The society of the polis was founded on the skillful use of the word {logos), and this sufficiently explains something we find abnormal: in Herodotus and Thucydides, in many respects the heirs of Homer, speeches in direct discourse hold the notable place that epic already acknowledged. For Thucydides’ readers, Pericles is the man who invades the Megarid at the war’s beginning or dies of the plague, but he is also, and most importantly, the orator who gives memorable speeches. Yet in the movement from epic to historiography, attention to the word created a question of methodology: in the absence of written evidence for speeches, how was one to present the thoughts, intentions, and plans expressed in public by various people?
Literal reproduction was admittedly impossible: the speeches were not transcribed documents and a certain amount of personal intervention by the historian was inevitable. According to some, this is one of the ‘‘weaknesses’’ of ancient historiographical method, namely that it was governed more by literary preoccupations than by scrupulous documentation {Murray 1996: 375). Before passing judgment, however, it is advisable to clarify what this personal intervention consisted of. Thucydides, in the famous ‘‘methodological’’ chapter of his History, puts forth the problem in this manner {1.22.1, tr. Smith):
Therefore the speeches are given in the language in which, as it seemed to me, the several speakers would express, on the subjects under consideration, the sentiments most befitting the occasion {hosd’an edokoun emoihekastoiperit(>n aieiparont(>n ta deonta malist’eipein), though at the same time I have adhered as closely as possible to the general sense ofwhat was actually said {echomeniii hoti eggutata tls xumpaslsgnSmls ton alothos lechthentOn eirOtai).
Reading this brief passage - in which almost every word has generated many questions and discussions - one is first struck by the contrast between the requirement of
Translated by Ginevra Adamoli and Kyle M. Hall.
Adhering to the substance of the spoken words ( echomenoi hoti eggutata tes xumpasls gnomes ton alethos lechthenton) and the requirement of verisimilitude (hos d’an edo-koun emoi hekastoi peri ton aiei paronton ta deonta malist’ eipein). Does Thucydides summarize real speeches or does he invent them? Or does he do both?
It seems that there are many ways of interpreting and justifying the ambiguities of the text. One could emphasize one aspect or the other: to consider, for example, the subjective element (hos d’an edokoun, etc.) dominant and the objective element as a small correction or a less significant addition (echomenoi hoti eggutata, etc.). A possible interpretation of this is that Thucydides would first have written what needed to have been said, including what was actually said only when the two coincided (Cole 1991: 104; Debnar 2001: 14-18 builds on the same presuppositions; on the opposite side, Vattuone 1978: 29-41, 216). Or one accepts both factors - subjective and objective - in their contradictory nature, as proof that ‘‘two hearts beat in Thucydides’ breast’’ (Hornblower 1994b: 45; cf. Hornblower, CT I.59-60). Another way is to distinguish the object of the two parts of the text: the freely created speeches on one side and on the other the speeches actually produced and in some measure reconstructable (Rusten 1989: 14); in other words, the form of the orations and, separate from this, their content (Garrity 1998).
I will leave to the reader the judgment on each of these hypotheses; none of them, however, truly resolves the contradictions of the passage: all presuppose that there will perceptibly remain a certain tension between elements that are discordant, or partly discordant. But is this necessarily the case? An attempt to reinterpret the text might bring a new insight. I tried to do this some years ago (Porciani 1999), and I present here the general concepts of that essay, adding a new parallel from Plato’s Phaedrus that appears to me to support it.
The idea brought forward in my essay is that an... malist’ eipein can be recognized as one of the numerous cases in which the adverb malista reinforces the potential modality of the action. In this interpretation, an. . . eipein is considered potential rather than unreal (as is perfectly possible, pace, e. g., Rusten 1989: 13 n. 46: see Porciani 1999: 119-121). Among the many Thucydidean examples of this construction we can look at the case of 4.80.3, when the Spartans falsely offer freedom to the helots: the Spartans, Thucydides says, thought that the first to request freedom would have been those who most probably would rebel against them (hcgoumenoi toutous sphisin... malista an kai epithesthai). Here the function of the adverb is to accentuate the possibility that the action will verify itself (cf. Jowett 1881: ‘‘would be... most likely to rise against their masters,’’ and Hornblower, CT II.266; Smith 1921-1930: ‘‘would be the most likely to attack their masters’’; Romilly, Weil, and Bodin 1953-1972: ‘‘etaient aussi les plus capables, eventuellement, d’une rebellion’’). In another passage (6.49.2), Lamachus supports the plan to attack Syracuse without delay; in this way the Athenians will have the greatest possibility of success (malist’ an spheis perigenesthai, cf. Romilly et al. 1953-1972: ‘‘c’etait la meilleure chance qu’on eiat de prendre le dessus’’).
