Bryn Mawr Classical Review

In this astonishingly ambitious book Carlo Scardino (S.) undertakes a close reading and comparison of the speeches in books seven through nine of Herodotus and books six and seven of Thucydides. His method is to describe each historian's speeches in narrative context, and on the basis of this description to analyze their functions and rhetorical structure. Finally, he compares his analysis of the speeches in Herodotus to his analysis of the speeches in Thucydides. S.'s determination to maintain continuous contact with the narrative context of the speeches determines the character of his descriptions of the texts, which resemble commentaries. Hardly enough good things could be said about these commentaries and the resulting analyses: over hundreds of pages S. balances perceptive reading with respect for each historian's particular choices and careful reflection on copiously cited scholarship. The outcome is that S. establishes a description of the historians' relationship to one another that is unprecedented in its comprehensiveness.

The introduction begins with a comparison between modern and ancient historiography. This comparison frames the work: S. returns to it at the very end (769-790). For the introduction S. reminds us that modern historical writing rejects the ancient format, in which the historian composes speeches for his characters and historical events are focalized through perspectives other than the narrator's, which is only rarely explicit (1-2). He provides a review of ancient criticism of the speeches which shows that in antiquity the speeches in historical works were taken for granted, and were analyzed for their rhetorical qualities, not their historiographical values (3-6). A review of the modern scholarship, on the other hand, shows our developing appreciation for both the historiographical and artistic values of the historians' speeches (6-26). No overarching analysis and comparison of the speeches had been attempted, however, until S. undertook his monumental work.

The second section of the introduction is devoted to Aristotle and Homer. To begin, S. outlines the view of history taken in Aristotle's Poetics, and establishes his differences from this view: S. will show that Herodotus and Thucydides resemble Aristotle's poets as much as his historians (34-35, cf. 58-59). Next, S. provides an outline of narratological method, justifying his use of this method by reference to the fact that the historians composed works characterized by techniques learned from fictional writing (36-46). To found this description of the historians S. delineates Homer's central importance for their speeches. The three main functions of the speeches, namely: to dramatize the events, to characterize the speaker (who is an historical agent) according to type, and to interpret and comment on events at the level of the fabula, originated in Homer and were adopted by the historians. Given this fundamental similarity between the two historians, S. sets out to discover what the differences between them may be, and what conceptual and rhetorical bases determine those differences (59).

S. begins his examination of Herodotus with a review of the first sentence of the Histories that isolates Herodotus' main interests (the spatium humanum, and particularly the great deeds performed in this arena, the analysis of conflict and of the responsibility for the war, 64), and continues by describing the corresponding structure of Herodotean narrative: "The structure of the narrative is in its foundation always the same: after an often modest beginning and successful growth the insatiable impulse to conquest brings it about that the (oriental) monarch at the height of his power attacks a comparatively poor but defensible people, and loses" (66).

Over the next ten pages (67-77) S. provides a chart that records the plot and shows the placement and main themes of all direct speeches in the text. The chart demonstrates the unity of Herodotus (80), and constitutes a defense of the historian; in the subsequent sections S. describes and defends Herodotus' historiographical method, including his ordering of material (82-83) and use of evidence, despite its well-known problems, as essentially "serious, reliable, and non-partisan" (86).

Finally, S. addresses the question of human and divine causation in Herodotus. S. argues that the divine constitutes "the strongest determinant" in Herodotus (101), but that parallel causes remain very important. In particular, and consistent with his explanations of the structure of the Histories, he argues that in Herodotus "since the human pursuit of power and property is hardly possible to avoid, a necessary process of history results (103)." The injustices that arise from this pursuit are corrected in the human, or if not, in the divine sphere (104-105). Regardless of whether divine or human forces act as a corrective, the process of justice in the world can be verified through empirical observation, i.e. ἱστορία (102).

This description of Herodotus' particular themes and world view is followed by discussions of more formal concerns: general characteristics of Herodotus' style (109-110), the complexity of the Herodotean narrator, and the role of speeches in indirect discourse (111-116) are described before S. undertakes to characterize the direct speeches.

It is actually surprising how little of Herodotus is devoted to speeches: about 18% of the entire Herodotean text is comprised of direct speeches, the portion of the text devoted to speeches being greatest in the final books, where it is closer to 25% (117). (By contrast, about 50% of the Iliad is composed of direct speeches, cf. 47.) The general character of these speeches will be familiar to Herodotus' readers. Speakers are not distinguished from one another through their style, but the style of the speeches is in general more elaborate than that of the narrative (118). Speeches occur in pairs and larger groupings (119-120), and are without exception the product of free invention (124).

