5.2: Writing Space, Living Space

Time, Agency and Place Relations in Herodotus’s Histories [1] 

Elton Barker, Stefan Bouzarovski, Chris Pelling, Leif Isaksen

Aristagoras’ Engraving and Herodotus’ Narrative

In the fifth book of his Histories, Herodotus describes the arrival in Sparta of one Aristagoras, tyrant of Miletus, there to win support for a revolt of the Ionian Greeks from Persian control [2]. Aristagoras enters into discussions with Cleomenes, one of the Spartan kings, armed with ‘a bronze tablet on which the way around the whole earth was engraved, and all the sea and all the rivers’ [3]. With this visual aid, Aristagoras points out the string of places to be conquered until he arrives at Susa, ‘where the great king lives and where the storehouses of his wealth are located. His promises are persuasive enough to gain another hearing; it is only three days later when Cleomenes asks how many days’ journey it was from the Ionian Sea to the king (50.1), that his mission fails. For Aristagoras makes the mistake of telling the truth – the journey from the sea was some three months (50.2) – at which news Cleomenes bids his Milesian guest leave Sparta before sunset (50.3).

Coming at a critical juncture in the narrative, as Herodotus bridges the transition from mapping out Persian power to detailing the first moment when Greeks and Persians clash, this episode brings to light a number of important issues for thinking about Herodotus’s representation of space [4]. For the purposes of this volume, we focus on the tensions between Aristagoras’s cartographic visualization and Herodotus’s discursive space – that is, how spatial ideas and concepts are put into words. Aristagoras’s bronze engraving attempts to capture the topography of the world – its ways, sea and rivers – in a form that is at once complete, abstract and fixed [5]. Indeed, Aristagoras puts these very qualities to use. Using deictics throughout his pitch to Cleomenes – ‘next to the Ionians here are the Lydians’ (5.49.5); ‘next are the Phrygians here’ (5); the Cilians ‘possess land that reaches this sea here’ (6), etc. – Aristagoras exploits the world in miniature to collapse the distance and differences between places and make the effort of traversing them seem easy [6]. Yet, this strategy and these qualities of completeness, abstraction, and fixity ultimately flounder when confronted by the ‘reality’ (50.2) of Cleomenes’s question, ‘just how far is it from the Ionian Sea to the King (i.e. Susa)?’

Where Aristagoras had concNentrated exclusively on general spatial properties like water bodies, territories and proximity to emphasise links and movement, Cleomenes introduces the concept of time that disrupts the flow – how long does it take? In other words, we are invited to consider what this space signifies.

Herodotus’s narrative neatly underlines the difference between space as abstractly represented and what it means to those on the ground. First, Herodotus records the delay in time (of two days) from Aristagoras’s initial appeal to Cleomenes’s answer, as if the very act of interpreting the engraving were resisting the attempt to capture the whole world all in one go. Second, Cleomenes’s decision is itself marked by shifts in location – initially ‘to the place that had been agreed’ (50.1), then to his house (51.1), finally ‘to another room’ (51.3) – as the engraving’s persuasive power is finally undone by the kind of detailed contextual mapping that its generalised representation glosses over and where each place is invested with a particular meaning [7]. Lastly, Herodotus supplies his own discursive representation of the equivalent space. Ostensibly going over the same ground in order to bear out the accuracy of Aristagoras’s (foolishly) true answer of thirty days, Herodotus rewrites Aristagoras’s bird’s-eye visual display as hodological – that is, from the perspective of someone making the journey [8]. And far from the ‘ease’ that had characterised Aristagoras’s abstract modeling, Herodotus’s narration emphasises complexity, the labor required to traverse the space, even the difficulty of apprehending it, an effort which takes him over two chapters to document (5.52-54). Where Aristagoras’s engraving presents an abstract product of contemplation, Herodotus’s text represents the idea of space as something lived [9]. We see this not only in the fact and manner of Cleomenes’s answer, which exposes the meaningless of a generalised topographical model if one cannot tell how far away places and peoples are from each other. It is also represented in and to a certain extent performed by Herodotus’s subsequent narrative, which fills the space with measurements to take, obstacles to overcome, and a sense of what it is like to travel from the Aegean Sea to Susa.

There are at least three important aspects to the idea of lived space that contributes to Herodotus’s discursive representation. The first of these, time, we have already mentioned: whether it is the thirty days’ journey or the simple chronological movement of Herodotus’s narrative, time provides the context in which space is experienced [10]. Herodotus explicitly confronts the issue of temporality in his prologue, when he describes his inquiry being based on him ‘seeking out towns of men both small and great alike: for of the places that were once great, most have now become small, while those that were great in my time were small before’ (1.5.3-4). A few paragraphs earlier, some unnamed Persians present a world that is sharply differentiated along ethnic lines between Greeks and barbarians, with the Greeks holding sway over Europe, the Persians Asia (1.4.4). By contrast, Herodotus suggests a different, more complex, path through the places of history, differentiated not so much by their spatial location as by their temporal dimension – how they do not remain the same over time [11]. Agency, which is already implied by our description of ‘lived space’, is important here too – whether we think of the Persians presenting their account of the history of the conflict with the Greeks, or Herodotus taking an active role in hunting out historical data from the places to which he travels. And agency permeates the Histories’ space, not only in the actions, thoughts, and accounts of the historical participants or of Herodotus himself, but even in non-human objects such as Aristagoras’s engraving, which attempts to present the world in a way that highlights movement (ways, rivers, seas), or the pillars by which Darius seeks to map out his dominion over the Black Sea territory (4.87), or the cult statues which the Aeginetans send as assistance to the Thebans – and which are sent back with a polite request for men instead (5.81). Thirdly, lived space is relational. For Aristagoras, the chain of places that he sketches out emphasise a linked world with clear and direct lines of communication, which in turn casts this space as attractive for a would-be conqueror. At the same time, however, the very fact that the man from Miletus is in Sparta seeking an alliance against the Persians shows the interconnectedness of the world, but in a different way, one that connects places from across the waters, which are not topographically proximate to each other (as Cleomenes’s question makes clear), but which are somehow conceptually and culturally linked [12]. Different relational models are available and differ according to the context, purpose, and persons involved.