Other passages in which modal accentuation with malista can be seen are 1.76.4; 2.48.3; 4.18.4; 5.22.2; 6.22, 38.4; 7.8.2 (cf. Porciani 1999: 108-109; for other possible attestations, like 3.53.1, cf. ibid. 109-110), where we have a potential in a dependent form, expressed by an infinitive with the particle an governed by a verb of thinking {hegoumai, nomizo, oimai, dokeo), or, in rarer cases, of speaking (cf. 6.49.2). The syntactical structure is the same as that which we find in 1.22.1: there past potential rather than the present potential is at issue: and in fact the historian looks to the past, where his characters concern themselves with questions of a pragmatic nature.
From this interpretation a precise idea of Thucydides’ procedure emerges: namely, the search for the most likely probability in the reconstruction of speeches. That part of 1.22.1 that we have defined as ‘‘subjective’’ sees its own axis shifting with that of the ‘‘objective’’ part. The discourses were written ‘‘as it seemed to me that each speaker was most likely to have spoken, etc.’’: one sees here the scrupulous nature of an investigation that seeks the best information on the content of the speeches, and not the creativity of one who writes what ‘‘each speaker might have been able to say’’ (i. e., an unreal condition: could have said, but did not). The link with the subordinate phrase, ‘‘holding myself as close as possible to the xumpasa gnome of what was actually said,’’ does not now create any difficulty: the coherence of the whole is attained.
It remains to understand precisely the sense of two key elements of the passage: the direct object of the infinitive eipein, i. e., ta deonta, and the genitive tes xumpasOs gnomes. In the language of oratory, to speak ta deonta signifies ‘‘to say what one needs to do,’’ ‘‘to give good advice’’ to the assembly (Canfora 1990: 322, 355; cf. esp. Classen and Steup 1892-1922: I.77; II.158), or ‘‘to indicate what to do.’’ Parallels with Demosthenes, in particular, confirm this (Canfora), and we can find relevant passages also in Thucydides (e. g., 1.138.3; 2.60.5; Porciani 1999: 113). It is clear, however, that ‘‘what one should do’’ must be considered from the perspective of individual speakers: in this way Thucydides’ two opposed discourses (the so-called antilogies) can be explained. It is conventional to render ta deonta as ‘‘that which was appropriate’’: that is possible, naturally, provided that ‘‘appropriate’’ is understood as being the speaker’s, not Thucydides’, point of view.
What is the xumpasa gnome? Is it the content or the general sense of the speeches (cf., e. g., Kruger 1858-1861: I.1, 30; Mazzarino 1965-1966: I.258; Rusten 1989: 11; Hornblower 1994a: 45; cf. de Ste. Croix 1972: 9-10 [‘‘main thesis’’])? Or the intent or general scope of the words actually spoken (e. g., Schwartz 1926: 80; Egermann 1972: 579-580; Badian 1992: 189; Winton 1999: 530; Plant 1988 seems heterodox but cf. Valla’s translation: communi opinione proxime ad veritatem accedere)? To understand this, we need to look at the passage where Nicias, during the Sicilian expedition, decides to send to Athens a written message, so that his report is as exact as possible and the Athenians will not underestimate the emergency (7.8.2). As in 1.22.1, this passage refers to the difficulties of oral transmission (those in the message entrusted to the ambassadors in 7.8.2, those of the speeches in 1.22.1), attributable to (in 7.8.2) ineffectual public speaking, defects of memory, or a tendency to flatter the audience, and (in 1.22.1) the limitations of memory.
