Herodotus’ conspiracy narrative: time episodic, relative, and absolute

I’ve been meaning for a while to write a blog about the current episode in British political life (relatively awful or absolutely the pits?). This is not it.

In Book 3 of his Histories, Herodotus reports a crisis in the Persian empire. King Cambyses has had his brother Smerdis secretly murdered and then gone on an expedition against Egypt. In his absence, two brothers, one of them called Smerdis, from the Median priestly caste of Magi, usurp the throne, with Smerdis pretending to be Cambyses’ brother of the same name. Cambyses dies in Syria, after a reign of seven years and five months. After the imposters have reigned for seven months, a group of conspirators forms to overthrow them. As the conspirators move towards the palace, Herodotus interrupts the narrative to describe events taking place at the same time in the court. The usurpers try to persuade Prexaspes, a courtier who had been given the task of killing Cambyses’ brother, to proclaim openly his support for the usurper Smerdis. But things do not go according to their plan. Prexaspes, addressing the Persians from a tower, exposes the Magi’s deception to the crowd and then kills himself by throwing himself from the tower. This dramatic event happens just as the conspirators are reaching the palace. Herodotus then switches back to them: forcefully led by Darius, they gain entry to the palace and complete the job (3.69‒79).

The Behistun inscription narrates the rise to power of Darius I
The Behistun inscription provides another source for the story of Darius’ rise to power (Mount Behistun, Iran).

Herodotus’ account of this incident receives detailed treatment in Donald J. Wilcox’s 1987 monograph The Measures of Time Past – a work that has influenced a number of scholars of anachronism ancient and modern. Wilcox’s book is based on a distinction between ‘relative time’, which prevailed until Newton, and ‘absolute time’, which was uncovered by Newtonian physics. In relative time, events themselves create their own time-frame. In absolute time, there is a time-line that contains the events. Absolute time is ‘objective, continuous, all-embracing’; in Newton’s own words, it flows ‘equably without regard to anything external’. Wilcox fleshes out this model further by tracing a path from relative to absolute time through figures such as Augustine, Bede and Scaliger, the final step before Newton being the invention of the BC/AD system ‒ a system founded on ‘a single, continuous, and linear time frame’ which proved ‘fit for the use of those who accepted the notion of absolute time’.

Wilcox sees Herodotus’ account of the conspiracy as an example of ‘episodic time’ ‒ a form of relative time which was ‘discontinuous, emphasizing process rather than the progressive building of events on one another’, and in which ‘the exact temporal order was not an important factor in the process that produced the final result’. He describes the account as episodic because he finds it impossible that Herodotus had a grasp of the temporal order of the events: how could the conspirators have entered the palace after the death of Prexaspes? Herodotus, he concludes, ‘was willing to sacrifice neither story to the demands of a linear sequence and had available a sense of time which allowed him to keep both’: ‘the determining factor … was the moral and social context of the events’ ‒ in this case ‘the remoteness of the eastern kings’, ‘the arrogance and intrigue of the court’, and ‘the boldness and decisiveness’ of Darius.

Wilcox’s argument helpfully stresses the cultural contingency of systems of time-measurement: he allows that our own dating system ‘would have been as inaccurate in the eyes of Herodotus and Thucydides as their relative dating is primitive in ours’. But there are historical and conceptual problems in his arguments, and problems, too, in the way that he reads the ancient historians.

Let us first return to Herodotus’ conspiracy narrative. One problem with Wilcox’s reading is that he ignores the clear explanation in Herodotus’ narrative of how the conspirators were able to gain entry to the palace: it was their high status that won over the guards. Herodotus also suggests that they enjoy some form of divine assistance, revealed by an omen. Wilcox ignores, too, the fact that the conspirators realize that Prexaspes’ action endangers their plan and that the eunuchs in the palace, once they realize what has happened, threaten the guards with dire punishment. Herodotus highlights the very problem that Wilcox claims he ignores because of his sense of ‘episodic time’.

Wilcox’s reading of Herodotus’ account ignores the strong thematic links that bind the parallel narratives together. The most important theme relates to truth and lies. (As I said, this is not about Brexit.) Darius tells his fellow conspirators that he will offer a lie (fake news from the king) to ensure that they can pass into the palace; he suggests that people tell the truth or lie entirely in accordance with their self-interest. Prexaspes’ self-sacrifice in telling the truth at the cost of his own life offers at least a partial modification of Darius’ sophistic assertion. (Oxford has some towers, too…)

