My thanks to the University of Toronto and the Earhart Foundation for their support of my sabbatical year in Rome in 1980–81, where I wrote a first draft of a partial version, and to the Library of the American Academy in Rome and the Librarian, Rogers W. Scudder, for their hospitality and assistance at that time. Thanks also to Clare Wolfowitz for her help with the manuscript, and to Joseph Solodow for his comments on that version. My work on Thucydides and this article has since been supported by the Earhart Foundation, the National Endowment for the Humanities, the Social Sciences and Humanities Research Council of Canada, and Harvard University, to all of which my gratitude. This is a better article thanks to the suggestions of my referees as well as of colleagues and students who have read it.

^1. Hobbes, , English Works (London: John Bohn, 1839–1845), 8: vii–viii;Google ScholarAron, Raymond, “Thucydide et le récit des événements,” History and Theory 1 (1960–1961): 108–109;Google Scholar cf. Dewald, Carolyn, “Practical Knowledge and the Historian's Role in Herodotus and Thucydides,” in The Greek Historians: Literature and History (Essays Presented to A.E. Raubitschek) (Stanford: Stanford University Department of Classics/ANMA Libri, 1985), pp. 56–57.Google Scholar I have used the Greek text of Alberti, G.B., Thucydidis Historiae Volumen I: Libri I-II (Rome: Istituto Poligrafico dello Stato, 1972);Google Scholar all translations from Greek as well as modern languages are mine.

^2. Cf. Strauss, Leo, The City and Man (Chicago: University of Chicago Press, 1978 [1964]), pp. 157–58;Google ScholarMoralejo, Juan Jose, “La ‘Archeologia’ de Tucidides,” in Nuevos Studios de Literatura Griega (Madrid: Fondacion Pastor, 1981), p. 11.Google Scholar

^3. Cf. de Romilly, J., Histoire et raison chez Thueydide (Paris: Editions “Les belles lettres” 1956), pp. 240–98;Google ScholarSayas, J.J., “La ‘Arqueologia’ de Tucidides, esquema de comprensión de un desarollo económico,” Revista de la Universidad de Madrid 20 (1971): 23–36;Google ScholarParry, Adam, “Thucydides' Historical Perspective,” Yale Classical Studies 22 (1972): 51–57;Google ScholarFinley, M.I., “Myth, Memory and History,” in The Use and Abuse of History (London: Chatto & Windus, 1975), pp. 18–19.Google Scholar

^4. Cf. Cook, R.M., “Thucydides as Archeologist,” Annual of the British School at Athens 50 (1955): 266–70;CrossRefGoogle ScholarFinley, , “Myth, Memory, and History,” pp. 19–20.Google Scholar

^5. Cf. Finley, , “Myth, Memory, and History” p. 14;Google ScholarHunter, Virginia, “Thucydides and the Uses of the Past,” Klio 62 (1980): 192–93, 196;CrossRefGoogle ScholarHowie, Gordon, “Thukydides' Einstellung zur Vergangenheit: Zuhörerschaft und Wissenschaft in der Archäologie,” Klio 66 (1984): 529–32.CrossRefGoogle Scholar For an earlier version of such arguments, see Decharme, Paul, La Critique des traditions religeuses chez les Grecs des origines au temps de Plutarque (Brussels: Culture et Civilisation, 1966 [1904]), pp. 84–85.Google Scholar

^6. Erbse, Hartmut, “Zur Geschichtsbetrachtung des Thukydides,” Antike und Abendland 10 (1961): 25–26.Google Scholar

^7. Cf. Connor, W.R., “Narrative Discourse in Thucydides,” in The Greek Historians, pp. 3–4.Google Scholar

^8. Cf. Schmid, Wilhelm, Geschichte dergriechischen Literatur, 15 (Munich: Biederstein, 1948), 143 n.3;Google Scholar and Verdin, Herman, “Les remarques critiques d'Hérodote et de Thueydide sur la poesie en tant que source historique,” in Historiographia Antigua (Mélanges W. Peremans) (Symbolae, sen A, vol. 6.) (Louvain: Katholieke Universiteit Leuven, 1977), pp. 68–70.Google Scholar On Thucydides 1. 5–6, see also Heubeck, Alfred, “Thukydides I 5.3–6.6,” Hermes 94 (1966): 308–14;Google ScholarErbse, Hartmut, “Über das Prooimion des thukydideischen Geschichtswerkes,” Rheinisches Museum für Philologie, N.F. 113 (1970): 52–53;Google Scholar and Tasolambros, F.L., In Defense of Thucydides (Athens: Grigoris Publications, 1979), pp. 85–101.Google Scholar

