Dialectical Logic

The Dialectical Logic of Thucydides' Melian DialogueAuthor(s): Hayward R. Alker, Jr.Reviewed work(s):Source: The American Political Science Review, Vol. 82, No. 3 (Sep., 1988), pp. 805-820Published by: American Political Science AssociationStable URL: http://www.jstor.org/stable/1962492 .Accessed: 14/03/2012 08:31

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 ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

American Political Science Association is collaborating with JSTOR to digitize, preserve and extend access toThe American Political Science Review.

http://www.jstor.org

http://www.jstor.org/action/showPublisher?publisherCode=apsa

http://www.jstor.org/stable/1962492?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp

THE DIALECTICAL LOGIC OF THUCYDIDES'

MELIAN DIALOGUE HAYWARD R. ALKER, JR.

Massachusetts Institute of Technology

If the realist tradition has underappreciated the formalizable quality of Thucydides' scientific investigations, neorealist teachers and writers have generally failed to see the normative and dramatical features of Thucydides' political science, each an expression of his dialectical epistemology and ontology. Nicholas Rescher's partial formalization of dialectics as a controversy-oriented approach to knowledge cumulation and Kenneth Burke's dramaturgical approach to textual understanding are both shown to fit Thucydides' argumentation in the Melian dialogue. Thus argumentation produces new knowledge about the inner determinants of Athenian imperialism; simultaneously it dramatically reveals the constituting practical rationale of Athenian actions to be unjust. Once Thucydides' determining essences of power politics are properly uncovered, their false "eternal, mathematical necessity" can be appropriately criticized. A case is thus suggested for a "neoclassical polimetrics" more fundamentally grounded in "political argumentation" about practical choices in particular contexts than in ahistorical laws, inductive statistics or deductive mathematics.

In writing his classic study of the Peloponnesian War, Thucydides sought "an exact knowledge of the past as an aid to the interpretation of the future, which in the course of human things must resemble if it does not reflect it" (1.22).1 For twentieth-century realists, this has typically meant seeking timeless truths, or even mathematical laws, about the state's eternal, self- interested search for power, or the need to balance against it. Werner Jaeger's ref- erence to "this political necessity, the mere mathematics of power politics" con- veyed such a message to the German- speaking world of the 1930s. Indeed, in a book widely renowned for its penetrating account of Greek civic culture, Jaeger asserts that as evident in Sparta's fearful response to the growth of Athenian power, this "necessity" "is defined [by Thucydides] as the true cause ... of the

[Peloponnesian] war" (1976, 488; for similar views of other modem realists see Wight 1978, 24 and Morgenthau 1978, 8-9, 38).

Would not any other social scientist in our own century, if desirous of emulating the increasingly universal validity of the natural sciences and committed to build- ing up reliable knowledge grounded in the deductive certainty of axiomatic formal argument and the "positive" evidence of our senses, also want to claim as their own Thucydides' search for "the clearest data," and his standard of objective precision: "the accuracy of the report being always tried by the most severe and detailed tests possible"? Would they not also want to "have written [such a] work, not as an essay which is to win the applause of the moment, but as a possession for all time" (1.23)? Influenced by apparently similar contemporary epis-

AMERICAN POLITICAL SCIENCE REVIEW VOL. 82 NO. 3 SEPTEMBER 1988

American Political Science Review Vol. 82

temological trends (developmentally described as "positivism," "logical positivism," "logical or analytical empiricism," and "critical rationalism" in Alker 1982), leading expositors of "scientific" neorealism focus their frequent, typically positive citations to Thucydides' classic on such instrumental, lawlike, causal analyses of state actions, possibly extended to include the structural -determinants of these actions (Keohane 1983, 1986; Waltz 1979). Despite their objec- tions to the imprecision, logical inconsis- tency, and empirical nontestability of too much traditional realism, contemporary neorealists still seek timeless laws of power politics, independent of moral praise or blame; they wish these laws to be objectively and falsifiably delineated with what Jaeger took to be mathematical precision.

Thucydides' work is equally well known today for the fatalistic, moral cynicism of the Athenian statement in the Melian dialogue (or conference) that "right, as the world goes, is only in question between equals in power, while the strong do what they can and the weak suf- fer what they must" (5.89). Associated with doubts concerning the possibility of deriving constructive, practical "lessons from history," one finds aspects of this critical tone in denials by structural neorealists that some policies or states have a more moral character than others (Waltz 1979, 127, 187n.). More historically accurate (but almost as bleak) is Keohane's ap- praisal that "classical Realism, with its philosophical roots in a tragic conception of the human condition, directs our atten- tion in the twentieth century to the exis- tential situation of modem humanity.... But Realism, whether classical or structural, has little to say about how to deal with that situation" (Keohane 1983, 519).

Surely Thucydides is one of the first2 'scientific historians" (a phrase used in Jaeger 1976 and Strauss 1978, chap. 3; and supported by Keohane's arguments

[1983, 506-101 for continuities in the "hard core" of the realists' "research pro- gram" from Thucydides to Waltz). But I think his epistemological conception of "scientific history" or "scientific politics" can more accurately and suggestively be described as a morally engaged version of dialectical logic, whose closest descendant is argumentation theory (Barth and Krabbe 1982; Eemeren, Grootendorst, and Kruiger 1987). His historiography combines a commitment to factual accuracy with what we would now call a dramaturgical perspective on human affairs (Harre and Secord 1972). Ignoring its dialectical origins, we seem almost to have forgotten that his scientific political history is at the same time one of the greatest morality plays ever written, in essence a Greek tragedy.

To better understand the relevance of Thucydides' insights and historical- political science for today's problems, I shall focus on the dialectical ways in which he presents, reconstructs and uses the Melian dialogue. I shall amplify the argument that Thucydides consciously used a sophistic dialectic in the writing of this dialogue (Finley 1939, 1942). He did this with a style and degree of formalizable rigor that compares favorably with more conventional contemporary modes of logical and statistical analysis. In contemporary language, his polimetrics sought to uncover, through critical, rational argumentation, the inner deter- minative structures, the criticizable and partly changeable constitutive practices of Greek politics.

The Melian Dialogue As Partially Formalizable Dialectics

Neither Popular Rhetoric nor Apodictic but Formal Debate

There are several levels of discussion in the dialogue, each with different epistemo-

806

Thucydides' Melian Dialogue

logical import. The first level concerns the type of negotiation to take place between the Melians and the Athenians; it leads to agreement in favor of formal disputation.