For the rest of this chapter, we want to think more carefully about the Histories’ lived space and explore how the picture that emerges from the narrative differs from abstract depictions of space presented within it. Such overly schematic representations we see articulated by the Persians at the very beginning of the Histories, or explicitly challenged by Herodotus when he ‘laughs at’ the maps produced by his Ionian contemporaries that similarly divide the world into two regions of equal size (4.36.2), or more subtly undercut when Aristagoras turns up with just such a map and puts it to service an argument in favour of conquest [13]. In particular, we want to challenge conventional readings of the Histories as presenting a polarised world of East versus West, which, while grounded in Herodotus’s concern to explore how ‘Greeks and barbarians came into conflict with each other’ (1.1), fail to take into account either Herodotus’s implicit rejection of the Persian model of an Asia-Europe divide in favour of an inquiry that recognises how places change over time or the extent to which Herodotus or his historical agents relate those places to each other [14]. Using the features of lived space-time, agency, and relation – that have emerged from the scene between Aristagoras and Cleomenes, we sketch out the beginnings of a network analysis of Book 5, which in turn helps pave the way for a close textual study of that book’s opening episode. Both methods help to give form to the idea of the Histories’ lived space that underpins and greatly complicates the historical agents’ own understanding of the world around them.

Towards a Network Analysis of Herodotus Book 5

Even a cursory reading of the Histories reveals the wide array of different places Herodotus hunts out in the course of trying to explain the conflict between Greeks and barbarians [15]. Because of their sheer volume, in order to make some inroads into understanding how space functions in the Histories requires a close analysis of one stretch of narrative. 

Book 5 offers a good example, since, as well as providing the broader context for the meeting of Aristagoras and Cleomenes, it stands at the centre of Herodotus’s Histories and marks the pivotal moment when Asia and Europe – the Persians and the Greeks – come into conflict for the first time [16]. But to approach the question of how space is represented in the Histories means first considering the ways in which Herodotus relates different places to one another and then exploring the nature, form, and significance of those relationships. In what follows, we briefly set out the beginnings of a network analysis of the place relations configured by Herodotus in Book 5 that will provide the broader cultural framework in and against which to read the opening movement of this book.

In recent decades network analysis has become increasingly important in the study of ancient history [17], with Herodotus even being used as a prime witness for many of the historical networks sketched [18]. Rather than using network theory to reconstruct a historical reality based on the Histories, our interest lies instead in using a study of the connections Herodotus makes between spatial concepts to represent his construction of geographic space and help prepare for a close textual examination of its lived experience. Above all, this has meant reconsidering what kind of spatial information ought to be captured. In previous work, we have used a digital text of Herodotus’s Histories to extract and analyse all place-name data [19]. In the episode discussed above, however, the two figures involved reflect in some way the places from which they come – Aristagoras represents the outward-looking city of Miletus on the Ionian coast, Cleomenes, the inward-looking military state of Sparta. Thus, in order to encompass those phenomena that occupy a physical space in the topographical reality described by Herodotus, the notion of place needs extending to encompass ‘proxies’ – the peoples, individuals, or even non-human agents that convey important spatial information [20]. From this point, our analysis of Book 5 proceeds on a clause-by-clause basis, assigning any single mention of a connection between two geographical concepts (places or their proxies) a single value (per verbal form), while noting the directionality of the relationship (how place/proxy x acts on place/proxy y) [21].

The results of this systematic approach are presented in Figure 1. At a glance, the sheer complexity of spatial relationships in Herodotus (Book 5) is evident, even after some necessary simplification to make the data readable [22]. All places and/or proxies mentioned in Book 5 are represented in the graph by a single point (or node), the size and boldness of which corresponds to the number of times they occur – the more they are mentioned, the larger and darker the node. The lines (or edges) mark a connection of some kind between two places and/or proxies, their thickness indicating the frequency by which the connection is drawn in the text. Arrows show the direction of the connection – whether place (or proxy) A is acting upon place (or proxy) B, or vice versa. Above all, the places are represented not in their ‘real-life’ geographical locations but according to their importance in the network: that is to say, those places/proxies which are mentioned most often in a relationship of some kind to others gravitate towards the centre of the graph, while those places/proxies on the fringes are those which are least related (though they may be important in other ways).

With careful analysis, the network graph can be used to paint a general picture of spatial relations in Herodotus Book 5. The places that occur most frequently, indicated by the largest nodes in the graph, are (in rough order of magnitude) Persia, Paeonia, Sparta, Ionia, Miletus, Athens, Sardis, Greece, Attica, Cyprus, Susa, Hellespont, Caria, Chios, etc. Many of these will come as no surprise for a reader familiar with Book 5, though even this basic information has the potential to flag up some unanticipated outcomes, such as the distinction that Herodotus makes between Attica and Athens – that is, between the settlement and its territory – or the strong unidirectional link from Sparta to Athens, which indicates Spartan dominance over Athens. It is this relationship that puts Athens under so much stress that Attica comes under threat. What the graph fails to depict, however, is Athens’s resistance to that influence and Sparta’s ultimate failure [23].

The most frequently occurring places cited above appear to divide roughly into two distinct, though to some degree interrelated, spheres – Sparta, Attica, Athens and Greece (along with Corinth, Croton, and Phoenicia) on the one hand, with Miletus, Susa, Ionia, Cyprus, Chios, Hellespont, Caria, Paeonia, Phrygia, the Aegean Sea and Persia on the other. These groupings would appear to substantiate the polarised split between East and West, which we had earlier contested; but the network picture is a good deal more nuanced than that. There is a significant asymmetry to the networks: while the ‘Greek’ network centres on two places, Athens and Sparta, the ‘other’ network more complexly builds around a number of places – Miletus, Ionia, Sardis, Paeonia, Cyprus, Susa, etc – with Persia at the centre, suggesting Persia’s influence both east and west. The presence in this sphere, moreover, of two important transit regions, the crossing point of the Hellespont and the Aegean Sea, underlines the coverage of Persia’s reach westwards, to the extent that Sardis sits next to Persia – Sardis is here almost synonymous with Persian power. But it is important to note the dual-directionality of the relationship between Persia and Miletus – Persia is not the only agent for change here, and Ionian culpability is an important part of the story (as we will see below in our narrative explication of the graph). Finally, the relationship between Persia and Miletus is reflected in that between Athens and Sparta, only this latter relationship is unidirectional and points to strong inter-Greek currents, which significantly complicates the impression of a coherent opposing sphere. This is no simple division between Greeks and barbarians.