^9. Cf. Plamondon, Michéle, La Mémoire et I'oubli dans la pensée grecque jusqu'à la fin du Ve siécle avant J.-C. (Paris: Editions “Les belles lettres,” 1982), pp. 265–66.Google Scholar

^10. Cf. Erbse, , “Uber das Prooimion,” pp. 57–58;Google ScholarTasolambros, , In Defense…, pp. 103–104;Google ScholarFunke, Hermann, “Poesia e storiografia,” Quaderni diStoria 12:23 (1986): 80–84.Google Scholar For an unusually forceful statement of the intransigence of Thucydides' rejection of the authority of the poets, see Detienne, Marcel, L'Invention de la mythologie (Paris: Gallimard, 1981), pp. 105–12.Google Scholar

^11. Just as it is not from the poets that he has concluded that “wealthy Corinth” was wealthy (1. 13. 5; cf. Iliad 2. 570).

^12. To say then, as Hunter does (“Uses of the Past,” pp. 196–98Google Scholar), that Thucydides “accepts Homer quite uncritically,” “and then goes on to interpret [the Homeric] data so as to prove his own personal thesis that fear motivated the other Greeks to accompany Agamemnon to Troy” is to miss the point entirely. For just by so interpreting “the data,” Thucydides is refusing to accept Homer uncritically—to say nothing of his extreme selectivity in accepting such “data” in the first place. Similarly, to proclaim as Täubler did a contradiction between Thucydides' stated attitude toward the poets and his reliance on them is to mistake the character of his reliance on them. Cf. Täubler, Eugen, Die Archäologie des Thukydides (Leipzig- Berlin: Tfeubner, 1927), p. 23.Google Scholar So too Moralejo, , “La ‘Archeologia' de Tucidides,” pp. 38–49Google Scholar, who argues that Thucydides accepts the tradition uncritically except where it disagrees with his own conclusions: a curious sort of uncritical acceptance!

^13. Romilly, , “A propos des commentaires de Thucydide sur la guerre de Troie,” Melanges Louis Jacob, Revue du Nord 36 (1954): 106.Google Scholar

^14. Cf. Pouncey, Peter R., The Necessities of War: A Study of Thucydides'Pessimism (New York: Columbia University Press, 1980), p. 45.Google Scholar

^15. “What is crucial is that [the theory of the remote past that Thucydides offers in the Archaeology] is a theory derived from prolonged meditation about the world in which [he] lived, not from a study of history” (Finley, , “Myth, Memory, and History,” p. 18Google Scholar). This is entirely true, and, once we recognize what it at issue in this passage, entirely necessary.

^16. On seeing and hearing compare Heraclitus, fragments 101a and 107 (Diehls), and Epicharmus, fragment 12 (Diehls).

^17. A number of writers have noted the role of eikos (“probability,” “plausibility”) in Thucydides' work as a whole and in the Archaeology in particular. Cf. most recently Gommel, jürgen, Rhetorisches Argumentieren bei Thukydides (Hildesheim: Georg Olms, 1966), pp. 48–49;Google ScholarMoralejo, , “La ‘Archeologia’,” pp. 32–36.Google Scholar These writers err, however, in contending that the invocation of eikos, common in the rhetoric of Thucydides' day (and so also in the speeches in his work) marks a merely rhetorical argument; they miss the deeper significance of the turn from hearsay to a standard supplied by human reason. It is often remarked by Thucydides' detractors that in fact he neither knows nor cares very much about the events of remote antiquity. (Cf. e.g., Finley, “Myth, Memory, and History”; Cook, “Thucydides as Archaeologist”) He claims, however, to know quite enough about them, not through having learned how things were but through having understood how men are. The knowledge that vouches for the authority of his account of the past is the same that vouches for its utility in the future; knowledge of to anthrōpinon (“that which is human,” 1. 22. 4). Cf. Strauss, The City and Man: “Thucydides makes one wonder whether anything can be known about the most ancient things. Yet the oldest things include the things which are at all times, and it is with things of this kind that the possession for all time is concerned” (p. 159). Thucydides presents himself as living in a privileged time, but it is a time which is privileged in good part because he lives in it.

