This is a good book, clearly presented, well argued, and concerned with a subject of importance for the understanding of Herodotus as well as the literary and historical milieu of his Histories. The author is well read in both Herodotus and the extensive relevant research and has a broad and detailed knowledge of Greek history and literature during the historian’s lifetime.
Referring to the work of Rosalind Thomas, Ian Oliver posits Herodotus as a sophos, a participant in fifth-century wisdom-culture in which oral performance was ubiquitous, and argues that before composing the text we know, Herodotus used to give lectures to live audiences, as is reported by, for example, Lucian. In the beginning of his work, the historian called his work an apodeixis; again, following Thomas, Oliver uses the term epideixeis for the preparatory lectures.
Oliver argues that specific passages of the Histories can be singled out as such epideixeis. His criteria for isolating them are narrative unity, clear delimitations in content or language, lack of internal cross-references, a spoken length suitable for oral performance, and an obvious historical audience. These passages he makes the object of his “audience-based analysis”.
After an introduction presenting the project, three battle narratives are analysed: the Battle of Plataea (Hdt. 8.133-9.70), the Battle of Salamis (Hdt. 8.1-8.96), and the Battle of Thermopylae (Hdt. 7.172-7.233). Each of them is treated in two chapters, one focused on the battle in question and one on what might be called the framing. At the end the book is summed up in a conclusion, followed by an appendix of potential other epideixis passages.
In his search for passages that seem to reflect the interests of specific, identifiable audiences, Oliver first concentrates on the Battle of Plataea. Other historians considered this battle to have been won first and foremost by the Spartans, and even Herodotus concludes his description by stating that Pausanias here “won the most glorious victory of all that we have seen” (Hdt. 9.64). Nevertheless, in his narrative of the battle, the Athenians are the uncontested heroes. Oliver retells Herodotus’s Plataea passage right from the preceding description of the Spartans and points out how, throughout this narrative, the Athenians act in a way that proves to be crucial for the success of the Greeks. He also spots situations that will have been understood symbolically by an Athenian audience, as when a Spartan officer throws a boulder into the air calling it a pebble for his vote not to flee (37). Thus, Herodotus’s Plataea narrative “prioritizes Athens’s contributions and marginalizes—even denigrates—Sparta and Pausanias” (38) in a way that contains an obvious appeal to an Athenian audience. Phocians and Macedonians are also treated favourably, and Oliver argues that an audience containing influential representatives of all three cities would suit the situation in the mid-fifth century rather than later periods. He imagines that Herodotus gave Plataea-epideixeis in Athens more than once and that the text would have developed with each telling. Herodotus will have observed how the audiences reacted, and members of the audience may have approached him after performances and related their own relevant stories (57). The two other battles, Salamis and Thermopylae, are analysed according to the same principles.
In the Histories the battle of Plataea is last of the three. Oliver begins with it because Herodotus differed from the standard narrative, which allows Oliver’s argumentation to come through most clearly. In the case of the battle of Salamis, it is more complicated for him to demonstrate conclusively that this narrative was tailored for Athenian audiences since, unlike the battle of Plataea, consensus already regarded Athens and Themistocles as decisive in the victory. A positive description might therefore have been accepted by most Greek audiences. Oliver shows how the narrative of the battle of Salamis is just as pro-Athenian as the Plataea one, and that the way praise and blame are distributed fits a mid-fifth century context. Athens and its allies are painted favourably while its enemies or rivals in the mid-fifth century such as Corinth, Thessaly, Thebes, and Sparta are negatively portrayed.
The description of the battle of Thermopylae is different since it does not point unmistakably to a specific audience. Athens is present, but so is Sparta, and Leonidas is the ultimate hero. Several other cities are also positively described, such as Thessaly, Phocis, and Boeotia, and even those that were compelled to fight for the Persians are positively characterised. Oliver concludes that wanting to perform a collective praise of the Greeks Herodotus must have looked around to identify an audience suitable for representing all Greek cities and found the Delphic Amphictyony. The venue where they gathered must have been the Pythian games.
