining and classifying by species vegetables, not except-ing the pumpkin. It is concerned not only with the construction andevaluation of arguments that a predicate simply belongs to a subject(i.e., belongs to it as an accident), but gives equal attention to those inwhich the issue is whether it belongs as a genus, proprium or def-inition, and elaborates a separate method of inventing syllogisms foreach of these, the four so-called predicables (1.6.102b27 ff.; 4.1.120b12ff.; 6.1.139a24-b5; 7.5.155a1636).23

As we shall see, Aristotles views about the relations betweendialectic and philosophy are surprisingly complex, but if this conclu-sion is on the right lines, he regarded the practice of dialectic, or onefavored form of it, as preparation or training or practice primarilyfor philosophy. Aristotles talk of would then be another

21E. Kapp,Greek Foundations of Traditional Logic(New York: Columbia UniversityPress, 1942), 1213; P. Moraux, La joute dialectique, cited in n. 15 above, pp. 289

90; J. Brunschwig, Aristotle on arguments without Winners and Losers, JahrbuchWissenschaftskolleg zu Berlin1984/5, 3140 (p. 34); O. Primavesi,Die aristotelische Topik:Ein Interpretationsmodell und seine Erprobung am Beispiel von Topik B (Munich: Beck,1996), 2021, 49 ff., Rapp,Aristoteles: Rhetorik, cited in n. 7 above, I.2512.

22I. Duhring, Aristotles use of examples in the Topics, in G.E. L. Owen,Aristotleon Dialectic, cited in n. 15 above, pp. 20229.

23J. Brunschwig, Aristotle on arguments, cited in n. 21 above, p. 32.



15/23

R H E T O R I C A100

instance of a widespread tendency to compare the training or prepa-ration of the soul to that of the body. We may compare Isocrates, who

in his use of the idea of counterpart disciplines makes philosophy, inhis sense of the term, the counterpart of, of which a partis (Antidosis 18082). And though he was notoriouslydismissive of the claims made by figures like Plato and Aristotleon behalf of what they call philosophy, Isocrates is also willing toconcede some value to it as training () and preparation forphilosophy properly so called (Antidosis2669).

There are a handful of clues in Plato which suggest that therewas a practice of argument in the Academy that was conceived as a

form of practice or preparation and was called . Thus in thefirst part of the Platonic dialogue that bears his name, Parmenidesraises difficulties for the theory of the forms, which is treated thereas the view of the young Socrates. Socrates has gone astray, Par-menides maintains, by attempting to define forms of beauty, jus-tice, goodness and the like before he had practiced or exercisedproperly. The practice he prescribes resembles Zenos manner ofargument, which had been discussed earlier in the dialogue, andhe insists that it is an essential preparation for grasping the truth

(136ce).24

Of course, it is another question why the practice of dialecticarose and took the form that it did. A fuller attempt to answerthis question would require considering the Academic milieu andthe wider philosophical world of the fifth and fourth centuries.The inquiry could be extended to the broader social and politicalconditions in the Greek speaking world that favored the growth ofpublic practices of argumentation and so encouraged reflection aboutproof, validity, refutation and the like. But to the extent that this kind

of inquiry appeals to contingent historical factors, it tends to drawus away from the kind of answer Aristotle himself would most likelyhave given.

I take it that he would have explained the practice of dialecticalargument as an expression of human nature. I have already cited theRhetorics claim that all human beings share in a way in dialectic to theextent that they attempt to examine and uphold theses (1.1.1354a36). The same sentiment can be found in the Topics, where all humanbeings are said to participate without art in that with which dialectic

is concerned as an art (Sophistic Refutations11.172a34). Remarks like

24On this passage from theParmenidessee Maier,Syllogistik, cited in n. 8 above, IIb 51, n. 1.



16/23


these suggest that Aristotle regarded the more specialized form ofphilosophical argument toward which the dialectical method set out

in the Topics is oriented as the form par excellence of the universalpractice of examining and upholding theses in argument by questionand answer. We may compare Isocrates attitude toward the kindof speech or logos that he teaches his pupils how to compose; heregards it as the expression par excellence of the human capacity forlogos, which encompasses all that is most distinctive of human beings(Nicocles610).