The evidence for treating the Melian dialogue as a classical disputation or debate is explicit within Thucydides' text. Con- sistent with their oligarchic political character, initially derived from Sparta, their parent state, the Melians "did not invite [the Athenian] representatives to speak before the people [as would be the custom in a democratic state like Athens], but asked them to make the statement for which they had come in front of the gov- erning body and the oligarchicc] few" (5.84). The Athenians put a good face on this choice by treating it as a preference for serious argument rather than for the set speeches of popular rhetoric by which the masses might be "led astray": they propose that the Melians "should instead interrupt us whenever we say something controversial and deal with that before going on to the next point" (5.85). The Athenians propose point by point philosophical debate about controversial matters rather than popular rhetoric.

The kind of argument envisioned is also not what the Greeks (and Aristotle) would call apodictic (deductive, axiomatic, or necessary) reasoning. I suspect their model of such reasoning was geo- metric deduction, which had already achieved significant results by the end of the fifth century B.C. My evidence is the Melian statement that disputants "should be allowed to use and to profit by arguments that fall short of a mathematical accuracy" (5.90). Since their lives depend upon their persuasiveness, they want to make arguments that they can not strictly prove but against which effective counter- arguments might not be made.

Both parties clearly have in mind some kind of uncoerced formal disputation (or "dia-logic," [Habermas 1971]) before a judge and an audience. To use Aristotle's terminology, we are in the realm of "dia-

lectic," which he considered appropriate for political inquiry, the realm of arguable opinions about contingent or "probable" truths. Here dialectical inferences "must start from premises that command general assent rather than universal or necessary truths" (Aristotle 1964, 83, 153). Except for his somewhat more positive, democratic conception of rhetoric, Aristotle's Rhetoric and his Topics-together the first great Western treatise on dialectical argumentation-will assert the same distinctions several decades later (Arnhart 1981, chap. 2).

Coercive Irrationalities in the Dialogue

This agreement on the desirability of a dialectical mode of political discussion occurs in the realm of practical reason, where arguments start from reasonable assumptions, where strict deductions are often not appropriate, where better arguments still do not claim deductive certainty. But this realm is not completely "dia-logical"; nor does agreement extend to a second level of argument about the topic of discussion. The Athenian proposal only to discuss matters of self- interest is described without objection as "scarcely consistent with" the expectation of a reasonable exchange of views in "a calm atmosphere." The Melians see the Athenian invasion of their island as a "present threat" of making war against them, evidence that they have come "pre- pared to judge the argument [them]- selves" (5.86). The Athenians do not deny this coercive element. A similar response occurs later (5.90), when the Melians note that the Athenians "force us to leave justice out of account and to confine ourselves to self-interest" right after the Athenians have asserted that for "practical people, the standard of justice depends on the equality of power to com- pel" (5.89). Evidently, Thucydides thinks like modem argumentation theorists who

807


Figure 1. Rescher's Simplified Formulation of Dialectical Move Sequences

A. PROPONENT MOVES i. OPPONENT COUNTERMOVES a C. PROPONENT REPLIES

(5) Provisoed counterassertion

(2) Cautious denial Vs P/O & 10 (a prima face case)

t-P__ _ _ _ _ _ _ _

Please prove that P" or (6) Categorical counterassertion (1) Categorical asserttert " P is compatible with all I-R.

1 p you have said or conceded."

"I assert that P" "a (3) Provisoed denial (7) Provisoed counterassertion

possibly associated with a --P/fl & t V s ,/TN & I T.

(Ia) Provisoed assertion 'Typically - P obtains (8) Strong distinction/exception provided that Ro,,"

p'o &K to which I cautlousjyasser ;" P/(R, &.T,) & l(R, & T,) "P generally, typically obtains t when 0 does, ceterls parlbus (4) Weak distinction/exception (6) Categorical counterassertion and I assert that O"

- P/(O & RI) & t(O & RI) _I (O & R2) which serves as a "Isn't -P the case, at least (fPoiodcutrseto prima facde case for P. for those 0 for which R2 holds?" (7 Provisoed counterassertion

X - (O & R2)/Tz & IT:

(8) Strong distinction/exception

PI(Q&R2&T4)&l(O&RA2&T4)

Source: Rescher 1977, 6-15.

"It would also be possible to claim an entailment, such as P 1 S1 or (vis-a-vis move la) Q FS), and then indirectly attack either P or Q with cautious or "provisoed" denials of SI or S2.

would consider this conduct "irrational in [such a] dialectical situation."3

I next want to show a continuity between past and present epistemologies that has been neglected by neorealist writers: despite the coercive irrationalities, the third level of substantive discussion in the Melian debate, the debate proper, can be partially formalized and scientifically ra- tionalized according to Nicholas Rescher's pragmatic version of dialectical inquiry (Rescher 1977; another application is Alker 1984). I shall emphasize the practical, truth-revealing contribution of this mode of scientific discourse. Except for her neglect of formalizability, Jacqueline de Romilly makes the same point about dialectical creativity in clarifying Thucydides' achievement: "Thanks to the dialogue form, the only one in the whole of the History, the particular and the general mingle together. This form en-

ables maieutics to take the place of formal demonstration, so that the complete pat- tern Thucydides has in mind is revealed little by little, in all its aspects down to its deepest foundations" (1963, 274). The Greek root meaning of maieutics is particularly apt: "midwifery."

Rescher's Dialectical Logic

This demonstration requires first a summary of Rescher's formalization of the move and countermove possibilities within formal disputation, perhaps "the clearest, and surely historically the most prominent, instance of dialectical process."4 This formalization is illustrated in Figure 1. Rescher recognizes that other, more or less formal characterizations of rational argumentation exist (e.g., Loren- zen and Lorenz 1978). But this simplified version is adequate for his primary pur-

808


pose: to exhibit the "social nature of the ground rules of probative reasoning," the establishing of our factual knowledge through a social process of controversy- oriented rational argumentation.

Rescher's dialectical logic starts from, and fits, a social-legal scenario involving four unequal, truth-seeking roles: the proponent, the opponent, the judge or "determiner," and an audience. The proponent tries to establish a thesis, which the opponent attempts to discredit by insightful questions, denials or counterassertions directed against the proponent's assertions or their deductive implications. The proponent, like a legal claimant under Greco-Roman law, bears the burden of proof. Not only must he or she make an assertion, or "commitment," as part of every turn and avoid repetitions, but these assertions should nontautologically develop the argument with new grounds, distinctions, supportive theses or condi- tional reformulations. (Reformulations, such as moves la, 5, 8, and 8' in Figure 1, are said to "discharge" the proponent's responsibility to defend a previous categorical assertion.)