What is the solution to these difficulties? In 7.8.2 it is a written compilation that allows the recipients of the letter to become aware of Nicias’ ‘‘undamaged’’ or ‘‘integral’’ thought (gnomOn meden... aphanistheisan), and therefore to make their decisions based on ‘‘authentic data’’ (tes aletheias). In 1.22.1 it is the search for the greatest adherence to xumpasagnOme ton alethos lechthentOn, which I would translate as ‘‘the entire thought unveiled in what was truly said,’’ or, to modernize a bit, ‘‘the entire logical structure of what was truly said.’’ In both of the passages, the truth ( alotheias, alOthOs) is the basis of authenticity, of the facts in and of themselves, of the reality inasmuch as it is independent from the subject; the idea or thought {gnomO: for this sense see 1.54.2; 2.20.1; cf. Porciani 1999: 128-129 with n. 59) reaches this level if it conceals nothing { mOden... aphanistheisan) or is complete, whole {xum-pasa).
In this interpretation, the adjective in the phrase tOs xumpasOsgnOmOs has its usual meaning of‘‘complete,’’ ‘‘total’’ {cf. Wilson 1982: 98), and not that of‘‘general’’ as the opposite of ‘‘exact,’’ ‘‘precise,’’ which is invoked by those who assign gnOmO the meaning ‘‘sense’’ or ‘‘intention.’’ This is confirmed by a passage in Plato’s Phaedrus {I thank Andrea Zambrini for having brought this to my attention), where Phaedrus explains to Socrates how he can report Lysias’ discourse about love {228d):
In reality, Socrates, I did not actually learn the words by heart; instead, almost all of the series of arguments {tOn mentoi dianoian schedon hapantoin) by which he showed the difference between the situation of one in love and one not in love I will go through summarizing it point by point, in order, beginning from the first.
Here dianoia is the thought or reasoning that manifests itself as the development of arguments for the demonstration of a thesis: arguments that have an order, as Phaedrus intends to lay them out beginning ‘‘from the first.’’ The term dianoia is one of the most frequently occurring synonyms of gnomO, and I would suggest that, in the Platonic passage, tOn... dianoian... hapanton expresses the same concept as tOs xumpasOsgnOmOsin Thucydides {even the adverb schedon, ‘‘almost,’’ finds a parallel in Thucydides’ more decisive and optimistic hoti eggutata, ‘‘as close as possible’’): the logical structure of a discourse, given by the succession of the arguments in a certain order.
We can now translate 1.22.1 as follows:
I wrote the discourses as it seemed to me that each speaker was most likely to have advised what had to be done in each situation, holding myself as close as possible to the entire reasoning laid out in the speeches that were actually spoken.
Characteristic points of this translation are: the force of the past potential attributed to an... eipein {‘‘that each speaker was most likely..., etc.’’); the concretely political, rather than rhetorical-literary, significance of ta deonta {cf. the different perspective of Nicolai 1998: 290; 1999: 281-282, 296 on ‘‘political exemplarity’’ in Thucydides); and finally the analytic rather than the synthetic value of the nexus xumpasa gnomO. The search for the greatest probability in 22.1 prepares for the demand for truth regarding the deeds {erga) of 22.2: therefore one can truly say that ‘‘the basic rule expressed in 22.1-2 is uniform: the reader will find in the xungraphOe of Thucydides the most faithful reproduction possible of all that was said and done. Only the degree of this fidelity differs in the two cases’’ {Fantasia 2004: 46-47).
To take up a position on Thucydides’ speeches has implications that go beyond the correct appreciation of a passage in a Greek text. ‘‘It would not occur to a modern historian to insert in his works versions of speeches delivered, or reputedly delivered, by historical characters’’ (Walbank 1965: 242); ‘‘the discourses constitute an insurmountable obstacle for all those who sought to treat Thucydides as a colleague’’ (Nicolai 1998: 289). There are several formulations with which academics, in the last decades, have expressed a seemingly natural sense of distance from this peculiarity of ancient historiography, direct speeches.
Today there are contemporary historians who, in looking at the period after the Second World War, give much space to very specific testimonies, such as the telephone conversations of US presidents or other widely known personalities; in this case, however, there is the aid of technology and the researcher can rely on the records of the White House or on the wiretaps of the FBI (cf. Branch 1988, 1998). Obviously, the ancients did not have similar access to technology; and also because of this, in the absence of external evidence for their research, many are ready to believe that their historiographic method was open, at least in some measure and especially in certain situations, to free invention.