By contrast to the introduction we have just reviewed, in which very large topics were introduced and summarized, the description of Herodotus' last three books (126-318) is detailed and specific. Descriptions of clauses and sentences construct the descriptions of the speeches and groups of speeches, each of which is treated with a short interim analysis as S. moves through the selected material. Narratological terminology is almost completely absent: the description relies on ancient rhetorical terms for its categories. The sound of phrases and sentences is frequently noted.

The analysis follows the description. Drawing together all of his examples, S. argues that his initial characterization of the complexity of the Herodotean narrator (326-331) and the role of indirect discourse in Herodotus (332-335) were justified, and shows that Herodotus' speeches fulfill the three Homeric functions (dramatization of events, characterization of speakers, and development of motifs that enable interpretation of the text). Of these the first two seem straightforward, and the last is most important for the comparison to Thucydides. In his analysis S. argues that "the direct speeches reveal to the recipient a polyphonic network of suggested meanings and frameworks for interpretation" (350). This network is constructed, in his view, through the use of recurring motifs that "transcend the specific situation" (350). These motifs sometimes correspond to words (νόμος, for instance), and sometimes to themes, such as envy, the justice of the gods, glory and honor, etc. that are referenced with a variety of words. They also correspond to rhetorical practices, such as the practice of offering interpretations of the past that intend to instruct us for the future, or of referring to difficulties of communication. Key to reader reception of these motifs is their repetition by different speakers in different situations, since this problematizes a simple or one-sided understanding and in this way draws us into the interpretation of events. In the closing section of his analysis S. explains a large variety of such motifs and demonstrates their development in the speeches he has discussed. For a list of motifs S. isolates as common to Herodotus and Thucydides, see footnote one.

The dumbfounded reader now turns to what is essentially another entire book, since S. gives Thucydides the same full treatment as he shows to Herodotus. S. begins his introduction to Thucydides by describing the Archaeology and its aims (to show the growth of power, 384). He pauses on Thucydides' statement of the causes of the war (Spartan fear, 385), and then offers a chart of the whole text, marking all direct speeches (387-394). After this S. reviews the debate of the analysts and unitarians (394-399; conclusion: we cannot tell where and how much Thucydides revised his work, 399), and provides an overview of Thucydides' historical method and use of evidence. On pages 402-416, S. discusses chapter 22, book one of Thucydides, which contains both Thucydides' statement of his divergent treatment of speeches and the narrative of events and also his statement of the utility of his work. Scholarly support is massive throughout, and S. maintains moderate views, while giving his opponents frequent opportunities to speak for themselves.

S. next discusses Thucydides' philosophy of history, and main theme ("The question of the nature and management of power, in particular its manifestation in Athenian "imperialism"...whereby the war gradually dissolves the ethical categories valid in peacetime", 418). His remarks on the centrality of the unchanging human condition for Thucydides' writing, in which the "reduction of reality to rationally intelligible patterns" leads to "stripping events of their temporal particularity"("eine Entzeitlichung des Geschehens" 425), argue that Thucydides' diagnoses, like medical diagnoses, apply to all potential readers. The ensuing argument takes note of Thucydides' didactic aims (426), and introduces some main teachings on topics such as the helplessness of hope and power of chance, rationality, military virtue, crowd psychology, and political leadership (432-440).

S. then addresses Thucydides' style and the structure of his narrative, discussing Thucydides' advertised adoption of chronological order, and the structural relation between the speeches and the narrative: in fact, thematic and didactic priorities determine the rhythm of the presentation (441-444). S. shows that Thucydides' narrator disappears to an extent not as evident in Herodotus, and addresses narrator interventions, internal focalizers, and speeches in indirect discourse (444-449) before providing a formal description of Thucydides' style (450-452).