I cannot agree with Elisabeth Young-Bruehl that 1. 22. 4 implies a rejection rather than an affirmation of the validity of the notion of “the human.” Her construction of the meaning of the sentence is highly implausible, and the only usage of “human” that she adduces from elsewhere in Thucydides (at 5. 68. 2) tends against her rather than in her favor. Young-Bruehl, , “What Thucydides Saw,” History and Theory 25 (1986): 4–5.CrossRefGoogle Scholar

^18. Compare Iliad 1.121, 6.467; Erbse, , “Zur Geschichtsbetrachtung,” pp. 22–23;Google ScholarMoralejo, , “La ‘Archeologia’,” pp. 46–47.Google Scholar

^19. Thucydides' use of Homer at 3. 104, where he once again accepts him as an authority for the times in which he lived, and where he offers his longest citation of him, must be read in light of the following facts, among others. In all the works, epic and lyric, extant in Thucydides' day under the name of Homer, only in the passage thus cited does the poet address his audience, speak of himself and state the object of his art; that object is the undying praise of a chorus of maidens, to which end Homer renders these maidens an homage which might be regarded as extravagant.

^20. Cf. Grosskinsky, August, Das Programm des Thukydides (Berlin: Junker & Dünnhaupt, 1936), pp. 14–24.Google Scholar

^21. While the discussion that follows will focus on Thucydides' statement on speeches, his statement on deeds is also a highly difficult and contentious one. My translation of it follows the persuasive exegesis of Schepens, G., L'‘autopsie’dans la méthode des historiens grecs du V siécle avant J.-C. (Brussels: Koninklije Akademie, 1980), pp. 94–146, 188–91.Google Scholar

^22. So Grosskinsky, , Programm, pp. 28–58;Google Scholar minimized by Gomme, , “The Speeches in Thucydides” Essays in Greek History and Literature (Oxford: Clarendon Press, 1937), pp. 159–60,Google Scholar and An Historical Commentary on Thucydides, vol. 1 (Oxford: Clarendon Press): 140–41;Google Scholar but cf. Pearson, Lionel, “Thucydides as Reporter and Critic,” Transactions of the American Philological Association 78 (1947): 40–42;Google ScholarEdmunds, Lowell, Chance and Intelligence in Thucydides (Cambridge: Harvard University Press, 1975), p. 166, n. 28;CrossRefGoogle Scholar and Jebb, R.C., “The Speeches in Thucydides,” in Hellenica, ed. Abbott, Evelyn (London: Longmans, Green, 1898), p. 252,Google Scholar n.l., who cites another feature of the antithesis between the two passages as having “the special effect of bringing out the antithesis between facts of speech and facts of action”

^23. Grosskinsky, , Programm, pp. 28–58;Google Scholar Pearson, “Thucydides as Reporter and Critic”; Strauss, , City and Man, pp. 164–65;Google ScholarSchepens, , “L'‘autopsie’,” pp. 94–146.Google Scholar

^24. Cf. Jebb, R.C., “Speeches,” pp. 252–53,Google Scholar who suggests what Thucydides might have written here had accuracy been his standard for the speeches. For the argument that Thucydides has aimed at the maximum feasible accuracy for the speeches, see Grene, David, Greek Political Theory (Chicago: University of Chicago Press, 1965), pp. 20–23;Google ScholarGomme, , “Speeches,” pp. 156–89,Google Scholar and An Historical Commentary, 1: 138–39;Google ScholarKagan, Donald, “The Speeches in Thucydides and the Mytilene Debate,” Yale Classical Studies 24 (1975): 75–77;Google ScholarCogan, Marc, The Human Thing (Chicago: University of Chicago Press, 1981), pp. x–xvi.Google Scholar Each of these arguments suffers, it seems to me, from inadequate attention to the contrast between Thucydides' statements of his handling of speeches and of deeds, and within the former, from an unconvincing minimization of the significance of the deonta clause.