The proponent's role, like his or her categorical assertions (symbolically pre- ceded by exclamation points in the styl- ized tree of possibilities in Figure 1), is thus a "strong" one; the opponent's posi- tion is, relatively speaking, a "weak" one. Every unattacked assertion of the proponent becomes, at least temporarily, a "concession" by the opponent. Whenever the opponent makes a distinction as a way of limiting the scope of strong assertion, the stronger form of the assertion is also treated as a "concession." Although the opponent may develop counterargu- ments, he or she is not required to. The opponent's role requires only "weak" or "cautious" assertions (symbolically pre- ceded by a dagger) at each turn. Charac- teristically, each cautious assertion may be reformulated as a question, that is to say, move 2 may be read as "Isn't the

opposite of your thesis consistent with everything you have so far asserted or conceded?"

Rescher's determiner makes judgments on either formal or material criteria: the former include illicit circularities in reasoning, inconsistencies and incoheren- cies within one's belief system, and formal logical errors; the latter focus on the plausibility or implausibility of the proponent's "living," i.e. undischarged or unconceded, commitments. Appropriate evidential plausibility standards for participants and judges include source relia- bility and authoritativeness; the relative strengths of supportive evidence for rival theses; and the simplicity, regularity, uni- formity (with other cases), and presumptive normality of such theses, where these features are seen as supportive of inductive systematization. A proponent must deepen the grounding and improve upon the reasoning supporting his or her original theses; 'an opponent must be judged for success or failure in finding weak links in such support.

The judge additionally oversees the conduct of the argument. A major initial simplification (presumably adequate for Rescher's purposes in the early stages of his book), is the assumption that errone- ous evidential claims either are not made by the participants or are immediately corrected by the judge. Thus the determiner role has implicitly limited the move possibilities of Figure 1, where, for example, no circularities of reasoning are exhibited. This assumption has the important consequence for Rescher's model of formal disputation that disputing parties address themselves to matters of evidential substance rather than to the proprie- ties of formal or evidential reasoning. For- mally, this will mean that the major dialogical connective of presumptive support, the (weak or strong) assertion of P "provisoed" upon grounds Q, P/Q, is never directly attacked in his examples. Indeed, when such questions are directly

809


discussed in his later chapters, Rescher still does not provide a direct, formalized way of attacking P/Q. Only indirect challenges, involving the making of distinctions (or the allowing of exceptions) in the grounds supporting such assertions, are allowed. For example, although the strong exception move 8 of Figure 1, P/(R1 & T3) & I(R1 & T3), indirectly attacks the opponent's "provisoed" denial -P/R1 in move 3, the focus formally is on the second part, the strong distinction !(R1 & T3), which "adds T3" to the opponent's weak assertion TR1 in order to pro- duce a strongly qualified reformulation, or "rewrite," of it.

Taking pains to subsume standard forms of sentential and syllogistic logic within his own approach,5 Rescher never- theless emphasizes the differences be-- tween his formalizations and the standard contemporary one. Thus - symbolizes "negation" or "the denial of" the subsequent proposition, as it does in modem sentential logic, but for Rescher the weak assertion T P suggests a need for substantive "refinement," not necessarily total rejection. As we have just seen, Rescher's & also sometimes involves a much more substantial combinatorial re- writing than the modern, purely formal logical and.

The "provisoed" or presumptive inference from Q to P, P/Q, is very different from either the Q D P ("Q implies P") or Q F P ("Q entails P") of standard modem symbolic logic. Consider the way P/Q is read: "P normally, generally, typically, or as a rule follows from Q, ceteris paribus." Because this key inferential relation is only presumptive, it does not un- conditionally support detachment of a conclusion Q from P & (P D Q). Hence, '.'in dialectical (as opposed to deductive) reasoning an assessment of the cognitive standing of a thesis never leaves its probative origins behind altogether." Alterna- tively, without detachment each dialectical argument requires inspection of the entire history of its derivation.

A Partial Formalization of the Melian Dialogue

As a test of the applicability of Rescher- ian formalizations, I shall offer and then discuss a formalization of the first major substantive debate in the dialogue, whether the submission of Melos to Athens, either voluntarily or through coercion, can be said to be in each of their interests. An italicized Rescherian formalization of each of the seven "moves" or "turns" in that debate will be placed under the corresponding Crawley translation text, with my interpretive insertions in that text indicated by square brackets. (I leave the formal analysis of the second part of the debate, concerning the pru- dence of Melos's hope for deliverance, to others.)

Move 1. Athenians. (a) We will now proceed to show you that we are come here [seeking Melian submission, voluntarily, if possible] in the interest of our empire, and that [what] we are going say [is in your interest, i.e., is] for the preservation of your country;

Let P1 = Melian submission is in Athenian imperialtinterest and P2 = voluntary Melian submission is in its (and Athens') interest.

IP = I(P1&P2)

(b) as we would fain [preferably] exercise that empire over you without trouble [through Melos's voluntary submission], Let Q = imperially, Melos's voluntary submission is desireable.

P1/Q & IQ (c) and see you preserved for the good of us both.

Let R = preservation of Melos via voluntary submission.

P2/R & IR

Move 2. Melians. And how, pray, could it turn out as good for us to serve. [submit, become slaves] as [it is] for you to rule? Let P2 = P21 & P22, where P21 = Athenian interest in rule and P22 = Melian good from voluntary submission.

t(-P21 & ~-P22)

Move 3. Athenians. (a) [You would] avoid. suffering the worst;

810


Let W = Melos's avoiding the worst.

P22/W & 1W

(b) [we would gain a more valuable tributary] by not destroying [you]. Let X = greater Athenian tributary gain.

P21/X & IX

Move 4. Melians. [Why could we not remain] neutral, friends instead of enemies, but allies of neither side?

Let N = Melos's friendly neutrality and l= Athens' interest.

-PA(I & N) & t(I & N)

Move 5. Athenians. (a) Your hostility cannot hurt us,

V (I & N) (b) as your friendship will be an argument to our subjects of our weakness,

Let S = subjects' regard for Athenian power. - I/(- S/N) & 1(- S/N)

(c) and your enmity of our power. I/(S/-N) & I(S/-N)

Move 6. Melians. Is that your subjects' idea of fair play- Let T = (- S/N) & (S/I- N), i.e., "that"

no distinction between people . . . unconnected with you and . . . your own colonists or else rebels whom you have conquered? Let -D = (making) no distinction.

T j -D& tD

Move 7. Athenians. (a) (Yesl Our subjects] see no difference between right and wrong;

I- D

(b) those . . . still . . . independently do so because they are strong, and if we fail to attack them it is because we are afraid.