This introduction accomplished, S. can turn to the speeches themselves, as he did for Herodotus. Twenty to twenty-five percent of Thucydides' text is composed of direct speeches (453). After discussing the speech frames (i.e. the narrator's introductory and closing remarks to direct speeches, 454-455), and the prominence of speech pairs in Thucydides (as opposed to the looser, more conversational organizations favored by Herodotus 456-457), S. broaches a description of the functions of Thucydides' speeches: the dramatizing and characterizing functions remain similar to those in Homer and Herodotus. An intensification of the third function is visible, however: "Most important is the function [of the direct speeches] as an unmitigated dramatic, analytic, and implicitly interpretive commentary in the place of explicit authorial reflection and interpretation, whereby the opinions expressed by the speaker do not agree with those of the historian, but rather must be brought into relation with narrative connections and authorial statements" (461).

The description of Thucydides' speeches follows (464-648). Once again, we enter the world of specificity; S. is perhaps an even more perceptive reader of Thucydides than of Herodotus: particularly impressive to this reviewer was a short essay on the Melian dialogue (468-483). The subsequent analysis of Thucydides' speeches (648-701) bases itself on the by now familiar Homeric triad of speech functions, clearly laying its weight on the last, or "commentary" function. Just as he did for Herodotus, S. offers a detailed presentation of the commentary function of Thucydides' speeches, reviewing a large number of motifs. The Thucydidean motifs that create "a toolbox for interpreting the events recounted by the narrator" (701) are largely the same motifs as were found in Herodotus' speeches, although important differences do emerge. The most important difference, perhaps, is a conspicuous enlargement in the number of motifs created from abstract terms (675, cf. 685). Again, see footnote one for a list of the motifs S. ultimately draws in to his comparison of the two historians.

The suggestion that Herodotus and Thucydides have such a similar cast of mind will be controversial for many readers. The comparison between the two (702-744) begins with a table showing the similarities between Herodotus' narration of Xerxes' expedition to Greece and Thucydides' narration of the Athenians' expedition to Sicily (702-705). This comparison is not new (cf. 707-709), but is here differentiated, detailed, and comprehensive. Though the parallels between the two narratives are clear, S. does not argue that Thucydides is imitating Herodotus. "The parallels speak less for direct dependence, above all when one takes account of the differences, than for a common intellectual foundation..." (710). He is also not dogmatic in his view that the similarities he has described are more important than the differences he can describe equally well. On the following pages S. clearly and persuasively describes the structural differences between the two narratives (710-711), and then goes on to describe other basic contrasts between the two authors. For Thucydides the human being according to his φύσις is the "exclusive subject and object of history" (713): this causes the fundamental difference to Herodotus. S. also discusses differences between the historians' narrators, their presentation of battle narratives, the status of religion and suffering in their narratives, and their divergent conceptions of the causes of military/political victory and defeat, for example, capping these with a discussion of indirect discourse in the two historians (716-717).

For the direct speeches, however, he argues that the similarities between the historians outweigh their differences (717, cf. 743). The functions of the speeches do show some differences: the dramatizing function of the speeches is less important in Thucydides than in Herodotus (717), whereas characterization of the speaker through characteristic speech (ethopoiesis) is more Thucydidean (718). S. pays most attention to a comparison of the themes ("motifs") of the speeches (723-44), however. Here individual differences and commonalities emerge; the main finding, however, is that Herodotus and Thucydides share many important themes, which they develop in the direct speeches.¹ Furthermore the direct speeches have a common and centrally important function in both historians: the direct speeches "augment authorial commentary and the thoughts expressed in the indirect speeches with detailed argumentation , [that is] the production and discussion of the categories relevant to the actors (most prominently in reference to the topical τελικὰ κεφάλαια), that breaks through the level of the particular and attains the general thanks to the use of gnomai, definitions, and through ana- and prolepses with which they [the speeches] widen the spectrum of possible interpretation..." (752).²

For S. the speeches thus serve a foundational role in enabling interpretation of the text, and not only this, but also in some way define historical writing against other available paths: The speeches "show the restrictions and possibilities of human action and transcend the unique historical situation, but at the same time [and] by contrast to the speculations of contemporary sophists and the fictional world of poetry, guarantee the knowledge they lift to the level of the general with a relation to historically verifiable reality" (752).

Needless to say, S.'s hypothesis about the character and function of thematic development in the speeches will be discussed for some time to come, as S. also suggests it should be (753). To the reader's final amazement, S. finishes his magnum opus with a description of Polybius' treatment of speeches and narratorial authority, which contrasts to that of the classical historians, and with his closing essay on modern historiography, in which he compares the goals and audience of ancient and modern historians. The book is thus a complete and closed form.