What of Wilcox’s general notion of relative and absolute time? The main historical problem lies in his account of BC/AD dating. He repeats the common contention that this ‘absolute’ system was invented by the French Jesuit Denis Pétau (Dionysius Petavius) in his 1627 Opus de doctrina temporum. In fact, datings before Christ’s birth are already attested (albeit rarely) in Bede’s Ecclesiastical History, written in the eighth century (Caesar crossed to Britain ‘in the sixtieth year before the incarnation of the Lord’). They become more common in the thirteenth century, and by the fifteenth they appear in graphic timelines (e.g. as the Carthusian Werner Rolevinck’s Fasciculus temporum (1474)). Prior to Petavius, Scaliger discussed the shortcomings of ‘ante Christum’ dating by contrast with his own Julian Period ‒ a period of 7980 years based on a 28-year solar cycle, a 19-year lunar cycle and a 15-year indiction cycle ‒ which provides a continuous timeline well adapted for astronomical calculations (and is indeed still used for this purpose). The shortcomings of BC dating were recognized by Petavius too: he explained that he included BC dates only for those who might be scared of the unfamiliar Julian Period, but he makes the Julian Period his main chronological anchor. It is true that BC dates were used by Newton in his posthumous Chronology of Ancient Kingdoms Amended (1728) ‒ but he also uses the Julian Period for an astronomical calculation. It was the nineteenth-century discovery of the age of the planet that rendered it (and rivals such as anno mundi dating) less useful (even though Scaliger conceived of the possibility of an infinity of earlier Julian Periods).

The conceptual problems in Wilcox’s model spring from his failure to distinguish between the measurement of time and time itself. With the ancient historians he looks at how time is presented in narratives; with Newton he explores ideas of time. And yet Newton’s time is in some ways not too far removed from some Greek conceptions of chronos ‒ which some classicists describe as an ‘absolute time’ by contrast with aiōn, ‘relative time’ (e.g. Collard on Euripides Suppliant Women 787‒8). Chronos is as universal as Newton’s ‘all-embracing’ time: ‘“all things” are regularly the target of time’s activity’ (Finglass on Sophocles Ajax 646‒7); it is ‘unwearying’, and ‘full in its ever-flowing stream it goes its round begetting itself’ (Critias TrGF 43 F 3.1‒3); it ‘has no father’ (Euripides fr. 303), but is itself the ‘ancient father of days’ (Euripides Suppliant Women 787‒8). In a comparative study of ancient Near Eastern cultures, Sacha Stern has indeed suggested that the ancient Greeks were unusually aware of the continuity of chronos as a separate entity.

Ideas about time, whether expressed through the imagery of a Greek poets or in Newton’s slightly more precise language, do not bear on the way time is measured. There is no such thing as an absolute chronology suitable for absolute time. All ways of measuring time are relative, and BC/AD dating is no more appropriate for Newtonian time than any other system of calculating years relative to a single point, including many of the systems used in Greco-Roman antiquity. If the term ‘absolute chronology’ is to be used at all, it should refer to a system which specifies the time in which events occur in relation to an external system (‘the first year of the twenty-third Olympiad’) rather than in relation to previous events (‘next year’).

Wilcox’s conception of absolute time seems as flawed, then, as his reading of Herodotus’ conspiracy narrative. And yet it is undeniable that Herodotus’ Histories do at times have an episodic quality ‒ and that occasionally his choices about where to reveal important information seem to us strange. Indeed, Wilcox’s analysis of ‘episodic time’ in Herodotus would have been better served by looking at the placement of other information that seems germane to the conspiracy and its aftermath. In his speech to the Persians before he jumps off the tower, Prexaspes retells the genealogy of the Achaemenids (the royal Persian line) as a way of asserting the correct order of rule in the empire. Much earlier in his work, Herodotus has told how Cyrus, the founder of the Persian empire, had a dream in which Darius’ rise to power was foreshadowed; and in recounting this dream Herodotus reveals that Darius himself is an Achaemenid (1.209). And yet Darius’ supposed Achaemenid descent plays absolutely no role in the conspiracy narrative ‒ even though it would have been highly relevant to Darius’ arguments for monarchy in the ensuing ‘constitutional debate’ (3.82). Darius is presented instead (as Wilcox notes) as a chancer who is last to join the conspiracy and whose ruthlessness and cynicism prove decisive in its success.

Wilcox does, then, capture something of the effect of reading Herodotus. But it is still questionable whether it is appropriate to speak of ‘episodic time’ ‒ as opposed to ‘episodic narrative’. Herodotus’ failure to integrate the whole of Darius’ past in a single story-line is not a cognitive clue to his sense of time but a sign, rather, of the extraordinary complexity and variety of the material that he was attempting to integrate and of the narrative artistry with which he nonetheless succeeded in shaping it.


S. Stern, Time and process in ancient Judaism (Oxford, 2003).
D. J. Wilcox, The measures of time past (Chicago, 1987).

For BC dating, see A.-D. von den Brincken, ‘Beobachtungen zum Aufkommen der retrospektiven Inkarnationsära’, Archiv für Diplomatik 25 (1979) 1‒20.

Author: Tim Rood

Researcher on ancient Greek historiography and its reception. Professor of Greek Literature, University of Oxford, and Dorothea Gray Fellow in Classics, St Hugh's College.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.