These considerations go some way towards explaining howAristotle could have regarded dialectic as the paradigm practice

of argument and the corresponding discipline as the primary dis-cipline of argument. It will also help to attend to a feature of dialec-tic that it owes specifically to its gymnastic character and whichcomplements its nature as a pure practice of argument, that is,a practice participation in which calls only for a logical under-standing, in Aristotles sense, of how to construct and evaluatearguments and not substantive knowledge of the matters underdiscussion.

It is clear that the answerers task is, in a way, to make the ques-

tioners work more difficult. He is to do this not by using any meansto obstruct the questioners progress, but rather by holding the ques-tioners argument to the highest standards, so that when the an-swerers thesis is defeated it will be because of its weakness notthe answerers (Topics 8.4.159a204). This is what is behind Aris-totles talk of a common task shared by questioner and answerer(8.11.161a20, 37).25 In Topics 8.5 the focus is on which premises theanswerer ought to allow. As we have seen, Aristotle holds that heought only to give his approval to those which are reputable or more

reputable than the conclusion for which the questioner is strivingto construct an argument. But throughout theTopicsAristotle makesnote of objections that the answerer can raise. Indeed he says thedialectician is not only a proposer of premises but also a raiser ofobjections (8.14.164b3). And this shows that the answerer is alsocharged to ensure that the questioners argument proceeds validly.The practice of gymnastic dialectic, then, puts a premium on care-ful step-by-step argument in which each step must be made explicitand pass muster with the answerer who is on the lookout for any

misstep This focus on the argument as such, with its attendant em-phasis on rigor and explicitness, furnishes another reason why the

25Brunschwig, Aristotle on arguments, cited in n. 21 above.



17/23


fundamentals of argument are a special concern of the dialecticaldiscipline.

By contrast rhetoric does not share dialectics interest in rigorand explicitness for their own sake. As Aristotle several times notes,rhetorical syllogisms or enthymemes need not state all the premiseson which the conclusion depends.26 He emphasizes that the kind oflisteners whom it is rhetorics task to address are unable to followlong arguments, but also that a full statement of the argument ishardly necessary if the listeners can take the point without it (Rhetoric1.2.1357a45, 1622; 2.22.1395b246). The orator aims to persuade andargument is his principal instrument. The point is not that this means

that anything goes in rhetoric. A noble orator will want to convincehis audience of the truth, or the conclusion that stands the best chanceof being true, and have them accept it for sound reasons, but even sohe lacks the dialecticians interest in the mechanics of valid argumentas such. He is able to adopt this attitude towards argument, however,because rhetoric can turn to dialectic for its understanding of thefundamentals of argument, which it then modifies and supplementsto suit its own ends.

So far I have argued that in Aristotles early thinking about argu-

ment the discipline of dialectic played a double role. It was the art ordiscipline of a practice of argument, so-called , which servedas its organizing principle, and it was at the same time the masterdiscipline of argument, to which disciplines corresponding to otherpractices of argument had to turn for an understanding of the funda-mentals of argument. I suggested that the discipline of dialectic wasable to perform both these functions for Aristotle because in his view,though it corresponds to one practice of argument among others, thatpractice is the paradigm form of argument, argument par excellence.

(I have left entirely out of account a separate question, which is nev-ertheless related to those pursued here, namely how early Aristotlerecognized that demonstrative syllogisms would require a specialinquiry of the kind pursued in thePosterior Analytics.)27

The perspective that we have now achieved should throw lighton the other uses for the method of dialectic that Aristotle puts beside. A closer examination of these uses will, I believe, lendfurther support to the thesis that I have been advocating. The seconduse of the method is

26Tradition turned this accidental attribute of the enthymeme into its essence.Cf. Burnyeat, Enthymeme, cited in n. 7 above, pp. 214, 3950.

27On this issue, see J. Barnes, Proof and the Syllogism, cited in n. 7 above.