The depiction of the broader conceptual categories, Greece and Europe, supports this more complex picture. It is notable that the former is situated in the ‘no-man’s land’ between Miletus and the sphere of Greek city-states, suggesting that Greece is an idea that is being fought over as the divisions begin to open and conflict ensues. More mysteriously, Europe is located in the Persian ambit, perhaps indicating that it is Persia’s activities that lead to the concept of Europe gaining hold in the narrative. Bringing together our initial impressions, we suggest the emergence of two spheres of influence coalescing around Persia, on the one hand, and Athens and Sparta on the other, with Miletus as the gateway between the two. But this scenario includes relationships that cut across that any notional boundary in ways that both underpin and undermine polarised readings of Herodotean space.

As is clear, while the network graph usefully brings to light interesting and not altogether anticipated relationships that are worthy of further investigation, it is not an end in and of itself. Rather, we suggest that such visualisations, while a product of close textual analysis, are inevitably also part of a continuing interpretative process. Their utility lies in helping us shine a light on the dense spatial relations that underpin the text, and in prompting further explanation and exploration of identifiable patterns. Hence, for the rest of this paper, we will tease out some of the patterns observed in this brief survey, notably the striking prominence of Paeonia in this network graph and in particular its relationship to Persia, as we read the opening episode of Book 5 in the light of this conceptual framework.

The Paeonians: A World on the Move

For the first twenty-three chapters of book 5, Herodotus uses the story of the Paeaonians to stitch together his narrative. As an ethnic group, the Paeonians are already a good example of the kind of additional information not captured by place-name alone – the category of ethnicity fixes people spatially just as the patronymic fixes them vertically [24]. But this section is also useful for picking away at the main narrative impulse toward the clash between Greeks and Persians. Through his description of the Paeonians’ spatial relations, Herodotus subtly suggests alternative routes through the gathering storm that does not easily or comfortably fall back on an East-West dividing line. In contrast to his historical agents’ hardening attitude towards conflict, Herodotus’s discursive space depicts an ever-evolving world that allows for a more complex and flexible understanding of the ties that bind them.

Identity Issues

The beginning episode of Book 5 introduces the various players. Herodotus describes the Persians as those ‘whom Darius left in Europe under the command of Megabazus’ (5.1.1), thereby defining this group through two key individuals, Darius, their primary signifying agent, and his proxy, Megabazus. In contrast stand the Thracian peoples to whom Darius now turns, both of whom Herodotus introduces in terms of natural features – the Paeonians live by the river Strymon (1.2), the Perinthians belong to the Hellespont (1.1). These definitions are important for setting the scene for the rest of the book. The Persians will be a catalyst for and protagonist in many of the relationships in Book 5, on the basis of Darius’ concern to extend the reach of his empire. The Thracians, however, defined by their land, introduce another important theme – the difficulty of enforcing control over other peoples and places because of the nature of the geography. Indeed, the Hellespont is going to be a critical ‘hotspot’ for Persian movement, as suggested already by the network graph. Similarly, the Strymon will be important for fluid notions of Thracian power [25], ironically not by virtue of being a river – for rivers in Herodotus tend to be markers of fixed lines that one crosses at one’s peril [26] – but as the one stable feature in the landscape that highlights a dangerous new turn of events – the founding of a city [27]. Thracian lived, and not lived (empty), space will become a major obstacle to the Persian conquest.

When discussing the graph above, we already had cause to mention Herodotus’s use of the large conceptual category ‘Europe’. To expand on that point here, its presence in the opening sentence of the book (as in ‘The Persians whom Darius had left in Europe’) harks back to the earlier narrative, which had been about Persian expansion into other areas (Egypt, Scythia), and also looks forward to the stakes in this and subsequent books [28]. Asia and Europe are coming together, again (cf. 1.2-4). In the beginning of the Histories, Persian wise men suggest a clear division between Europe and Asia – Asia is theirs, they say, and all the barbarian peoples who inhabit it, while Europe and the Greeks are separate (1.4.4). Book 5 is going to complicate that picture, even as it reignites the conflict.

Other relationship types set the scene for what is to follow. This is clearest in the link between the Paeonians and Perinthians, whose conflict extends over five clauses in 5.1 and leads to one of the groups, the Perinthians, being transformed as a result – the Paeonians triumph over them. But, Herodotus is keen to observe, this conflict is just an example of a time ‘before’. Now, in the chronological present of Herodotus’s narrative, the Perinthians were the ‘first’ to be subdued by the Persians. Thus, Herodotus looks both forward and back, back to a time when neighbourly conflict led to the defeat of the Perinthians, forward to a time when the Perinthians will be just the first of many groups subjugated by a far superior invading force. That earlier conflict with their neighbours cannot have been all that catastrophic apparently since the Perinthians have retained their identity enough to be now subject (again) to invasion – but will it be the same here, or are the Persians the game-changer? That certainly seems to be the implication of Herodotus’s description of how the Perinthians now fought bravely ‘for their liberty’ (5.2.1) – the first mention of the war against the Persians as a fight for freedom [29]. Only then, when they have been (irrevocably?) conquered, does Herodotus record the actual city of Perinthus (2.2). This is a good example of the meaning of space-changing before our eyes. With the people having lost their liberty and the city having fallen, the place is named. Now, it is simply a plot of land or resource to be possessed by its conquerors.

Filling in the Gaps, Making the Connections

Darius is the prime mover behind the change in function of this territory. But, while the Perinthians may have been conquered (again), there is a suggestion that the rest of Thrace might not be all that easy, even as Megabazus marches through the land subduing ‘every city and every people of those living in that place’ (2.2). The distinction between place and people picks up on the capture of Perinthus – a place to be possessed now that the people are defeated – as well as on Darius’s most recent (and unsuccessful) campaign in Scythia, where there were no cities and the people remained unconquered. Something similar will happen here, in spite of the apparent complete success of Megabazus’s blitzkrieg. Indeed, at this point, with the conquest of Thrace in Persian sights, Herodotus breaks off to supply details about this place. The sudden shift in tone and break from the chronological account of Persian conquest disrupts that narrative. And, by filling in our knowledge of Thrace, Herodotus sets the scene for its resistance to Persian rule [30].