Three remarks. First, a small point: Unless I somehow missed this, S.'s many discussions of indirect speech in the historians neglect to mention that any speech expressed in indirect discourse falls under the narrator's direct control (i.e. speeches in indirect discourse record the narrator's selection from the things the speaker said). This control is sometimes made explicit through narratorial interventions, and more frequently emphasized through verbs of speaking and grammatical structures that demonstrate that the speech is being reported. S. remarks for both historians that it is difficult to tell why a speech might be set in indirect, rather than direct discourse (e.g. 334, 659, 717); perhaps a consideration of which speeches the author wishes to subordinate to the narrator's control could be a useful additional aid for considering this question.

Second (another small point), S. argues that in Herodotus no character appears who understands both the compulsions of power and his own desires, and who knows how to control the passions of the people, whereas in Thucydides, Pericles achieves this, by contrast to the leaders who follow him (749-750). This is a difficulty: Herodotus has Solon (cf. e.g. Rengakos 2006), and Thucydides' Pericles is definitely not immune to the allure of power and glory (cf. e.g. Samons 2007).³

Finally, a tentative suggestion about the main point of the book. Scardino has done yeoman's labor in providing a detailed, differentiated, and comprehensive comparison of the two historians in a breadth never before attempted. His book is a watershed publication that will at a single stroke improve the quality of our comparisons of the two historians. His attention to the motifs of the speeches provides much (but definitely not all, as the basic line of comparison to the Homeric paradigm does not rely on this) of his proof for the similarity between the two authors. Clearly, one has a range of such motifs to choose from, and the motifs one chooses to discuss will determine the similarities and differences one describes. S.'s choices struck me as absolutely accurate, but also as representing a perspective on the speeches that privileges abstract and therefore shared motifs. As for abstraction, I missed basic motifs such as polis, παρασκευή, or even navy or wealth. S. frequently mentions these concepts, but they do not rise to the dignity of motifs: yet they are important for constructing precisely the relationship to reality he argues separates poetry and philosophical speculation from historical writing. In my view, therefore, the final analysis of the speeches as Scardino has so responsibly produced it for us represents them as somewhat more abstract than even Thucydides' speeches really are. This could be a problem for readers who do not take into account S.'s entire argument (i.e. description as well as analysis), since abstract concepts in particular may appear to be related when they are considered in general, whereas in context the arguments they undergird may not be as similar as the connections of the extracted concepts suggest. This observation is not to detract from anything S. does say, but to add a perspective and a suggestion.

Notes:

1. Common motifs of speeches in Herodotus and Thucydides: I am providing these motifs untranslated for fear of misrepresenting some concept and its relation to the pertinent Greek words.

Inhaltliche Motive und Argumente der Reden

1. Abstrakte handlungsmotivierende und exegetische Kategorien (e.g. Herodotus: νόμος, ἐλευθερίη, δουλοσύνη; Thucydides: ἀνθρωπεία φύσις, ἀρχή, ἀνάγκη, ἐλευθερία, δουλεία) 2. Gerechte Vergeltung und Verwandtschaft als offizieller Grund (δίκαιον) 3. Die ethnische Zugehoerigkeit und der Volkscharakter 4. Ruhm und Ehre (als καλόν beziehungsweise αἰσχρόν) 5. Der Nutzen (συμφερόν), die Leichtigkeit (ῥᾴδιον) und die Machbarkeit (δυνατόν) 6. Das Thema der Sicherheit (δέλος, ἀσφάλεια) und des Praeventivkriegs 7. Die Tapferkeit und die Kriegskunst 8. Die irrationalen Affekte 9. Die Interpretation der Vergangenheit als Erfahrungsschatz fuer die Zukunft 10. Der Wert der Wohlberatenheit (εὐβουλία) und rationale Elemente 11. Das Problem der Kommunikation 12. Die Uneinigkeit und Zerstrittenheit der Parteien 13. Ueberlegungen zur Verfassung 14. Das Thema der Jugend (ἡ νεότης) 15. Argumentum a persona

2. (The τελικὰ κεφάλαια referred to above are words that directly provide motifs for a speech: e.g. δίκαιον at Thucydides 1.32.)

3. References: Antonios Rengakos, "Thucydides' Narrative: The Epic and Herodotean Heritage," in Brill's Companion to Thucydides, Brill, Leiden and Boston, 2006, and L. J. Samons, "Pericles and Athens," in The Cambridge Companion to the Age of Pericles, Cambridge University Press, New York, 2007.