In the Conclusion, Oliver sums up four main angles of his textual analysis: historical, ideological, cultural, and narratological.
It is exciting to follow Oliver’s efforts to find individual epideixeis and thus reveal the progress of the text from local oral performances to a unified apodeixis addressing a Panhellenic reading public. Following stylistic criteria Rosalind Thomas had already found that the geographic and ethnographic parts of Herodotus especially appear to have been made for performance, but also stated that “We cannot simply divide the text of the Histories into possible lectures”[1]. Nor does Oliver attempt that, but he challenges and develops her view by isolating clearly defined epideixeis in the historical part of the work. In the appendix he discusses whether it would be possible to single out similar passages elsewhere in the work, and is tempted to see Hdt. 2.2.1-2.99.1 as building on an epideixis. He stresses that his analysis of the three battles must not be understood as excluding the chance of finding other traces of the epideictic compositional method in the Histories.
There is much in Oliver’s approach with its focus on oral performance as a key for understanding the transmitted text that resembles the oralist trend in Homeric scholarship. The Herodotus scholar has the advantage that chronology and localisation is relatively well known. That is still disputed in the Homeric field, but during the last decades the tendency has been to move the epics closer to historic times, and so the chronological distance between Homer and Herodotus has diminished. With Rosalind Thomas’s insistence on the importance of orality during the archaic and classical periods, the socio-historical context of literature has become almost the same for the two.
When Oliver compares Herodotus’s narratives of the battles of Thermopylae and Salamis, his analysis of the historian’s methods reveals that they follow the same pattern down into details (145-46). He points specifically to how both battles begin with the Greek navy meeting at Artemisium. This is problematic for the Salamis-narrative because there the Artemisium event is told as if it were the first time the Greeks met there and realised how enormous the Persian army was. There is no reference to the fact that a similar story was told at the beginning of the Thermopylae epideixis (Hdt. 7.173-174 vs. 8. 1-4). Oliver’s explanation of this inconsistency is that the two passages have originally been composed for different audiences, an Athenian and an Amphictyonic respectively.
Patterned narrative and inconsistencies are, of course, terms well known from Homeric scholarship. You might point to the problem of how Achilles first turned down an embassy from Agamemnon pleading for reconciliation and two books later says, as if there had been no embassy: “Now I think that the Achaeans will come and beseech me, for their trouble is no longer bearable” (Il. 9.89-655 vs. 11.609-10). To the analysts such an inconsistency was a proof of multiple authorship of the poem,[2] whereas the oralists of today see it rather as a witness of a technique in which one poet handles multiple stories. Oliver’s Herodotus is one author addressing multiple audiences.
Concerning both Homer and Herodotus, we are left with the question of why the inconsistencies were not removed in the work’s the final version. Generally, Oliver does not seem interested in the details of the process of composition though he touches on it now and then in passing. He mentions that Herodotus must have anticipated the reception of an epideixis (82) and that he had intentions to perform to specific audiences (103, 130), thus imagining the situation when the historian was preparing his performances, even though the possibility of extemporaneous performance is kept open (131). Oliver also envisages the time when Herodotus wrote the epideictic material into the larger unitary work (77) and a point when he transitioned fully to his pre-written material (111). At some points during the compositional work Herodotus also thought of readers (109, 110), and he was concerned with the suitability for re-performance of his Histories (137-138). This last-mentioned point is part of a highly interesting comparison between Pindar and Herodotus concerning questions of writing and performance (135-138).
The present reviewer would have enjoyed a coherent and detailed description of how Oliver imagines Herodotus’s epideictic compositional method operating between written drafts and oral performances. However, it may, of course, be seen as an advantage of the book that it puts a thesis and limits itself strictly to defend it.
Notes
[1] R. Thomas, Herodotus in Context. Ethnography, science and the art of persuasion. Cambridge: Cambridge University Press 2000, 260.
[2] Forcefully argued by D.L. Page: History and the Homeric Iliad. Berkeley, Los Angeles: University of California Press 1959, 297-315, 324-334.