First, Herodotus compares Thrace to India, which has been used as a point of comparison before, when Darius himself had tested cultural norms (3.38). There, Herodotus had again broken with chronology in order to use Darius to raise issues about cultural relativism in the context of Cambyses’s assault on Egyptian customs. Here, Herodotus returns to that same comparative example to turn the tables on Darius. At first sight, the comparison seems somewhat exaggerated – after all, Thrace is not quite at the ends of the earth (certainly not so far out on the margins as was Scythia). However, Herodotus’s point seems to be more political than geographical: in his judgment, Thrace would be by far the strongest of all nations if they were ruled by one man or had the same intention in mind’ (5.3.1). There is more to a place than natural geography or size alone: power rests largely on the kind of political settlement a place enjoys and the relationship between the inhabitants it promotes. The irony is that it is precisely Thrace’s plurality – the tribes have many names, each according to its place (3.2) – that will prove difficult to control.

How to place Thrace will be important for Darius’s Persians, of course, but no less so for the reader. Are they Greek or non-Greek (2.2) [31]? Or, rather, how (non-) Greek are they, and from whose perspective? And how do factors like ethnicity and topography impact on the ability to resist Persian expansion? Continuing his Thracian ethnography, Herodotus recounts a ritual for the dead (5.4) that, while appearing to be typically outlandish – they celebrate when someone dies! – actually hints at the pessimistic vein that runs deep in Greek ideas about the human condition [32]. Similarly, he describes the Thracians as worshipping only Ares, Dionysus, and Artemis while their kings worship Hermes – all still Greek gods (5.7). This case of defining non-Greek beliefs by Greek means includes a fascinating description of the distinction between kings and citizens (5.7). ‘Citizens’ is a highly charged term in the context of a narrative that places so much stress on the importance of being free politically. But it begs the question of what status these citizens have if they are governed by kings and live in tribes, not cities. Thus, this episode continues many of the themes hanging over from Darius’s failed invasion of Scythia, raising the spectre, as Darius moves westwards, whether his invasion will be replaying the Scythian catastrophe over again. Furthermore, it anticipates another marginal group whom Megabazus’s Persians are soon to encounter, the Macedonians, somehow closer to the mainland Greeks topographically and culturally, but also crucially different too, though in what ways and with what significance is hard to say. Uncertainty about Macedonian identity – Are they Greek? How Greek are they? – has consequences for thinking about the space they occupy too and how it is lived in [33]. What is clear is that the political and cultural alterity of these quasi-Greek groups skirts around their geographical marginality to confuse foreign invaders and historical inquirers alike.

Herodotus demonstrates the problem of conceptualising Thracian space in the chapters that follow. If he had earlier couched some of his language in vague terms (such as ‘those above’ 5.3.2; cf. 5.5.1), now the flow of his prose becomes almost choked with uncertainty. ‘As for the region which lies even north of this country’ (5.9.1) – the interesting addition of ‘even’ marks a serious distance from the chartered territory. ‘What lies already beyond the Ister’ – the shift in time signified by already underlines a conception of space that is both experiential and hodological, as if Herodotus were following some (unspecified) route. The way north signals the direction, but the river marks a boundary between what can be known and what really cannot. The men who live out here are very broadly defined as ‘persons’ while the land itself ‘appears’ boundless and empty, unlived in.

The narrator’s rhetoric abounds with similar expressions of ignorance: ‘none can tell with certainty’ (9.1); ‘I am able to learn only of certain persons dwelling beyond the Ister’ (9.1); ‘they say’ (9.3); ‘I am not able to work out’ (9.3) [34]. The space seems so foreign that, it is said, its borders reach almost as far as the Eneti on the Adriatic, while the inhabitants themselves, ‘Sigynnae’, say they are colonists from Media. Herodotus does not pass judgment on this remarkable claim; he simply comments that ‘all is possible in the long passage of time’ (9.3), before noting that the people who dwell about Massalia use the term ‘Sigynnae’ to describe traders, while Cyprians use it for spears. The shift to the longue durée to account for the comparably stretched spatial dimension is a feature of the Histories, in which time and space go hand-in-hand. Above all, it represents a world in flux, neatly captured by Herodotus’s prose, as he brings together in some kind of conceptual network places as far-flung as Thrace, Media, the Adriatic Sea, Massalia and Cyprus [35]. Even the etymological wordplay on the name of these people to mean traders hints at movement and exchange or, in the language of the Cyprians, a particular kind of movement and exchange – spears foreshadow the onset of war from Media [36]. Cyprus is another place highlighted in the network graph. Soon enough it will be part of a rhetoric of conquest (5.31.3; 49.6), and soon after that, it will be full of spears (5.104-116), winning and losing freedom. The Medes are coming.

These borders of Thrace are another matter, however, and the difficulties confronting whichever force would seek to conquer such a place are clear – what cannot be grasped cannot be controlled. In lieu of apprehending this inland territory, Megabazus keeps to what the Persians do know, the area by the coast (5.10), which he now subjugates. While this happens, Darius returns to the Persian sphere of influence (Sardis, after crossing the Hellespont, 5.11.1), and offers his Greek allies a choice of place – rewards in this political climate take the form of land possession. Accordingly, one of these special advisors, a man by the name of Histiaeus of Miletus, asks for Myrcinus in Thracian Edonia (5.11.2). Histiaeus’s desires seem to match Darius’s narrative, for the time being at least.

Moving Through Space and Time

An essential feature of the Histories’ lived space identified above is the idea that space does not stay the same – groups move in and out, cities wax and wane. Book 5 begins with the Persians moving into Thracian territory with an eye on controlling (and thereby) changing it. Similarly Darius invites his Greek allies to Sardis to offer them a choice of place over which to preside. This story already suggests that the division between East and West is not going to be so clear-cut as all that – Greeks are involved in this business of land-grabbing too [37]. But the clearest example, which introduces another kind of occupying movement, relates to a people – and here we return to the Paeonians.

Following hard on the heels of Histiaeus and Coes’s dream locations comes the story of two Paeonian brothers, who travel to Sardis in the hope of becoming tyrants of their people – more space-power dynamics. In their gambit for absolute power, they already seem to have grasped the Persian way of ruling; but, by inviting the Persians in, they get much more than they bargained for. Perhaps they should have taken greater note of their present location: a Lydian city presided over by a Persian king – as Herodotus keenly observes (5.12.2), while the two brothers are made to wait for Darius. Certainly, they fail to appreciate the full consequences of parading their sister before the king. ‘Darius took note of the woman as she passed by him, for what she did was not in the manner of the Persians or Lydians or any of the peoples of Asia’: once again a woman is used as both the instrument through which power is sought and symbolic of it [38]. Her attractions mirror those of their land, Paeonia, which is precisely the danger that the brothers have evoked. Held in thrall (13.1) by her, Darius wants to know more. ‘What persons are the Paeonians and what land do they live in and why have they come to Sardis?’ (13.2), he asks in the manner of a proto-investigator [39]. But his three questions, relating to the people, the land in which they dwell, and their relationship to Sardis, also reveal the nature and extent of the king’s gaze. He seeks to possess knowledge of their land as a precursor to possessing the land itself.