Also because of this, we said; but not only because of this. There is a cultural background of great importance that, even today, justifies and favors a similar view. During the 1970s talk began, for example, of the birth of a ‘‘new Thucydides’’ (Connor 1977). The detached and rational historian, pre-illuminist or pre-positivist of the ‘‘tradition’’ seemed to give way to the emotional writer, caught up in the vicious and dreadful material that he narrates, thus, to the artist and, at times, the philosopher of history who arranges the facts in an architectural structure. In the meantime, the portrait has been enriched, and much in particular has been said on the relationship between Thucydides and the literary ambit of rhetoric and oratory (cf. Cole 1991; Nicolai 1998; for a well-balanced position, see Rood 1998); the tendency at any rate was already well delineated in the 1970s. This resulted in a ‘‘new’’ image of Thucydides: more than a historian proprio sensu, he appeared rather as a writer in whose work it became interesting to emphasize the subjective aspects - theoretical, political, literary, oratorical - while historical understanding of human events took a secondary position or passed into the background; the same concept of historical understanding that inevitably implies the search for the truth tended to appear as too modern a category, and therefore not applicable to an ancient author.
Intuitively, every reader of Thucydides 1.22.1 who assumes a degree of creative liberty, whether greater or smaller, falls into this general interpretive frame. What is the origin of this? Is one dealing with a philological and extreme historicizing effort, which seeks to keep a good distance between ancient and modern, as well as to understand the former in its own features, far from every modernizing temptation? This is what definitions of Thucydides as ‘‘a human being and thinker eminently representative of his age’’ - a recurrent interpretation in the work of those who defend the non-positivist image of Thucydides (Hunter 1973: 5; cf., e. g., Loraux 1980: 70) - would lead us to believe. However, the same vision of Thucydides is that of someone who is aware of the influence that, in the interpretation of antiquity, contemporary culture exercises in all the variety of its components, from political concerns to new tendencies in literary criticism (Connor 1977; 1984: 3 ff.). Connor spoke of a ‘‘postmodern’’ Thucydides: and postmodern, paradoxically, is also the attitude of all those who attempt to circumscribe antiquity in a more or less absolute irreducibility to historiographical categories that we today are a part of. One characteristic of postmodernism, in fact, is the hypertrophy of the subject, incapable of an authentic dialogue with the object, and, in the best case, available to a distant listening: the hiatus that classicists today use to place between themselves and their object of study in the investigation of an absolute past, whereby every continuity recedes, is exactly one of the ways in which that distance manifests itself.
Thus, we have a philology of ‘‘alterity’’ that, even if unconsciously, feeds itself on the latest cultural stimuli. Therefore, one of the fundamentals that sustain this philological approach falls: the belief of being, thanks to objectivity itself, a privileged observer of the past. The results to which this philology leads are not readily picked from the texts: they are, rather, conditioned by an important part of modern reflection on historiography. The image of a Thucydides little interested in truth and much more interested in the rhetorical and didactic efficacy of his narrations owes much, it seems to me, to general theories of historiography such as those of Hayden White and Frank Ankersmit, defined as ‘‘constructivist’’ with respect to the alternative ‘‘documentary’’ model of history as investigation that, through proofs, makes a judgment on reality (on these two models see LaCapra 2001: 1-42). ‘‘Constructivism’’ celebrates the aesthetic, rhetorical, ideological, and political aspects of historiography, limiting referential assertions and the ‘‘truth claims’’ to the interstices of the discourse: in general the constructivist model admits the possibility of affirmations on reality, but only with regard to single events and not to structures such as plot and interpretations. It follows that historical discourse tends to contract to a textual dimension and to forget its natural opening to the world.
The Thucydides who, at least to a certain degree, invents speeches for aesthetic, rhetorical, didactic, or political ends is a ‘‘constructivist’’ Thucydides, rather than a writer seen within the appropriate framework of his time. The alternative image, that of the historian of facts and truths, was considered the result of an anachronistic projection of completely modern categories (‘‘retrojection’’: Bicknell 1990: 172). As I think I have demonstrated, however, the same could be said of Thucydides as ‘‘artful reporter.’’