Let N1 = still being independent, - SI = their strength, and N2 = not being attacked by Athens, and -S2 = Athens' being afraid to attack.

- D/(Ni/ - S1 & N2/ - S2) & I(N1/S1 & N2/S2)

(c) [Sol that by conquering you we will increase [the] . . size ... [and] security of our empire. We rule the sea and you are islanders,

Let I, = increase in imperial size and security, 12 = Athenian interest as sea ruler via a vis island-

ers, and I3 = particular Athenian interest vis a

vis weaker islanders.

I I12

(d) and weaker islanders too than the others; it is therefore particularly important that you should not escape.

P/IMI & I2 & IW & I(II & 12 & IO)

The flow of Thucydides' argumentation evidences a truly remarkable fit with Rescher's rules of dialectical logic. As one kind of evidence, note the clear and repeated use of the specialized terminology of argumentation: proceed to show; and how, pray; argument; distinction; and therefore are terms of argumentative art that any broadly literate philosophical scholar will recognize. Secondly, there is the explicitly indicated "turn taking" man- dated by Rescher; the more demanding requirement that previous arguments be met with new argumentation also holds. Although compound theses were not mentioned in Figure 1-they are an easy, natural and common complication of that formalism-the argument forms, from categorical assertions like (la) to the "provisoed" counterassertions of (3a-b) and the weak distinction of move 4, clearly conform to Figure l's evolving grammar of move, countermove, and reply possibilities.

Coincident with their major show of force, the Athenians are the proponent, they articulate the major theses of this discussion, and they offer an argument sketch for them. As opponents, the Melians regularly question, or weakly assert their views (2, 4, and 6). As I observed first hand in Uppsala at Lars Udehn's 1987 public doctoral thesis defense, the polite, cautious, carefully questioning, cooperative yet critical truth-seeking style of Melian interven- tions is still practiced in European univer- sities 24 hundred years later (Steven Lukes was the "opponent"; see also the neglected discussion of debates in Rapo- port 1960.)

811


Further Reflections on This Correspondence

Going into the details of the argumentation formulae above opens up a deeper examination of some key issues raised by Rescher's proposal. Thus one finds there strong support for the centrality of ceteris paribus inference in (at least classical) political argumentation, thought of as a kind of "provisoed" assertion that sub- sumes standard formal inference and en- courages novel, more adequate substantive reformulations. As for Rescher's sub- sumptive approach toward standard formal logic, it is harder to map Thucydides' approach into the modem sentential cal- culus. But there are at least several cases where roughly equivalent versions of the same proposition, e.g., P1, P2, I, S. N, T, and D (and their negations) are logically substituted for each other in a modem way. Moreover, an apparently analytic claim appears when the Athenian interest in controlling islands is somehow deduced without further evidence from its interest and internal nature as a power who "rules the seas" (7c). (Aron 1984 [p. 29] nicely spells out the requirements for maritime domination.)

Beyond including something like standard deductive inference within his infor- mal but formalizable dialectical logic, Thucydides here seems to have encom- passed important aspects of the "internal relations" of Athenian domination. Brief- ly, "internal relations," like Hegel's master-bondsman relationship, are those that (at least partially) constitute the identities (or essences) of the entities being related. Variously formalized in the Hegelian, Marxian, and analytical traditions, these identity-affecting logical relationships are usefully clarified in Bhaskar 1986, Elster 1978, and Olman 1971. Since the argumentation shifts to another topic at this point, one is led to believe that at least the sketch of a deeper, provisionally plausible (and in effect conceded) ex- planation has been introduced.

Unlike standard formal logic, Thucydi- des' logic-his dialectic-is not tauto- logical. Rather it is a rule and standard- governed, hence partially formalizable, practical epistemology. Within it there is always the possibility of new supporting or critical reformulations: both P1 and P2 are attacked and defended in a "provisoed" way, with (7d) being a very significant restatement of probative grounds for P1, clearly more persuasive than those offered in the preliminary argument sketch.

In this light, the Melian attempt to reformulate Athenian imperial interest so as to make it consistent with Melian neutrality-the weak distinction of move 4-is an especially creative, revealing, as well as practically crucial, reformulation.7 Not only does it make possible the knowledge enhancement due to the "internal relations" account of (7c), but practically it is a valiant, if ultimately inadequate and tragic, effort to deter Athens' attack.

Note that in the subsequent round the Athenians are put on the defensive, to the extent that they are induced explicitly to assert the denial of the Melian neutrality thesis. The argumentative requirement to support this new, negative claim produces (5b-c), which in turn allow the most revealing-and for a relatively neutral Greek determiner, probably decisive- move by the Melians: their inferential assertion of the Athenians' underlying warrant of no difference, no differentia- tion, no distinction in conduct expected from neutrals, colonists, and rebels, which at (7a) the Athenians clearly ac- cept. For a society in which (according to Aristotle), justice is defined among edu- cated citizens in terms of proportionality, or appropriate distinction, not only does this admission speak against the Atheni- ans on this point, but it contradicts their own preference not to talk of the justice of their actions.

Rescher talks of dialectical negation as more than displacement, as "refinement." The formally correct law of double nega-

812


tion (. -p = p) is pragmatically aban- doned, replaced by the possibility of an original thesis being aufgehoben, preserved in a refined form through a feed- back "cycle of probative dialectic" that arises out of constructive, critical debate (1977, 67). His claim that his controversy- oriented dialectics produces cumulative, new knowledge, is either clearly or tenta- tively supported by each of the admis- sions (7a-d) of the seventh round of the Melian debate.

Rescher's preference for a dialectics of probative reasoning, one that does not challenge "provisoed" assertions per se, however, seems somewhat too restrictive. If one recognizes that P1 might aptly be respecified as I/M, where M = (Melian) submission, one sees more clearly the inti- mate connections between the earliest and later stages of the argumentation. Indeed, the challenges to P1 can all be said to challenge this implicit assertive connective. (See Devereux 1985 for the next steps here; reliable formalizations concerning contextually inferred elements are not, and may never be, easy to come by.)

Consonant with the possibility of deeper, more internally or characteris- tically revealing reformulations, a contemporary restatement of Thucydides' dialectical logic is as follows: Think (like Thucydides) of reality as layered and define scientific progress as the uncover- ing of the deeper, dialectically inner, determining layers or characteristic dis- positions. (Systemic, political, economic, sociocultural, and ecological relationships may also play such inner, identity-affecting roles.) Recognizing that political systems are open and "autopoietic" (Bhaskar 1986), hence not apodictically decom- posable, it is appropriate that investiga- tive inferences are all in the classically ten- tative, ceteris paribus mode of reasoning rather than the ontologically flat statistical sense of ceteris paribus relationships going back to Ando and Fisher 1962 and Ando, Fisher, and Simon 1963. Be-

cause classical ceteris paribus reasoning corresponds closely to the "default reasoning" and "nonmonotonic inference" of contemporary artificial intelligence research on commonsense reasoning, which allow rationally defensible, nonadditive reconceptualizations, I am optimistic that Rescher's approach can be improved upon by more rigorous, inter- nality-sensitive formalizations of dialectical, political reasoning (see Winograd 1980 and Reichman 1985).