The brothers subsequently, and unwisely, advertise the benefits of their land to the king: Paeonia is on the Strymon, close to the Hellespont – water here is described in terms of facilitating communication rather than as a boundary to it, while the Hellespont represents the point up to which Persian influence is clear and through which they can pass over to the other side. And, even more unwisely, they advertise themselves: ‘we Paeonians are apoikoi of the Teucrians from Troy’. They are, and always have been (according to the brothers’ account at least) under the control of an Eastern potentate. What they hope to get out of this bargaining is left hanging. Certainly, they fail to sense the danger of making their request and their sister too desirable. The catastrophic consequences are sudden. Darius sends a letter to Megabazus back the way that the brothers had come with the order to bring all the Paeonians to him (5.14). The Paeonians are coming home.

The reader has already been alerted to the outcome of the brothers’ desire for tyranny right at the beginning of the episode (5.12.1). There Herodotus had used the large conceptual categories of Asia and Europe to mark the high stakes involved. That is replicated too in the very grammar describing the messenger’s journey, for whom it now takes two clauses to pass the Hellespont, one to get to it, the other to cross it (5.14.2) – a movement and pause that will be echoed when Megabazus returns, carrying the Paeonians away in captivity (5.23.1). The conflict itself is swift. The Paeonians march to the sea (5.15.1), thinking that the Persians would attack there, as they had done before (5.10), but the Persians instead use local guides to take the high road (15.2). From there they fall upon cities deserted of men, resulting in the Paeonians each going their own way and giving themselves up. (Even in Thrace, the identity of the group is inextricably tied to cities – where people live and rule together.) The Paeonians who are taken are then defined as ‘Siriopaeones, Paeoplae and those dwelling as far as the Prasiad Lake’. This strikes a concluding note with respect to Persian activities in Thrace. All these groups are forcibly relocated into Asia (5.15.3; cf. 17.1, 23.1).

Will a Greek campaign be the same as these events in Thrace, or will it be different because of Greek cities (poleis)? For sure there are echoes here of Xerxes’s invasion. The Persians’ use of guides to take the high road recall the later, more famous ruse by which they eventually succeed in circumventing the first line of Greek resistance at Thermopylae. Or the fact that the Persians meet so little opposition even in Greece, as their opponents simply melt away and relocate to fight another day – as the Athenians paradigmatically demonstrate or the Scythians before them. Even here, however, the Persians meet with resistance, as this story of relocation gets more complicated than either we or the Persians could imagine.

Location, Location, Location

The Paeonian story, introduced from the beginning of Book 5 in the context of their victory over the Perinthians, is dominated by repeated movement – a theme which points to the slipperiness of control, the instability of lived space in the Histories, and the diffusion of power through the unwinding narrative. The movement of coming into conflict with their neighbours; the movement of the two brothers to meet the king in Sardis; the movement of the messenger who returns the same way carrying Darius’s instructions; the movement of Megabazus who carries out those instructions to the letter, first rounding up the Paeonians, then forcibly relocating them to Asia. Even the stories relating to the Paeonians’ origins stress movement – their ancestors come from Troy and made apoikia settlements in Thrace. But this latter story, which introduces a historical perspective, unsettles the otherwise straightforward tale of an Asian conquest of Europe. Though Herodotus uses these conceptual labels to show what is at stake in these movements, and he marks out the Hellespont as the fulcrum around which the two worlds revolve – or, better, the place through which one must travel either way –, the neat division between East and West appears more fluid than that. The Paeonians do not even come from Europe. Should their forcible relocation, then, be viewed as something of a return?

The challenge to the story of a neat division does not end there – it was after all a pair of Paeonians who invited the Persians to intervene in the first place, and it was after all the Paeonians who started this narrative of conquest in Book 5 when they overcame the Perinthians. (The Perinthians begin as the group defeated at the hands of the Paeonians, but subsequently anticipate what will happen to their former rivals: as Herodotus points out, while they, the Perinthians, are the first group to fall to the Persians, the Paeonians will soon follow suit.) After this apparently concluding note on Paeonian forced movement (5.15.3), Herodotus’s tale takes a surprise twist that has not been trailed: he turns his attention to all the Paeonians whom the Persians did not conquer (5.16.1). Of the Paeonians who resisted Persian domination are those who live ‘near the Pangaean mountains and the Prasiad Lake itself’. In what follows, it is not only the case that the natural features (mountains and lakes) define the conquered peoples; those features help perhaps to explain why they avoided capture. Megabazus ‘did in fact try’ (16.1) to take the lake dwellers, but the difficulty he encountered is hinted at in an ethnographic excursus that traces this people’s habitation back, Herodotus supposes, ‘to olden times’ (16.2). The struggle of the narrative to capture this type of terrain goes some way towards explaining why the Persians failed in their attempt to capture them.

The point is underlined by a far lengthier excursus of some six chapters about Persian misadventures in Macedonia. At first, Herodotus states that ‘there is a shortcut from the Prasiad Lake into Macedon (17.2) as if the going was going to be easy. But the nearby mine is the one from which Alexander later draws his daily talent of silver. Note that at this unspecified later date the land still serves the treasury of Macedon’s king, not the Persians’ – already the tale anticipates (successful) Macedonian resistance. Once in Macedon, Herodotus records another parade of women, mimicking the land to be conquered, only this time it is a trick, and the women turn out to be armed assassins; when the Persians come in search for their lost men, Alexander buys off their leader and the Persian rescue party depart none the wiser. The story the Persians are trying to tell of supreme military might and world domination is resisted by the narrative’s meandering path [40]. Not that the story can be reduced to a simple clash of East versus West either, as we have seen from Paeonians being implicated in their relocation and hints of their non-European origins. It should be noted too that Herodotus ends his Macedonian excursus with a discussion of the claims to the Greek heritage of this Alexander, son of Amyntas (5.22). What that will look like in a few books time for a reader, when Alexander mediates between the Mede and the Athenians on the eve of war (8.140-144), is anybody’s guess [41].