Even so this position is far more credible than the opposite one that the speeches owe next to nothing to their originals and even that Thucydides has fabricated them out of whole cloth, inventing speeches where none were given. Cf. Eduard Schwartz, review of Taeger, Fritz, Thukydides, Gnomon 2 (1926): 73–80;Google ScholarFlashar, Helmut, Der Epitaphios des Perikles (Heidelberg: 1969), pp. 5–6.Google Scholar To this the proper response is that of Kagan, “Speeches”: “The fact is that no one has shown that there is a single speech in Thucydides … that could not have been given in something like its Thucydidean form” (p. 77). So too Gomme, “Speeches,” and Adcock, F.E., Thucydides and His History (Cambridge: Cambridge University Press, 1963), pp. 28–33;Google Scholar cf. Grant, J.R., “A Note on the Tone of Greek Diplomacy,” Classical Philology 15 (1965): 261–66.Google Scholar

^25. Edmunds, , Chance and Intelligence, p. 166,Google Scholar following Luschnat, Otto, “Thukydides,” Realenzyklopädie, Supp. 12 (1971), cols. 1167–79.Google Scholar Although I concur only partly with Edmunds's thoughtful interpretation, I have learned much from it—above all that the crucial question here is the meaning of ta deonta, “the needful things” with which Thucydides embellishes the speeches.

^26. Wilson, John, “What Does Thucydides Claim for His Speeches?,” Phoenix 36 (1982): 97.CrossRefGoogle Scholar

^27. “Esprit,” Girard, Jules, Essai sur Thucydide (Paris: Hachette, 1884), p. 58;Google Scholar “Willensrichtung im ganzen,” “praktisches Zweck,” Schwartz, review of Taeger, pp. 79–80; “Gesamtintention,” Flashar, Epitaphios, 5; “general intention,” Proctor, Dennis, The Experience of Thucydides (Warminster, England: Aris and Phillips, 1980), p. 150;Google Scholar “senso generale,” Canfora, Luciano, Tbtalità e selezione nella storiografia classica (Bari: Laterza, 1972), p. 33;Google Scholar “Gesamttendenz,” Grosskinsky, , Programm, p. 33;Google Scholar “proposal,” Dover, K.J., Thucydides (Oxford: Clarendon Press, 1973), p. 22.Google Scholar “Main thesis etc.” expresses the position of de Ste-Croix, G.E.M., The Outbreak of the Peloponnesian War (London: Duckworth, 1972), pp. 7–12.Google Scholar

^28. Wilson, , “What Does Thucydides Claim for His Speeches?” pp. 97–98.Google Scholar While I disagree with Wilson's overall conclusions, his is by far the best discussion of xympas. As for the substantive to xympan (“the main thing,” “the long and the short of it”), which we find in the mouths of Thucydides' speakers, we do not find it here, and I cannot follow De Ste-Croix, (Origins, p. 9)Google Scholar in construing hē xympasa gnōmē as its synonym.

^29. Grene, , Greek Pblitical Theory, p. 22.Google Scholar There being, however, no clause of this difficult passage which has not been ardently disputed, Wilson has suggested that “keeping as closely as possible to” “gives far too specific a sense to the middle voice of echō, which has the very general meaning of ‘hold onto,’ ‘cling to,’ or ‘keep to’(but not necessarily in the sense of follow).” Wilson proceeds to argue that while Thucydides claims to “take the complete gnōmē into account” in composing a speech, he does not in fact claim to furnish us with it (“What Does Thucydides Claim …?” p. 99). This seems, however, an odd interpretation of “keeping to as closely as possible”: the superlative adverb argues against too weak an interpretation of the verb. It should be added that in Thucydides' other uses of the middle voice of echō in conjunction with gnōmē, the sense of the verb seems an emphatic “hold fast to”; 1.140.1; 8.81.1. Tasolambros, , In Defense…, p. 134Google Scholar, suggests an equally imaginative interpretation of “as closely as possible” (hoti engytata), which however depends on interpreting xympasa gnōmē as “general purpose.”

^30. 1. 44, 45. 1; 2. 20; 3. 92; 6. 47. 1; 6. 50. 1; 6. 71–72. 1; 7. 48. 1 (egignosken); 7. 52. 3, 4; 8. 87. 2, 6.