Other Dialectical Elements in Thucydides' Historiography

We have now a better understanding of Thucydides' "logic," in the classical, dialectical sense of his (only partly formalizable) standards and procedures of valid argumentation and knowledge cumulation. In the one case where we have carefully examined its exercise, it has moved us more and more toward his ontology of things, his conception of the inner character of imperial systems, states, and states- men. We have not yet clarified and ex- plained the exceptional, dramatic, moral force of the Melian dialogue. Nor have we linked Thucydides' exceptional historical analysis of the Melian episode to the rest of his book. Each of these concerns also has a dialectical aspect.

Events Data and Speeches in Thucydides' Science

Let us begin with the way Thucydides interrelates events and speeches. Just as Finley observes (Thucydides 1951, xii), Thucydides' historiography "uniquely combined opposite tendencies of dialectic and observation, of generalization and observation." Throughout the book, speeches regularly come before the great actions (or "motions"). This sequencing is an explicit aspect of his interpretive, explanatory methodology: 'My habit has been to make the speakers say what was

813


in my opinion demanded of them by the various occasions, of course adhering as closely as possible to the general sense of what they really said" (1.21).

But this search for contextual appropri- ateness does not override the awesome sense 'of free, responsible, criticizable moral agency that this sequencing also conveys. Given the likelihood that many of the speeches were written or rewritten after the narrative was nearly finished (as Jaeger [1976, 342-47] and Romilly [1963, 275-86] argue), dialectically it is highly plausible to look for Thucydides' own critical interpretations of events, their motives, and consequences in the oppositions apparent in almost all of them. Indeed, the central role of the Melian massacre in postwar denunciations of Athenian imperialism suggests to Romilly a distinctive, quasi-legal significance for Thucydides' last revisions of the work, making it a "study [of] the policy of con- quest as a whole, in relation to certain criticisms that had recently been made of it" (p. 286).

Does not Thucydides' account of Athens' response to the Mitylene revolt (a more serious affront than Melian neutrality, but occurring earlier in the war [3. 1-501) contrast favorably with Athenian behavior at Melos? Cleon, "the most vio- lent man at Athens, and at that time by far the most powerful with the commons" argued for the same standard as imple- mented in Melos-killing all male defend- ers and enslaving the others-with similar statements like "if, right or wrong, you determine to rule, you must carry out your principle and punish [all] the Mitylenians as your interest requires"; but Diodotus recommends "moderate chas- tisements," killing the oligarchic leaders of the revolt, not potentially pro- Athenian democratic masses, who had indeed turned against the Mitylenian oli- garchs. Diodotus carries the day in a close vote with a final appeal that "good policy against an adversary is superior to the

blind attacks of brute force" (3.49). Given Thucydides' stated preference for a much more limited form of democracy (8.97), Cleon's excesses may be seen as struc- turally linked to a commons that is too large.

Even more momentous comparisons critically link Athenian rationalizations in the Melian episode with Pericles' original statements of Athenian war aims and the immediately following Sicilian debate between Nicias and Alcibiades, the principal instigator of the fateful Sicilian invasion. Nicias argues against "risking things present for the sake of things future and un- certain," against the "madness of attacking a land which, if they prevail, they cannot hold," and against falling "sick of a fatal passion for what is beyond your reach" (6.1-13). Alcibiades, on the other hand, not long after having helped under- mine Nicias's interim peace with Sparta by a spiteful trick (5.44), boasts that his "folly" (or "madness") brought benefits in alliances against Sparta and asserts that the Sicilian cities are unpatriotic, in- habited by "motley rabbles." He defends his extravagant sending of a record seven chariots to the Olympic games, scorns Nicias's do-nothing policy, calls for supporting their allies on distant Sicily, suggests that the initiation of a second front will "humble the pride of the Pelopon- nesians," and concludes that "a city not inactive by nature could not choose a quicker way to ruin" than inaction: "The safest rule of life is to take one's character and institutions for better and for worse, and to live up to them as closely as one can" (6.17-20). The interest of the multitude in the allies' exaggerated (and not disinterested) reports of great wealth in Sicilian temples and treasuries helps Alcibiades carry the day.

How clearly this debate, which repeats the theme of post-Periclean Athens' inher- ent aggressiveness, ironically replays the devastating strictures against groundless hopes in the Melian dialogue! How Alcibi-

814


ades' remarks contrast unfavorably with Pericles' speech at the beginning of the war! Like Nicias much later, Pericles had been "more afraid of [Athens'] own blunders than of the enemy's devices," pos- sessing many "reasons to hope for a favorable issue, if [the citizens] can con- sent not to combine schemes of fresh con- quest with the conduct of the war" (1.143). On this contrast, Thucydides himself makes a rare explicit judgment, not in a speech, but in comments that constitute his encomium for Pericles at his death, 21/2 years after the war began:

The correctness of his previsions ... became better known by his death. He told them to wait quietly, ... to attempt no new conquests.... [Rather,] what they did was the very contrary, allowing private ambitions ... to lead them into projects whose success would only conduce to the honor and advantage of private persons. [Moreover,] committing even the conduct of state affairs to the whims of the multitude ... produced A host of blunders, and amongst them the Sicilian expedition; though this failed not so much through a miscalculation of the power [of the Sicilians] . . . as through a fault in the senders, .... choosing rather [than continuing to help] . . . to occupy themselves with private cabals for the leadership of the commons, . . . [which] first introduced civil discord at home. (2.65-66)

The above extensive and rather exceptional quote confirms, then, the interpretation we have made on other grounds, that deliberative speeches like the Melian dialogue are to be compared for their implicit judgmental implications about the motives of the principal actors in Thu- cydides' history (see Mefford 1985).