18/23


for encounters because, having reckoned up the opinions held by themany, we shall speak to them not from the opinions of others, but from

their own, changing their minds about anything they seem to us notto have stated wellTopics1.2.101a3034.28

The third is

for the philosophical sciences, because, if we have the ability to gothrough the difficulties on either side, we shall more readily discern thetrue and the false in each matter

Topics101a346

There are two ways in which the method of dialectic may beuseful to these ends, however. Is it useful because (a) the practiceofgymnastic argument, of which it is in the first instance the method, isin turn useful in relation to other ends, or (b) the method itself, or partsof it, are directly useful to practices of argument other than dialecticand without being put to use in the practice of dialectic? The positionthat I have been defending would lead us to expect the answer tobe in way (b) or a combination of ways (a) and (b). And certainly

in the case of the second use of the method, in encounters, (b) is veryplausible. The command of a mass of reputable opinions, organizedwith reference to the type of person to whom they are likely to bepersuasive, will be of the greatest assistance in conversations of a notspecifically dialectical character, as Aristotle also notes in the Rhetoric(1.1.1355a28; 2.1356b324; cf. 2.22.1396b211).

The philosophical use of the method is more complicated andhas been the object of much attention. Aristotle expands on thedescription of the philosophical use of the dialectical method in the

immediate sequel to the remarks cited above.29

And further the treatise or the method it contains [the pragmateia] isof use regarding the first of the matters that concern each science. Forit is impossible on the basis of the principles of the science at issue tosay anything about them, since the principles are prior to everythingelse, but it is necessary instead to tackle them via the reputable opinionsregarding each of them. And this is peculiar or most proper to dialectic

28These translations and that of 101a36b4 below are adapted from R. Smith(trans.),Aristotles Topics: Books I and VIII(Oxford: Oxford University Press, 1997).

29Unless they are meant to introduce a fourth use in addition to the threepromised at 101a26. On this question, see J. Brunschwig, ed. and trans., Aristote:Topiques, Livres IIV(Paris: Les Belles Lettres, 1967), xii, 116 n. 1.



19/23


since, because of its capacity to examine, it has a way to the principles ofall the disciplines.

Topics101a36b4

Aristotle seems to mean that inquiry about principles belongs to themethod of dialectic rather than to other methods, or in any case toit more than to them, and not that it is the most proper use of dialecticby comparison with the other uses to which it may be put.30

It is noteworthy that on the rare occasions when he touches onthe issue at all in theTopics, Aristotle seems to view the philosopheras a solitary inquirer whose characteristic activity is distinguished

from dialectic in part by the fact that it does not involve relations toanother (Topics 8.1.155b1016; Sophistic Refutations 16.175a912). Thusin Topics 8.1, Aristotle notes that what has gone before, viz. the discus-sion of thetopoi, may be of use to the philosopher engaged in solitaryinquiry as well as the dialectician, whereas the discussion of how toorder and pose questions that is to follow will not be relevant to hisconcerns since it applies only to arguments conducted with others(155b716). This implies that philosophers can derive benefits fromthe dialectical method, or parts of it, without engaging in the practice

of dialectic and while remaining ignorant of some of its precepts.This seems to be confirmed in what is by far Aristotles mostilluminating remark about dialectics usefulness to philosophy:

Withregard to knowledge and philosophical wisdom being able to graspand to have grasped what follows from each hypothesis is an instrument() of no little value. For it remains to choose one of them correctly,and for this one must have a good nature; this is the good nature thatregards the truth, the ability to choose the true and flee the false.

Topics163b916

The context for this remark is Topics8.14, which is about dialecticaltraining (). The training in question is not that in which thepractice of dialectic consists, however, but training or preparation forthis practice. Such training can be performed alone or in company(163b34), and it seems that the philosopher may profit from itwithout engaging in the practice of dialectic by going through thearguments that one might employ in an actual dialectical encounter.Of course, there is every reason to suppose that philosophers will

also profit from such participation. The abilities and understandingdeveloped by preparatory exercises will be further sharpened by the

30Brunschwig,Topiques, 117 n. 2.



20/23


actual practice of dialectic, and there may be others which can only beacquired in this way. It is clear, then, that Aristotle takes the method

of dialectic to be of use to philosophy in both ways, directly withoutbeing applied in the practice of dialectic and through being used inthat practice, though it is the first that he emphasizes in theTopics.

The picture that emerges differs in some ways from the expecta-tions we might take away from, e.g., Nicomachean Ethics 1.1. ThereAristotle tells us that every art and every method has its own end.But although it is clear that an arts end does not always tell us whyanyone might care to practice itfor that we must look to the broadercontext and perhaps ultimately to practical wisdom or political sci-

ence and the human good which is their objectit is implied thatwhen an art or method is of use relative to goods other than itsown, proper end, it is through achieving that end. As we have seen,however, more often than not the uses of the dialectical method thatAristotle has in mind are available to those who do not participatein the practice of dialectic.