But arguably the biggest turn in this meandering narrative comes when Megabazus returns with the captured Paeonians (5.23.1). Along the way back (first to the Hellespont, then to Sardis – the two foci of Persian power at the present time) he passes Myrcinus, the place on the Strymon that Histiaeus had been given by Darius. Megabazus reports back to Darius what he sees, and he finds it worrying: not only is the Greek typically building a settlement (polis), but the location is a potential crossing point, there is plentiful wood for ship-building, mines of silver are nearby, and many people both Greek and foreign (Megabazus uses the term barbaros!) dwell about – all the tools necessary for turning this living space into a revolution of sorts. The almost ethnographer-like Megabazus views the topography in military terms and sees a threat to Persian security from someone both like them (a tyrant) and not like them (free-thinking). As a result, Megabazus warns Darius against allowing Histiaeus to return ‘to Greeks’ (23.3). Darius, ever alert to threats to his power, takes the advice and recalls Histiaeus in the pretense of doing him greater honour than anyone – an honour that manifests itself in the form of Histiaeus too undergoing a physical relocation. And not just anywhere in Asia but all the way to Susa – the distance neatly encapsulates the threat he poses – in the company of Darius, who now will not let him out of his sight. Thus is set in train the next movement of Herodotus’s narrative, with Histiaeus doing all he can to return to the Sea (like a good Greek) by joining forces with his cousin Aristagoras of Miletus, who will also fall out with the Persians (5.33-35). These are the humble, and self-motivated, origins of the Ionian Revolt, a movement that will set the cities of Ionia ablaze with ideas of freedom, and lead, however meanderingly, to Athens being set on fire by Xerxes.

Conclusion: A World in Flux

This chapter has investigated the Histories’ lived space in terms of a relational model that has identified and questioned the connections that Herodotus makes between different spatial concepts in his narrative. It has extended the idea of space from a set of geographical coordinates to encompass the peoples and individuals who occupy the land or who may be said to represent a particular place. Above all, it has sought to chart Herodotus’ discursive representation of space, based on a close study of spatial relations in Book 5, and using a network graph to help provide a conceptual framework for our own discursive interpretation.

Initial results show a much more complex picture than the usual East-West polarity would suggest. It is not that there is no division between Asia and Europe – in the case of Book 5, in particular, we see the battle lines being drawn up in a series of clusters, as indicated in the network graph. Rather, the divisions are frequently temporary or partial or are subject to constant revision or even challenge. The picture that emerges is of a world not rigidly and schematically divided into distinct territories – a model which Herodotus directly criticises – but one that is interconnected in various ways on various levels at various times.

As the book nears its end we find the Paeonians still looking for ways to return home, still resisting Persian rule. Far from being the end of the story when Darius forcibly relocates them to Phrygia (5.15, 17, 23 – and then not all of them: 16), the Paeonians turn up in Chios. From there they apparently make it back to Paeonia (98.4), back home – what they seem to regard as home, at least, the one from which the Persians removed them not the one to which the Persians relocated them, even if that could be considered their ancestral home; some spatial concepts are too deeply embedded in a people’s consciousness for total fluidity (5.98) [42]. Writing space as it shifts this way and that is no easy task, and it is no easier reading it. If Darius finds territories and peoples more difficult to comprehend and control than anticipated, the same goes for Herodotus’s reader, for whom this partial and provisional narrative brings to life the engraved world according to Aristagoras.

Bibliography

de Bakker, M. (forthcoming) “An uneasy smile: Herodotus on maps and the question of how to view the world”, in: Barker E.T.E. et al. (eds.) New Worlds out of Old Texts: Developing Techniques for the Spatial Analysis of Ancient Narratives, Oxford

Barker, E.T.E. (2009) Entering the Agon: Dissent and Authority in Homer, Historiography and Tragedy, Oxford

Barker, E.T.E., Bouzarovski, S., Pelling, C.B.R., and Isaksen, L. (2010) “Mapping an ancient historian in a digital age: the Herodotus Encoded Space-Text-Image Archive (HESTIA)” Leeds International Classical Journal 9.1 (http://www.leeds.ac.uk/classics/lics/)

Barker, E.T.E., Bouzarovski, S., Pelling, C.B.R., and Isaksen, L. (eds.) (forthcoming), New Worlds out of Old Texts: Developing Techniques for the Spatial Analysis of Ancient Narratives, Oxford

Barker, E.T.E. and Bouzarovski, S. (forthcoming) “Towards a qualitative network analysis of Herodotus 5”, in: Barker, E.T.E. et al. (eds.) New Worlds out of Old Texts: Developing Techniques for the Spatial Analysis of Ancient Narratives, Oxford

Barker, E.T.E. and Pelling, C.B.R. (forthcoming) “Space-Travelling in Herodotus 5”, in: Barker, E.T.E. et al. (eds.) New Worlds out of Old Texts: Developing Techniques for the Spatial Analysis of Ancient Narratives, Oxford

Braund, D. (1996) “River frontiers in the environmental psychology of the Roman world”, in: D.L. Kennedy (ed.), The Roman Army in the East. JRA Supp 17, 43-7

Cartledge, P.A. (20022 [1993]) The Greeks: A Portrait of Self and Others, Oxford

Chamberlain, D. (2001) “ ‘We the others’: interpretative community and plural voice in Herodotus”, ClAnt 20, 5-34

Christ, M.R. (1994) “Herodotean kings and historical inquiry”, ClAnt 13, 167-202

Dewald, C. (1981) “Women and culture in Herodotus’ Histories”, in: H. Foley (ed.), Reflections of Women in Antiquity, Philadelphia, 91-125

Dewald, C., Marincola, J. (eds.) (2006) The Cambridge Companion to Herodotus, Cambridge

Eide, Ø. (forthcoming) “Verbal expressions of geographical information”, in Barker, E.T.E. et al. (eds.) New Worlds out of Old Texts: Developing Techniques for the Spatial Analysis of Ancient Narratives, Oxford

Fabian, J. (1983) Time and the Other, New York

Fearn, D. (2007) “Narrating ambiguity: murder and Macedonian allegiance (5.17-22)”, in: Irwin, E., Greenwood, E. (eds.) Reading Herodotus: A Study of the Logoi in Book 5 of Herodotus’ Histories, Oxford, 98-127