In reality, what is decisive for an interpretation is not how much it owes to contemporary suggestions, but how much it justifies and illuminates its object. Let us take up again the subject matter of the speeches: how did Thucydides work to reconstruct them? We have examined the methodological statements and know what they provide for. But what was the concrete behavior of the historian? Let us admit that it is clear that in many cases the historian cut, synthesized, and condensed. The reports given to him were not always sufficient to reproduce the ‘‘entire reasoning’’ developed in the speech actually spoken: in such cases, other discourses by the same orator would have been useful for reconstructing an argumentative sequence.
The historian certainly would have avoided inserting passages or supplements that only resulted from his own imagination: that would have been quite incompatible with the interpretation proposed here for 1.22.1. What remains, the interpretive elaboration that aids in reconstructing with the highest probability the words actually spoken, is in line with the methodological principles of that passage. Between the historian’s creative liberty and interpretive ability there is a significant difference - that which Hegel indicated when he observed that the historian is not free to ‘‘subject to his own design the circumstances, the characters, and the events that are given’’: that is, to give a subjective unity to his material (Hegel 1842: III.260; cf. Koselleck 1985: 112[= 1979: 153]: ‘‘The [historical] sources provide control over what might not be stated’’).
Operations such as cuts, syntheses, and reductions are eminently interpretive acts. History, that of Thucydides just as that of the modern age, is judged for its openness to objectivity: this does not indicate that it should be considered a science or that our Thucydides wears the mantle of a positivist scientist. And this is for two reasons: first, a truism that bears repeating, Thucydides was not a positivist historian, but it was positivism that was Thucydidean. The second reason concerns the epistemological status of history which, even if it is an art, has never lost the characteristic of being a discourse on human reality: its peculiarity in the fifth century bce just as in our own times. Paul Veyne wrote that history is a work of art not ‘‘despite’’ but ‘‘for its efforts towards objectivity, in the same way in which a beautiful drawing made by a draftsman of ancient monuments, that shows the evidence without making it banal, is in some measure a work of art and presupposes that its creator possesses talent’’ (Veyne 1971: 272; italics original). Talent here is not a creative liberty without limits, but the ability to render the object per figuras. The talent of Thucydides - who knows if on this Veyne (cf. 1983, 1988) would agree or not? - was the talent of rendering per verba actions and speeches.
FURTHER READING
The literature on Thucydides’ speeches, and the so-called “methodological” chapter in particular, is vast, so much so that most recent contributions lean towards an extremely selective bibliography. An exhaustive examination of the studies published to about thirty years ago can be found in Luschnat 1970/1974: 1146-1183; 764-768; for later works, Stadter 1973; Nicolai 1992: 66-67 n. 61; and Porciani 1999 are rich in bibliographical references.
A useful introduction, with a collection of translated texts, is Harding 1973. The fundamentals of the twentieth-century discussion on the speeches (inventions or faithful recordings?) are found in Schwartz 1919: 23-27, 105, 131; 1926: 79-82; and Gomme 1937. According to Schwartz, Thucydides applies the rules of rhetoric, adopting them to single orators and, by having the protagonists of the story speak, explains the links between events: the logoi are therefore the intellectual property of the historian and belong in the sphere of ‘‘Fiktion.’’ Schwartz sees in the speeches the expression of a historical-political thinking which aims at explaining the connections among events (a view of Thucydides which clearly lies at a great distance from the modern, rhetorical-political picture of the ancient historian as a collector of exempla). Gomme by contrast insists on the historical accuracy of the Thucydidean rendering: it is not a matter of‘‘naive belief’’ (Badian 1992: 190), as Gomme’s analysis is very acute in emphasizing the influence of certain modern conventions, such as inverted commas, on
Our perception of the direct speeches in an ancient historiographical work. Gomme also clarifies the arbitrary nature of considering a summary in indirect discourse to be more objective. For attempts to prove the authenticity even of speeches traditionally held to be fictions, such as Pericles’ Funeral Oration, see Bosworth 2000b (cf. Porciani 2001a: 73-74 with n. 21), who demonstrates its appropriateness to the historical situation in which Thucydides places it.
World History