Thucydides' Symbolic Oppositions and Principled Contradictions

Frequently, the debaters are leaders of different city-states or their contending factions; sometimes they are spokesper- sons for-or almost mythical symbols of -a frail but larger collective will polit- ically composed out of contradictory in- puts. Finley elaborates an oppositional

theme we have already seen in the Melian dialogue's explanatory account (Thucydi- des 1951, xii-xiii): Athens is characterized as a naval power, based on an extensive commercial economy and political democracy (excluding women and slaves but typically including many of the mili- tarily and economically necessary oars- men); Sparta's nature is that of a land power, based on helot agriculture, con- trolled by a conservative oligarchy and a weak, rotating monarchy. Athens "en- courages enterprise and initiative; Sparta emphasizes tenacity and tradition." These differences are discussed at length in the introductory sections (5.84-85) and the later parts (5.97, 99, 107-11) of the Melian conference.

Strauss puts some of these oppositions (and others) into a partial hierarchy of dialectical distinctions: "Just as humanity divides itself into Greeks and barbarians, Greekness in its turn has two poles, Sparta and Athens. The fundamental opposition of motion and rest returns on the level of Greekness; Sparta cherishes rest whereas Athens cherished motion" (1978, 157). Even more remarkable (and extreme) is his argument, "[In] the Pelo- ponnesian War . . . one sees Greeks at their peak in motion; one sees the beginning of the descent. The peak of Greek- ness is the peak of humanity. The Pelo- ponnesian War and what it implies ex- hausts the possibilities of man. . . . All human life moves between the poles of war and peace, and between the poles of barbarism and Greekness. By studying the Peloponnesian War Thucydides grasps the limits, . . . the nature of all human things. It is for this reason that his work is a possession for all times (1978, 157).

Despite their ethnocentric excess, consonant with our Rescherian analysis, these truly remarkable quotations point to the deepest answers Thucydides gave to the quest signaled by his words quoted in my opening paragraph. Raymond

815


Aron has made rather similar points: "Politics is dialectic when it unfolds between men who mutually acknowledge each other. It is war when it brings into opposition men who . .. wish to remain strangers to one another.... By the same token, we perceive why war is the com- pletion of politics and at the same time its negation" (1984, 23). The common thought is at once dialectical in its revela- tory search for fundamental, substantial oppositions and transformations and her- meneutical in its reconstructive sugges- tions of a grammar of motives, and of opposed organizing principles spanning or generating the space of possible human political activity (see Alker 1982 and Habermas 1973).

Thucydides' Greek Tragedy

At last we are ready for the crucial second thesis in this reconstruction of the Melian dialogue: Thucydides' scientific study of power politics is essentially dramatic and ultimately tragic. Because the failures of both the Melians (a Spartan colony) and their opponents, the Atheni- ans, are revealed in that dialogue, a larger Greek tragedy is suggested by it.

The first argument is a textual one. At the same time that it is a turning point in Thucydides' entire "scientific history," the Melian dialogue is pure drama (cf. Aron 1984, 28). To quote Cornford, at the point in the Melian conference where the dialogue begins, "the historian changes from narrative to full dramatic form, pre- fixing as in a play, the names-'Atheni- ans,' 'Melians'-to the speeches" (Corn- ford 1971, 174-87). The dialogue was in fact a principal feature of Greek tragedy of the period as practiced by Euripides and others. By Thucydides' shifting into this form, Cornford argues, the dramatic ironies of this pivotal turning point are highlighted. These concern both the blind hubris, Eros, insolence, and pre-Christian

hope infecting both the Athenians and the Melians.

The second argument for the dramatic nature of Thucydides' Melian dialogue is more substantive, with essentially the same tragic conclusion. It begins by not- ing how Thucydides' use of dialectical substance, "the overall category of dramatist," conforms to the dramatist's search for the roots of human action in verbal terms, for instance, symbolic oppositions, and principled contradictions. Strauss and Aron have already shown us how, "when men are treated in terms of other things, men may even be said to speak for the dumb objects [or forces] of nature" (Burke 1969, 33).

Burke then offers a directly relevant, dialectical conception of tragedy (and science):

Galileo speaks of experimental testing as an "ordeal." Stated broadly, the dialectical (agonis- tic) approach to knowledge is through the act of assertion, whereby one suffers or calls forth as counter-assertions the kind of knowledge that is the reciprocal of his act. This is the process embodied in tragedy, where the agent's action involves a corresponding passion, and from the sufferance of the passion [by the original agent or the empathetic observer] there arises an understanding of the act, an understanding that tran- scends the act. In this final state of tragic vision, intrinsic and extrinsic motivations are merged. ... Although purely circumstantial factors participate, . . . they bring about a representative kind of anecdote . . . that belongs with the agent's particular kind of character. (1969, 33)

As a Burkean "representative anecdote," the Melian episode reveals the blindness of "tyrannical Eros,". of blind Athenian hubris and its insolent defiance of the gods. (Cf. Burke 1969; Cornford 1971; and Strauss 1978, 225-27; Alker [1987] suggests formally reconstructing Thucydi- dies' narrative in terms of the fundamental tragic notion of a "mythical actor's self-annihilation.")

Burke even claims that "we can . . . catch glimpses of a relation between dialectics and [Platonic] mathematics . . . in the fact that mathemata means both

816


things learned . . . and the mathematical sciences" and that "we can discern something of the 'tragic' grammar behind the Greek proverb's way of saying 'one learns by experience'; 'ta patemata mathemata,' the suffered is the learned" (1969, 39-40). As Strauss says, the Peloponnesian "War surpassed the Persian War in regard to human suffering," caused (intrinsically) by men and (extrinsically) by nature (1978, 150).

It turns out, then, that the Melian dialogue is a key scene in a classical morality play about might and right, not simply an eternal statement of the mathematical truths of realpolitik. An increasingly blind, arrogant, lustful, imperious Athens will pay for its failings with the lives of many of its citizens and, eventually, with its independence as well. Like the most thoughtful modern realists, Thucydides joins normative, descriptive and explanatory investigations pointed toward the suffering-based "learning" of moral lessons.

On the Appropriation of the Classics

In a recent, postmodernist treatise, On Diplomacy, James Der Derian has at- tempted "to explain a[n international political] system by studying the genesis of its internal relations, which are seen as expressions of alienated powers." A stu- dent of a leading classicist of our era, Hedley Bull, Der Derian, too, cites the authority of Aron and Thucydides as support for his own approach, quoting a pas- sage pointing ironically to subsequent Athenian injustice: the Corcyrans say to Athenian and Corinthian representatives on the eve of the outbreak of the Pelopon- nesian War, "Every colony, if it is treated properly, honours its mother city, and only becomes estranged when it has been treated badly. Colonists are not sent abroad to be slaves of those who remain

behind, but to be their equals" (1987, 2, 6; Der Derian's emphasis). As equals, they could exhibit their practical, moral capacities for arriving at reasonable, just bases for concerted political action in times of need. His book thus suggests that certain aspects of a neoclassical orientation can legitimately be attributed to the present effort as well.