Friedman, R. (2006) “Location and dislocation in Herodotus”, in: Dewald, C., Marincola, J. (eds.) The Cambridge Companion to Herodotus, Cambridge, 165-77

Greenwood, E. (2007) “Bridging the narrative (5.23-7)”, in: Irwin, E., Greenwood, E. (eds.) Reading Herodotus: A Study of the Logoi in Book 5 of Herodotus’ Histories, Oxford, 128-45

Gurevich, A.J. (1985) Categories of Medieval Culture, London

Hall, E. (1989) Inventing the Barbarian: Greek Self-definition through Tragedy, Oxford

Harley, J.B. (1989) “Deconstructing the map”, Cartographica: The International Journal for Geographic Information and Geovisualization 26, 1-20

Hartog, F. (1988 [1980]) The Mirror of Herodotus: The Representation of the Other in the Writing of History, Berkeley

Harvey, D. (1985) Consciousness and the Urban Experience, Baltimore

Hornblower, S. (2008) A Commentary on Thucydides: Volume III Books 5.25-8.109, Oxford

Hornblower, S. (forthcoming) Herodotus Histories Book V, Oxford

Immerwahr, H. (1966) Form and Thought in Herodotus, Cleveland

Irwin, E. (2007) “ ‘What’s in a name?’ and exploring the comparable: onomastics, ethnography, and kratos in Thrace (5.1-2 and 3-10)”, in: Irwin, E., Greenwood, E. (eds.) Reading Herodotus: A Study of the Logoi in Book 5 of Herodotus’ Histories, Oxford, 41-87

Irwin, E. and Greenwood, E. (eds.)(2007) Reading Herodotus: A Study of the Logoi in Book 5 of Herodotus’ Histories, Oxford

Irwin, E., Greenwood, E. (2007) “Introduction: reading Herodotus, reading Book 5”, in: Irwin, E., Greenwood, E. (eds.) Reading Herodotus: A Study of the Logoi in Book 5 of Herodotus’ Histories, Oxford, 1-40

Janni, P. (1984) La Mappa e il Periplo. Cartografia antica e spazio odologico, Macerata

Jackson, P. (1994) Maps of Meaning: An Introduction to Cultural Geography, London

Jebb, R.C. (19003 ) Sophocles: The Plays and Fragments, part II. The Oedipus Coloneus, Cambridge

Latané, B. and Liu, J. H. (1996) “The intersubjective geometry of social space”, Journal of Communication 46, 26-34

Malkin, I. (1998) The Returns of Odysseus, Berkeley

Malkin, I. (2011) A Small Greek World: Networks in the Ancient Mediterranean, Oxford

Malkin, I., Constantakopoulou, C. and Panagopoulou, K. (eds.) (2009) Greeks and Roman Networks in the Mediterranean, New York

Marincola, J. (1987) “Herodotean narrative and narrator’s presence”, Arethusa 20, 121-37

Munson, R.V. (2001) Telling Wonders: Ethnographic and Political Discourse in the World of Herodotus, Ann Arbor

Munson, R.V. (2006) “An alternate world: Herodotus and Italy”, in: Dewald, C., Marincola, J. (eds.) The Cambridge Companion to Herodotus, Cambridge, 257-73

Payen, P. (1997) Les Îles nomades. Conquérir et résister dans l’enquête d’Hérodote, Paris

Pelling, C.B.R. (1997) “East is east and west is west – or are they? National stereotypes in Herodotus”, Histos 1, 51-66

Petropoulou, A. (1988) “The interment of Patroklos (Iliad 23.252-57)”, AJP 109, 482-95

Purves, A.C. (2010) Space and Time in Ancient Greek Narrative, Cambridge

Romm, J. (2006) “Herodotus and the natural world”, in: Dewald, C., Marincola, J. (eds.) The Cambridge Companion to Herodotus, Cambridge, 178-91

Rood, T. (2006) “Herodotus and foreign lands”, in Dewald, C., Marincola, J. (eds.) The Cambridge Companion to Herodotus, Cambridge, 290-305

Rood, T. (2012) “Herodotus”, in: Jong I.J.F, de (ed.) Space in Ancient Greek Literature: Studies in Ancient Greek Narrative, Leiden, 121-40

Rood, T. (forthcoming) “Mapping spatial and temporal distance in Herodotus and Thucydides”, in: Barker, E.T.E. et al. (eds.) New Worlds out of Old Texts: Developing Techniques for the Spatial Analysis of Ancient Narratives, Oxford

Thomas, R. (2000) Herodotus in Context: Ethnography, Science and the Art of Persuasion, Cambridge

Thrower, N.J.W. (1996) Maps and Civilization: Cartography in Culture and Society, Chicago

Turnbull, D. (2007) “Maps, narratives and trails: performativity, hodology and distributed knowledges in complex adaptive systems – an approach to emergent mapping”, Geographical Research 45, 140-49

Notes

1. This chapter is based on the work carried out by Barker, Bouzarovski, Pelling and Isaksen for the Hestia project, funded by the UK’s Arts and Humanities Research Council. See Barker et al. (2010), (2014). We have benefitted greatly from Klaus Geus’s common sense criticism; any lack thereof remains our own.

2. On this scene, see especially: Purves (2010: 132-138, 144-150); Rood (2012); de Bakker (2014); cf. Rood (2006: 294-296); Barker et al. (2010: 6-9).

3. All translations are our own.

4. On Herodotus’s treatment of space, see especially: Purves (2010: 118-158); Rood (2012), (2014); de Bakker (2014). On the trope of the text (of Herodotus) as topography, see: Hartog (1988); Payen (1997); Chamberlain (2001); Greenwood (2007).

5. As Purves (2010: 133-138) argues, by virtue of being engraved, Aristagoras’s map only offers a snapshot of space and fails to capture its full meaning. Cf. Barker et al. (2010: 4-5); de Bakker (2014).

6. On ‘easy’ being a watchword in Aristagoras’s discourse, see Pelling (2007: 179-180, 183).

7. Cf. Purves (2010: 140).

8. Janni (1984). Purves (2010: 126) suggests that Herodotus’s periodic style deliberately challenges a cartographic approach to spatial representation. The picture is even more complicated than we suggest, since Aristagoras’s description superimposes a ‘route’ approach to space on top of the bird’s-eye view that the map presents.