Although it answers our question about the dialectical methodsusefulness to philosophy at a certain level of generality, this leavesunclear precisely how, as a form of , the actual practice

of dialectic is supposed to contribute to philosophy. One kind ofinterpretation sees a sharp distinction between dialectical activityand philosophy, i.e., between what one does as a dialectician andwhat one does as a philosopher. On this view, the practice of dialecticcontributes skill or virtuosity in argument to philosophy rather thana deeper understanding of philosophical issues.31 At the oppositeextreme are interpretations which suppose that in practicing dialecticone is, at least much of the time, practicing philosophy and thata good part of the philosophers time, as a philosopher, will be

occupied with the practice of dialectic.A distinction of the kind posited by the first kind of interpretationis easiest to see when the skill and argumentative facility beingdeveloped by dialectic are exercised in relation to objects differentfrom those tackled by philosophy. We may compare the way inwhich the participants in a Platonic dialogue will sometimes warmup, as it were, by defining an item or working through a divisionof little intrinsic interest before tackling a more serious question (e.g.,Politicus, 285d-286a). In fact, the issues regularly cited as examples in

the central books of theTopicsare serious and not the trifling mattersthat one would expect if the sole object in discussing them were to

31Moraux, La joute dialectique, cited in n. 15 above, p. 308.



21/23


cultivate facility in argument.32 But interpretations of the second kindare hard to square either with the Topics conception of dialectic as

a form of training or the way in which Aristotle sometimes contrastsdialectic and philosophy in the treatises.It may help to bear in mind that some forms of practice or

exercise are manifestly different from the activities for which theyprepare or train those engaged in them, while others differ chieflyby being performed as exercises or for the sake of practice andthat there are many degrees in between. In Topics 1.11 Aristotleconsiders what is to count as a dialectical problem. The examples hecites, e.g., whether the universe is eternal or not (104b16), show that

serious philosophical questions are not excluded. He does, however,insist that those engaged in dialectic will not tackle questions thedemonstration of which is near to hand or those of which it is toodistant; the former because they do not present a challenge, the latterbecause the challenge they present is more than accords with thepurposes of gymnastic (more than is ) (105a79).Though some questions of interest to philosophy may be excluded,discussion of many others is plainly permitted. I suspect that weshall come closest to grasping the distinction between dialectic and

philosophy if we suppose that the qualification so far as accordswith the purposes of gymnastic that Aristotle uses to restrict thescope of dialectical discussion can also be used to characterize theway dialectic tackles questions that it shares with philosophy. Anargument or discussion will be dialectical to the extent that it accordswith the purposes of gymnastic, and this will depend in part onwhether it confines itself to the resources available to the dialecticianbut also how far and in what spirit the discussion is pursued.

Nothing prevents the practice of dialectic from deepening the

participants understanding of the issues at the same time as itsharpens their argumentative skills. Putting philosophical theses tothe test () and subjecting them to investigation () shouldserve both these ends. But when the understanding amounts toknowledge, dialectic is left behind. That the transition from thedialectical discussion of an issue to one grounded in knowledge andunderstanding of the field to which it belongs will sometimes be hard

32Moraux, La joute dialectique, acknowledges this, but takes it as furtherevidence for the old view that the peripheral books, 1 and 8, are more recent than thecentral books. On his view, book 8 marks the emergence of a new perspective, whichno longer identified philosophical research with dialectical discussion, but insteadviewed the latter as a form of intellectual training sharply distinct from the former.



22/23


to mark is a point Aristotle makes himself, when he observes that itis possible to pass by degrees without noticing it from discussing

a matter in the way proper to dialectic to discussing it in the waythat belongs to the science under which the matter falls (Rhetoric1.2.1358a236).

The last question, about the second and third uses of the dialecti-cal method, has taken us some way from our main purpose, whichwas to discover how Aristotle could have regarded dialectic as themaster discipline of argument, but it has helped advance that purposeby throwing further light on the complex unity that the Aristoteliandiscipline of dialectic possessed. We are now in a position to see how

dialectic has been able to present such a variety of aspects to differentobservers. Viewed from one angle, it is a method tailored to the spe-cial demands of a certain form of philosophical discussion. Viewedfrom another, it is the proper home of general reflections about ar-gument of interest to every pursuit of which argument is a part.

