BMCR 2020.11.24

Herodotus and the question why

, Herodotus and the question why. The Fordyce W. Mitchel memorial lecture series. Austin: The University of Texas Press, 2019. Pp. xv, 360. ISBN 9781477318324 $55.00.


Herodotus’ keen interest in investigating historical explanations or causes is already evident in the Histories’ first sentence, which ends by asking ‘for what reason they [i.e. Greeks and Persians] fought with one another’. While this interest on Herodotus’ part has long been noted by scholars, Pelling’s book is the first monographic study of the topic.[1] In Pelling’s view, asking the question why, rather than merely the question what, in relation to past events is fundamental to Herodotus’ task as a historian.[2] Pelling does an outstanding job of analyzing all the varied rhetorical and thematic strategies Herodotus uses in his efforts to answer this question.

The book is divided into sixteen chapters. In Chapter 1 Pelling surveys the different ways in which Herodotus thinks about historical causes. The subject of Chapter 2 is the assignation of blame (aitiē) by Herodotus or his characters (such as in the abductions of women, 1.1-5) and the related theme of “payback” (tisis), whether with Candaules (1.8-12) or with Xerxes (9.108-113). Chapters 3-6 treat the intellectual and rhetorical background to Herodotus’ methods of explanation, especially as found in contemporary scientific and medical writers (Chapters 3, 5, and 6) and such writers as logographers (Chapter 4), from Hecataeus to Gorgias.[3] For example (Chapter 6), Herodotus can accumulate explanations, without preferring one over the other (as he does with the factors that drove Cyrus to continued conquests, 1.204.2), no less than can the Hippocratic authors of Airs, Waters, Places or of On Ancient Medicine. In Chapter 7 Pelling argues that with the stories of Candaules and of Croesus in Book 1, Herodotus trains readers to expect that some explanations are provisional and must be reevaluated over time (as Apollo himself demonstrates, when he points out the reasons why Croesus’ empire fell, 1.91). All the rest of the chapters except the last one focus on individual themes around which Herodotus seeks out explanations. Chapter 8 is on empire: the reasons why empires like Croesus’ rise and fall or imperial enterprises like Xerxes’ invasion of Greece succeed or fail usually depend on both the shrewdness and the folly of aggressor and of would-be victim alike. In Chapter 9 Pelling finds that Herodotus shapes his “Persian stories” to reflect certain thematic patterns: the influence of royal women, the difficulty of speaking truth to power, the imperialist led to ruin by overreach and by the gods. Although Herodotus privileges the human element as the determining factor in how historical events turn out, Chapter 10 shows that sometimes the divine element is inescapable, whether it is Xerxes’ dream figure commanding him to invade Greece (7.12-18) or the mountain peaks that tumble toward the Persians at Delphi (8.37.3, 39.2). Chapter 11 shifts from Persian defeat to Greek victory: the Greeks won at Marathon and against Xerxes’ forces for reasons both commendable (superior military equipment and training, 9.63.2) and not (Greek disorder at Plataea paradoxically leading Mardonius and his army to attack without good order, 9.59). Other reasons for the Greek victory are freedom (Chapter 12) and democracy (Chapter 13). Herodotus praises those who fought to preserve Greek freedom from the threat of Persian autocracy (7.145.1) and acknowledges that Athens only started to grow great once it freed itself from its tyrants (5.78). Democracy, says Pelling, is freedom taken to an extreme: the Athenian dēmos was free to make decisions that were providentially good (building two hundred ships at Themistocles’ instigation, 7.144) and bad (sending ships to help the Ionian Revolt, 5.97). Chapter 14 deals with the interplay between individuals and collectives: among Persians, individual success is ultimately the purview of the king alone, but even he must operate in the shadow cast by previous kings; among Greeks, the Spartan Pausanias is as free to achieve individual success (and prospective failure) as is the Athenian Themistocles. Herodotus’ allusions to events later than the end (479 BCE) of the main narrative, including the downfalls of Pausanias and of Themistocles, lead Pelling in Chapter 15 to consider how Herodotus, along with his first audiences in particular, may have sought connections between and even explanations for contemporary events (above all, the Peloponnesian War) in the past events narrated in his work. Chapter 16 is a brief epilogue, in which Pelling stresses the explanatory power that Persian imperialism (as described by Herodotus) may have held for Greeks living at the time of an imperialist Athens.