Neoclassical Polimetrics

I have objected to the view of mathematical realpolitik derived from Thucydi- des' work and attributed to him by many of Thucydides' realist and neorealist inter- preters. By showing how much of his analysis they (and deductive behavioral modelers like Zinnes 11975]) have lost, I have tried to motivate and to legitimate a more comprehensive form of rigorous, partly formal, political analysis. The dialectical methodology of Thucydides' classic work has been available in principle since the dawn of scientific history, yet the above formal analysis of the Melian dialogue still seems, to this author, full of as yet unrealized potential.

For one trained in the powers of formal analysis, that exercise suggests what a neoclassical polimetrics of political activi- ties and constitutional structures would look like (cf. Alker 1975). It would be a polimetrics dialectically grounded in human experience, focused on contextually appropriate practical-normative standards of just conduct and institutional worth. Whether addressing empirical topics or arguing policy alternatives, its preferred "logic" would be argumentation theory, using classical ceteris paribus reasoning. Quasi-experimental inference, conventional mathematical probability and statistics, and deductive modeling would be seen as special cases or com- ponents of this more general approach. (Dunn 11982] anticipates many of these arguments; Campbell [1982] concedes most of them.)

817


We have seen how political argumentation, if grounded in a cooperative, uncoerced, truth-seeking orientation and used skillfully to ask the right questions- those that critically probe the essential justifications-can suggest key determinants and consequences of socio- political identities, actions, relations, institutions, or contexts. Polimetrics built up from a foundation in political argumentation can help reveal the underlying, contestibly legitimate political orders (constitutions), the principled practices and constitutive relationships that, if known, will make possible further, even more penetrating, policy-relevant discus- sions of a critical and reconstructive sort. Its cumulative results are unlikely to be timeless laws; its universals are more like- ly to reflect hegemonic or pluralistic political achievements.

Der Derian's genealogical approach is justified as an effort to refute the existence of any ahistorical, timeless, defining essence of international politics. To study discursively, genealogically, historically the internal relations of power politics is to deny its eternal reality. My own re- reading of Thucydides' Melian dialogue suggests a similar, open-ended, dialectical epistemology of practical knowledge cumulation, ideally to be combined with competing sets of hypotheses concerning the inner and outer determinants of Athenian (and Spartan) actions. It thus locates itself where any epistemologically self-conscious contemporary approach to international politics ought to be-in the midst of contending claims about lessons of the past and possibilities for the future.

Thus we too may participate in the engaged, dramatic, but relatively objective and rigorous kind of historical political analysis that Thucydides was engaged in. Like the Athenians and their Greek friends before the war, we can scientifically appeal to dialectical judgements, relatively unconstrained argument and coun- terargument, "impartial laws . . . [and]

differences settled by arbitration" (1.76- 77). We can join different sides of the inner contradictions of "Greeks" and "barbarians," enlisting Thucydides' (or Euripides' even more antimilitaristic) insights in our own political-scholarly debates. By dialectically engaging ourselves in the endless, passionate search for objectively accurate and motivationally superior historical accounts, we critically renew the past and help create the future.

Notes

Since the 1980 draft of this paper, supported by NSF grant number 7806707 to the Center for Inter- national Studies at MIT, a John T. and Catherine D. MacArthur Foundation grant to that center and to the Swedish Collegium for Advanced Studies in the Social Sciences have furthered its refinement. Among the many commentators on earlier drafts, I am most indebted to Karl W. Deutsch, Erik Devereux, Roger Karapin, and F. J. Tickner. Dwain Meffordproposed I try to formalize Thucydides' dialogues.

1. The customary form of citation is to the books and paragraphs of the Greek text. Except where a secondary source is cited, my quotations are taken from the Crawley translation (Thucydides 1951), now in the public domain.

2. Jaeger (1976, 483) argues that Hecataeus first took "the scientific and rational approach to the facts of human life as the essence of history," while Herodotous gets the credit for introducing "the religious and dramatic element" into history. A more globally valid view should take into account Chinese Confucianism, Legalism, and Taoism (Frei- berg 1977).

3. Barth and Krabbe (1982, 58) associate with the opponent's role "an unconditional right to criticize" proponent's assertions, and reconstruct the pragmatic rules of logical argumentation to include rule FD E5super: "When party N performs a speech act that is not among those permitted to N by the rules of this system of formal dialectics, or if it performs a non-permitted, nonverbal action that reduces the other party's chances of winning the discussion, then N has lost all its rights in the discussion and N's behavior is to be called irrational with respect to the present dialectical situation by the company that has adopted this system of formal3 dialectics" (p. 63).

4. My partial summary of Rescher's (1977) argument, including quotations, are taken primarily from chapter 1, pages 1-34; see also chapters 2 and 5.

5. Indeed, Rescher suggests that his formal dialec-

818


tics goes nicely with Aristotle's formal syllogistic reasoning, modified to include an "all-things-being- equal" operator (1977, 13-14). He follows Aristotle on this point, who argues that syllogistic principles apply both to axiomatic demonstration and to dialectic (Aristotle 1964, 153-59).

An even more ambitious synthesizing move between the dialectical and standard, that is, formal and deductive, logical traditions is due to Paul Lorenzen, who defines the standard logical constants -&, v, -, if... then-dialectically (i.e., argumen- tationally), in terms of strip rules for their use in "critical dialogues" between proponents and opponents. "Such a 'social' definition of the logical constants may be regarded as a theoretical elaboration of Wittgenstein's notion of a language game." It suggests a new "dialogical or argumentational 'garb"' for "modem propositional and predicate logics" shown by Paul Lorenzen to be demonstrably equivalent to them. Lorenzen equates "classical logic with a logic of cooperative debates (dialectics, in Plato's and Aristotle's sense), and [twentieth-century] intui- tionistic (constructive) logic with a logic of competi- tive debates (eristics)" (Barth and Krabbe 1982, 12, 24, 55). The set of 21 definitions and 40-odd rules proposed for bringing both Cartesian monological axiomatics and post-Hegelian associative thinking into the discursive, socially and rationally arguable realm is given in chapter 3 of the Barth and Krabbe text, from which one rule was cited in note 3. The relevant original papers are collected in Lorenzen and Lorenz 1978. Further integrations and elabora- tions, of direct relevance to political science, are clearly needed.

6. Apparently, Sergeev's recent, rich, philosophi- cally informed formalization of Thucydides' argumentation omits the discussion of internal relations (Sergeev 1986). His published paper cites the Euro- pean literature of note 3 and the 1980 version of the present paper.