^31. “What Does Thucydides Claim…?” p. 99.Google Scholar

^32. Cf. Hudson-Williams, H.LI., “Thucydides, Isocrates, and Rhetorical Composition,” Classical Quarterly 42 (1948): 76–81.CrossRefGoogle Scholar

^33. A reader of an earlier version of this article has suggested that I erred in translating the adverb malista as “best” or “in the highest degree,” that is, as possessing true superlative force; that here as often elsewhere in Thucydides it means rather “approximately,” “more or less.” An exhaustive survey will confirm, however, that malista conveys this latter sense in Thucydides only in certain expressions of number, and that elsewhere it invariably possesses true superlative force. It clearly does so in the other three cases where it modifies cognates of deonta; i.e., 3. 13. 7; 4. 5. 2; 4. 92. 3.

^34. Benardete, S.G., “Dei and Chrē in Plato and Others,” Glotta 93 (1965): 293, 295 n. 2.Google Scholar

^35. So Forbes, W.H., ed., Thucydides Book One (Oxford: Clarendon Press, 1895),Google Scholarad loc: “‘what was wanted,’ what was appropriate to each occasion, what to the best of Thucydides' own judgment the circumstances called for … the best arguments for peace and war, or for severity or mercy to revolted allies, the most appropriate praise or blame for Athens or Sparta, Thebes or Plataea … on any given occasion.” So too Marchant, E.C., ed., Thucydides Book One (Bristol: Bristol Classical Press, 1982 [1905]),Google Scholarad loc: “the best arguments that could be found to support the xympasa gnōmē ofthe speaker.” Cf. also Maddalena, Antonio, Thucydidis Historiarum Liber Primus (Tbmus I) (Florence: La Nuova Italia, 1951),Google Scholarad loc.

^36. Hobbes, pp. vii–viii; Aron, , “Thucydide,” p. 110;Google ScholarDewald, , “Practical Wisdom,” pp. 57–58.Google Scholar

^37. Cf. Girard, , Essai, pp. 41–48.Google Scholar

^38. The speeches abound in interplay which cannot have figured in the originals. On this crucial feature of Thucydidean composition see Romilly, , Histoire et raison;Google ScholarStrauss, , The City and Man, pp. 163–74;Google ScholarWallace, W.P., “Thucydides,” Phoenix 18 (1964): 251–61;CrossRefGoogle Scholar and (more recently) Pouncey, , Necessities of War, pp. 13–15;Google ScholarRawlings, Hunter R. III, The Structure of Thucydides' History (Princeton: Princeton University Press, 1981);CrossRefGoogle ScholarConnor, , Thucydides (Princeton: Princeton University Press, 1984).Google Scholar

^39. Finley, J.H., Thucydides (Cambridge: Harvard University Press, 1942), pp. 95–99;CrossRefGoogle ScholarEdmunds, , Chance and Intelligence, pp. 166–68;Google ScholarTasolambros, , In Defense…, pp. 132–33;Google Scholar and others have argued for the distinction between the deonta appropriate to political speeches and those appropriate to “historical” ones, i.e., to speeches intended not to sway a crowd but to inform posterity, and have suggested that the deonta with which Thucydides furnishes his speeches are of this latter sort. On my reading of 1. 22, if he provides the “needful” in the “historical” sense it is only within the bounds imposed by the “needful” in the “political” one.

^40. The most impressive examples of speeches that communicate with listeners (and readers) on more than one level are that of Hermokrates at Gela (4. 59–64: the most neglected of the great speeches in the work), Perikles' famous Funeral Oration (2. 35–46), and the debate between Kleon and Diodotos (3. 36–49). All not only practice carefully nuanced communication but make clear (albeit and necessarily only by means of such communication) the necessity of practicing it. On this side of Kleon and Diodotos see Orwin, Clifford, “Democracy and Distrust: A Lesson from Thucydides,” The American Scholar (1984): 313–25;Google Scholar and “The Just and the Advantageous in Thucydides: The Case of the Mytilenaian Debate,” American Political Science Review 78:2 (1984): 485–94.Google Scholar

^41. Cf. Strauss, , The City and Man, pp. 163–74.Google Scholar

^42. Cf. Thibaudet, Albert, La Campagne avec Thucydide (Paris: Editions de la Nouvelle Revue Française), p. 49.Google Scholar