There is much that is excellent in Pelling’s book. It is comprehensive in scope, treating an impressive number of Herodotean passages. At the same time it is well-structured and argued, with each chapter flowing easily into the next. Pelling’s discussion (Chapters 3-6) of Herodotus’ explanatory modes vis-à-vis that of other prose authors stands out for its thoroughness. He provides a useful annotated list (Chapter 15) of Herodotus’ allusions to later events – external analepseis in narratological terms. Perhaps the best part of the book are the many overarching statements Pelling makes about the Histories. A few of these literary critical gems are: “As so often, the richest explanatory passages explain more than the single context in which they figure” (43); “But it was not coincidence that Herodotus happened now; nor was he writing for an audience that would find his thinking too alien to accommodate” (79); “For Herodotus knows that choices can be overdetermined” (103); “Herodotus has more than one approach to multiple explanations” (269 n. 25); “This is a two-way street: past illuminates present, present illuminates past” (215); “Herodotus too will puzzle, and expect his readers and hearers to puzzle too, over the complex events he describes, and he allows us to share the different ways of looking at them and making sense of them” (162); “And Herodotus surely does expect a very active reader, one who is given plenty to work on and plenty to puzzle at” (11).

While Pelling himself seems to expect “a very active reader” for his own book, this can at times cause problems. Indeed, there are some statements Pelling needs to explain more fully. While discussing the mercurial nature of the Athenian dēmos (“swift to decide and swift to change their mind”, 192), Pelling refers to “the Athenians’ swift recrimination of their envoys for offering Persia earth and water (5.73.3; cf. 5.96.2).” Herodotus says that the Athenians had sent their envoys to meet with the satrap of Sardis in ca. 507 BCE and to form an “alliance” (symmachiē, 5.73.1) with the Persians. Pelling implies that the Athenians are upset with their envoys for doing exactly what they had been sent to do (i.e. form an alliance). But it could be instead that the Athenians belatedly realize that the envoys, by offering earth and water to the King, had subjected, not allied, Athens to Persia; this would have made the Athenians rebels in the eyes of the King when they later participated in the Ionian Revolt.[4] Elsewhere, Pelling rightly observes that the idea that Persian subjects were tantamount to slaves is “a Greek conception of the way Persians thought, and real Persian ideology was more nuanced” (286 n. 10). On the other hand, Pelling’s assertion that among the Persians “[t]he degree of slavery doubtless varied greatly even at a much lower level” (176) is not very helpful, and his claim that “‘vassalage’ is a better term” (286 n. 10) for this supposed slavery is also untenable because the Persian “slavery” in question was broader than the concept of vassalage suggests. Rather, in the context of Greek (mis)conceptions of Persian slavery mention should be made of the Old Persian term bandaka (literally “bondsman”), a word which ranged in meaning from whole subject peoples (e.g. Darius’ Bisitun inscription [DB], paragraph 7) to individual Persian aristocrats who had ties of loyalty to their superiors (much like vassals to a feudal lord), particularly to the Persian king.

Errors in the book are rare. Fortunately, the typographical errors tend not to mislead; they occur mostly near the end of the book or in the bibliography.[5] The three outright factual errors are: “Aristagoras arrives in Athens” not in 494 (191-2), but probably in either 500 or 499; Cyrus is recognized by Astyages not “at the age of fourteen” (129), but at the age of ten (Hdt. 1.114.1), as Pelling correctly says later (151: “ten-year-old Cyrus”); it is not Harpagus who is ‘“stricken by the gods”’ (152, referring to Hdt. 1.127.2), but Astyages, as Pelling correctly notes earlier on the same page (152).

Pelling has produced a book that has a definite Herodotean feel to it: sweeping, rich, and thought-provoking. Readers will come away with renewed respect for Herodotus’ overall historiographic achievement and especially for the dogged nature of Herodotus’ attempts to get at the reason why things happened.


[1] See esp. H. R. Immerwahr, “Aspects of Historical Causation in Herodotus,” TAPA 87 (1956): 241-80 (reprinted in R. V. Munson, ed., Herodotus: Volume 1: Herodotus and the Narrative of the Past (Oxford 2013) 157-93).

[2] In a self-professed companion piece to the current book, Pelling examines the role that Homeric epic played in establishing the roots of historical explanation for later Greek historians: C. Pelling, “Homer and the Question Why,” in C. Constantakopoulou and M. Fragoulaki, eds., Shaping Memory in Ancient Greece: Poetry, Historiography, and Epigraphy (Newcastle upon Tyne: Histos Supplement 11, 2020), 1-35.

[3] In these chapters, Pelling builds upon Thomas’ work to ground Herodotus in the thought world of the Hippocratics: R. Thomas, Herodotus in Context: Ethnography, Science and the Art of Persuasion (Cambridge 2000).

[4] See M. Waters, Ancient Persia: A Concise History of the Achaemenid Empire, 550-330 BCE (Cambridge 2014) 84.

[5] Potentially misleading errors include: “euteteōs” for eupeteōs (205); “160.3” for 1.60.3 (205); “above, n. 000” (285 n. 4) Dewald 1993 is not “repr. Munson 2013” (306); “Xeus” for Zeus (326).