7. "Recognition of the central role of distinctions in the dialectical enterprise-based on the division (dihairesis) of key concepts-goes back at least to the Socrates of Plato's Phaedrus and is doubtless present in the theory and practice of the early Sophists" (Rescher 1977, 12, n. 13). Later (pp. 66-67), Rescher's description of "constructive negation" exactly fits the creativity of Melian distinction making: 'When P/Q is succeeded by - P/(Q & R), there is not just the displacement from P to - P, but also the refinement (amplification, improvement) from Q to (Q & R) . . : it advances the discussion and shifts the issue onto a more sophisticated ground."

References

Alker, Hayward R., Jr. 1975. Polimetrics. In Hand- book of Political Science, vol. 7, ed. Fred" I.

Greenstein and Nelson W. Polsby. Reading, MA: Addison-Wesley.

Alker, Hayward R., Jr. 1982. Logic, Dialectics, Politics. In Dialectical Logics for the Political Sciences, ed. author. Poznan Studies in the Phi- losophy of the Sciences and the Humanities, vol. 7. Amsterdam: Rodopi.

Alker, Hayward R., Jr. 1984. Historical Argumenta- tion and Statistical Inference. Historical Methods 17:164-73, 270.

Alker, Hayward R., Jr. 1987. Fairy Tales, Tragedies and World Histories. Behaviormetrika 21:1-28.

Ando, Albert, and Franklin M. Fisher. 1962. Two Theorems on Ceteris Paribus. American Political Science Review 56:108-13.

Ando, Albert, Franklin Fisher, and Herbert Simon. 1963. Essays on the Structure of Social Science Models. Cambridge: MIT Press.

Aristotle. 1964. Prior and Posterior Analytics. Ed. and trans. John Warrington. New York: Dutton.

Arnhart, Larry. 1981. Aristotle's Political Reason- ing. DeKalb: Northern Illinois University Press.

Aron, Raymond. 1984. Politics and History. New Brunswick: Transaction Books.

Barth, E. M., and Erik C. W. Krabbe. 1982. From Axiom to Dialogue. Berlin: de Gruyter.

Bhaskar, Roy. 1986. Scientific Realism and.Human Emancipation. London: Verso Books.

Burke, Kenneth. 1969. A Grammar of Motives. Berkeley and Los Angeles: University of Cali- fornia Press.

Campbell, Donald T. 1982. Experiments As Argu- ments. Knowledge 3:327-37.

Cornford, Francis M. 1971. Thucydides Mythistori- cus. Philadelphia: University of Pennsylvania Press.

Der Derian, James. 1987. On Diplomacy: A Gene- alogy of Western Estrangement. New York: Blackwell.

Devereux, Erik. 1985. Processing Political Debate: A Methodology for Data Production with Special Application to the Lincoln-Douglas Debates. Bachelor's thesis. Massachusetts Institute of Technology.

Dunn, William N. 1982. Reforms as Arguments. Knowledge 3:293-326.

Eemeren, F. H. van, R. Grootendorst, and T. Kruiger. 1987. Handbook of Argumentation Theory. Providence: Foris.

Elster, Jon. 1978. Logic and Society. New York: Wiley.

Finley, John. 1939. The Origins of Thucydides' Style. Harvard Studies in Classical Philology 50: 35-84.

Finley, John. 1942. Thucydides. Cambridge: Har- vard University Press.

Freiberg, J. W. 1977. The Dialectic in China: Maoist and Daoist. Bulletin of Concerned Asian Schol- ars 9:2-19.

Habermas, Jurgen. 1971. Knowledge and Human

819


Interests. Boston: Beacon. Habermas, Jurgen. 1973. Theory and Practice.

Boston: Beacon. Harri, Rom, and Paul Secord. 1972. The Explana-

tion of Social Behavior. Totowa, NJ: Rowman & Littlefield.

Jaeger, Werner. 1976. Paideia: The Ideals of Greek Culture. Vol. 1. Trans. G. Higet. New York: Oxford University Press.

Keohane, Robert. 1983. Theory of World Politics. In Political Science, ed. Ada W. Finifter. Wash- ington: American Political Scienue Association.

Keohane, Robert, ed. 1986. Neorealism and Its Critics. New York: Columbia University Press.

Lorenzen, Paul, and Kuno Lorenz. 1978. Dialogische Logik. Darmstadt: Wissenschaftliche Buchgesell- schaft.

Mefford, Dwain. 1985. Changes in Foreign Policy across Time. In Dynamic Models of Inter- national Conflict, ed. Urs Luterbacher and Michael Ward. Boulder: Lynn Rienner.

Morgenthau, Hans. 1978. Politics among Nations. 5th ed. New York: Knopf.

Ollman, Bertell. 1971. Alienation. Cambridge: Cambridge University Press.

Rapoport, Anatol. 1960. Fights, Games, and De- bates. Ann Arbor: University of Michigan Press.

Reichman, Rachel. 1985. How to Get Computers To

Talk Like You and Me. Cambridge: MIT Press. Rescher, Nicholas. 1977. Dialectics. Albany: State

University of New York Press. Romilly, J. de. 1963. Thucydides and Athenian

Imperialism. New York: Barnes & Noble. Sergeev, Victor. 1986. Strucktura politicherskoy

argumentatsii v (<Meliyskom dialoged Fukidida (The structure of argumentation in Thucydides' Melian dialogue). In Matematika v. izuchenii srednevekovykh povestvovatel'nykh istochni- kov, ed. B. M. Kloss. Moscow: Nauka.

Strauss, Leo. 1978. The City and Man. Chicago: University of Chicago Press.

Thucydides. 1951. The Peloponnesian War. Trans. Richard Crawley. Introduced by John H. Finley, Jr. New York: Modern Library.

Waltz, Kenneth N. 1979. Theory of International Politics. Reading, PA: Addison-Wesley.

Wight, Martin. 1978. Power Politics. Ed. Hedley Bull and Carsten Holbraad. New York: Holmes & Meier.

Winograd, Terry. 1980. Extended Inference Modes in Reasoning by Computer Systems. Artificial Intelligence 13:5-26.

Zinnes, Dina. 1975. The Scientific Study of Inter- national Politics. In Handbook of Political Sci- ence, vol. 8, ed. Fred I. Greenstein and Nelson W. Polsby.

Hayward R. Alker, Jr. is Professor of Political Science, Massachusetts Institute of Technology, Cambridge, MA 02139.

820

Dialectical Logic

Documents

categorical

mit press

der derian

melian episode

dialectical

melian debate

formal disputation

symbolically