This complex unity also helps explain other features of the dialec-tical method as it is set out in theTopics. As we have seen, Aristotleholds that the consideration of every syllogism without distinction() is the business of dialectic, either dialectic as a whole or a

part of it (Rhetoric 1.1.1355a810). Presumably he raises the possi-bility that general discussion of the syllogism belongs to a part ofdialectic rather than the whole because, as I have already suggested,the whole of it will inevitably contain much that is of use only to prac-titioners of specifically dialectical forms of argument. Yet though theTopicscontains much that has a bearing on forms of argument otherthan the dialectical, Aristotle makes no effort to gather this materialin one place corresponding to the part of dialectic that he envisagesor to tackle it in a way that makes it readily available to other dis-

ciplines. The dialectical method unites those elements necessary toequip practitioners for successful participation in gymnastic argu-ment, and it tackles facts about argument of wider interest when andin ways that are dictated by this purpose.

In addition, dialectics standing as an art of argument means thatneither its boundaries nor even those of the part of it concerned withthe syllogism in general will coincide with those of analytic. TheTopicsdoes, to be sure, contain much informal discussion of mattersthat are tackled with the aid of formal logical theory in the Prior

Analytics. It has something to say about the premises and conclu-sions of syllogisms, though in a way that is peculiarly adapted to thedialectical arguments where the differences between the predicablesmatter as they do not in rhetoric (Topics1.4, 1011). But it cannot betaken for granted that the understanding of from what and how



23/23


a syllogism arises, that the orator derives from dialecticlanguagethat is remarkably similar to that of the Prior Analyticsis confined to

the form or structure of syllogistic premises and conclusions (Rhetoric1.1.1355a1112;Prior Analytics1.4.25b267). It may embrace the rep-utable opinions which will be of so much service in encounters withthe many (Topics1.2.101a3034;Rhetoric1.1.1355a279, 2.1356b324,2.22.1396b411). And it very likely extends to the topoi,theelementsofthe method of invention on which, as Aristotle emphasizes, both di-alectic and rhetoric rely (Rhetoric1.2.1358a1032) and to which muchthe largest part of theTopicsis dedicated. Here too, however, rhetoriccannot simply take from the method of dialectic what it needs. The

Topicspresentation of thetopoidivides into four parts, one for eachof the four predicables. The difference between the different ways inwhich a predicate belongs to a subject is immaterial in rhetoric, andthough many of the topoi in the Rhetoric resemble dialectical topoi,all reference to the predicables disappears. Othertopoiin theRhetorichave no analogues in theTopics (Rhetoric2.23).

What is more, to judge by theTopicsandSophistical Refutations, aknowledge of the syllogism in general will require a thorough under-standing of how to construct fallacious arguments. To the surprise of

present-day students of Aristotle, the fullest discussion of the syllo-gism is not found in the Topics proper, but in Sophistical Refutations6. There Aristotle analyzes each of the seven forms of fallacy not dueto language as instances ofignoratio elenchiby showing how each ofthem violates a part of the definition of the syllogism (168a23; cf. 8,169b40). Some of the apparent oddness of this procedure disappears,however, if we keep in mind that Aristotle has the needs of the partic-ipants in a practice of argument constantly in view. One must knowhow fallacious arguments arise if one is to detect and solve them,

that is, to reveal why an apparently valid syllogism is in fact invalid(Sophistical Refutations 24, 179b234). This ability will come into itsown most obviously when one is faced with deliberately fallaciousarguments put to one by others. Yet here too a double perspective isin evidence; Aristotle maintains that it will also help the philosopheravoid inadvertent errors in his own reasoning (Sophistical Refutations16, 175a912).

If the argument of this paper is on the right lines, then admirationfor Aristotles invention of logic should not blind us to the existence

of an earlier phase in his thinking about argument. In it the place ofthe master discipline of argument later to be occupied by analytic wasoccupied instead by dialectic, a discipline that owed its priority not,as analytic was to do, to a concern with the form of valid argument ingeneral, but to its special relation to a favored practice of argument.

Aristotle on the Disciplines of Argument Rhetoric, Dialectic, Analytic - J. Allen

Documents