9. For a discussion of some of the differences between the textual and visual mapping of geographical information, see Eide (2014). On the flattening out of spatial understanding as a result of Cartesian mapping, see e.g. Harvey (1985: 253); cf. Fabian (1983); Gurevich (1985). On political and cultural constructions of space, see e.g. Thrower (1996); Harley (1989: 1-20); Jackson (1994).

10. On space and time in Herodotus, see Rood (2014).

11. See Friedman (2006: 166) on how “Herodotus’ perspective will be a deterritorialised one”.

12. Or, in this case, not, as Aristagoras’s appeal for Ionian unity in the face of perceived Persian aggression falls on death Dorian ears. The same Ionian call-to-arms will enjoy more success when he delivers the same promises (and more) to the Athenians (5.97).

13. On his rejection of overly schematic representations see Munson (2001: 84-86); Thomas (2000: 98-100); Purves (2010: 111-112, 128-132). Rood (2012: 133-135) nuances this view by drawing attention to the various moments in the Histories when the narrator takes a bird’s-eye view of a particular place. Cf. de Bakker (2014) on Herodotus’s more authorial geography of Egypt.

14. E.g. Hall (1989); Cartledge (2002). For qualifications and some re-nuancing of Herodotus’s East-West polarity see Hartog (1988), along with the review by Carolyn Dewald in Classical Philology 85 (1990: 217-224); and Pelling (1997).

15. ‘Herodotus’ inquiry into other lands and customs proves to be as central to his project as his inquiry into the wars fought by Greeks and non-Greeks’: Rood (2006: 290).

16. On the centrality of book 5, see Irwin and Greenwood (2007); cf. Hornblower (forthcoming).

17. See now Malkin (2011). Cf. Malkin, Constantakopoulou and Panagopoulou (2009).

18. Malkin (1998).

19. See Barker et al. (2010).

20. Herodotus provides a classic statement of this problem, when he depicts Themistocles threatening the fragile Greek coalition with the prospect of the Athenians taking to their ships and moving Athens to Southern Italy (8.62.2). Thucydides offers further evidence: as the catastrophic Sicilian Expedition nears its terrible climax, Nicias tries to rouse his Athenians for one last battle on the basis that ‘men are the polis, and not the walls or ships empty of men’ (7.77.7) – the mention of ships even seems a deliberate echo and distortion of Themistocles’s threat. For a collection of ancient sources, which diverge on the relative importance of people over place, see Hornblower (2008: 720-721).

21. See Barker and Bouzarovski (2014), in which the analysis is refined according to idea that geographical spaces convey movement and/or transformation. On ‘the creation of knowledge spaces in which people, practices and places are discursively linked’, see Turnbull (2007) 143. Cf. Latané and Liu (1996).

22. For the sake of intelligibility and in order to explore the possibilities afforded by general patterning, only those places that occur most frequently are labelled and discussed here, though in theory any point could be explored for its connections (or lack thereof) to the network. For the visualisation, we are indebted to the help provided by Scott Weingart for use of the Sci2 tool developed at Indiana University: https://sci2.cns.iu.edu/user/index.php.

23. See Barker and Pelling (2014).

24. We owe this description to Simon Hornblower.

25. The Hellespont and river Styrmon symbolise themes of great significance, perhaps at the expense of strict topographical accuracy: Hornblower (forthcoming) 76-77 questions the accuracy of the geography here. On the Hellespont as a hotspot of communication, see Barker et al. (2010) and ‘moving through space and time’ below. See also Greenwood (2007) on the Hellespont ‘as a symbolic space between two continents that has geographical, ethnographical, and historical significance, [which] represents the kinds of repeated crossings that the reader of Herodotus has to make in order to comprehend the significance of the different dimensions of the narrative (128).

26. For rivers as demarcation boundaries and markers of transgression, see especially Immerwahr (1966); cf. Braund (1996). But Herodotus’s depiction of man’s control over the natural world is more nuanced: see Romm (2006).

27. The importance of the Strymon is picked up in ch.23, when Megabazus observes the strategic importance of Histiaeus’s town Myrcinus. See ‘location, location, location’ below.

28. Hornblower (forthcoming: 73) talks about how these opening chapters at first sight seem more in keeping with Book 4, ‘but there are also important pointers forward’. On how book 5 bridges the narrative see Greenwood (2007). Cf. Irwin and Greenwood (2007: 9-19).

29. Hornblower (forthcoming: 79).

30. On the Histories as a territory that resists conquest, see Payen (1997). Irwin (2007: 43) notes how the Paeonian ‘narrative is configured as topography’.

31. Irwin (2007: 43).

32. The idea of not being born is best (or, if that is not possible, of dying as quickly as possible) is most memorably expressed by Sophocles’s chorus in Oedipus at Colonus (1225-1226), but is hinted at elsewhere (e.g. Theognis 425-427), as Jebb (1900) ad loc. notes. This aspect of Thracians identity is even more strongly marked if we hear resonances with Patroclus’s funeral games: Petropoulou (1988: 492-493); cf. Munson (2001: 167).

33. On the ambiguous identity of the Macedonians, see Fearn (2007) and Hornblower (forthcoming: 104-105), who in passing mentions the analogous banquet of the suitors in the Odyssey. (Of course, slaughtering guests, even if they are barbarians behaving badly, is not exactly the sign of a good Greek heritage, but, if Odysseus did it…) Note too how Alexander prevented the truth coming out ‘by guile’ (sophiê, 5.21.2) – a fifth-century gloss on Odysseus’s famous mêtis. Herodotus is using Homer, the paradigm of Greek culture, to interrogate the values and assumptions of that culture.

34. Herodotus’s language here recalls his description of Egypt (a comparison at the other extreme), where he talks about the Nile Delta as the one area that is known about or knowable (like the places next to the sea in the geography of Thrace). Nevertheless, this waterway (the Nile) does give access (like an umbilical cord) to these places, however unknown they are. On Herodotus’s description of Egyptian topography and his rhetoric of unknowns, see Marincola (1987); Purves (2010: 129-130); de Bakker (2014).

35. For this phenomenon, in which Herodotus leaps from place to place, drawing a web of association across the Mediterranean, see Munson (2006: 258).

36. On the wordplay for ‘Sigynnae’, see Irwin (2007: 59-60, 83-87), which she relates to criticism of Athenian imperialist ambition. Certainly Herodotus’s shifts in time encourage multiple ways of viewing the same space.

37. People who trigger the great east-west movements carried out by the Persians are often westerners. See Barker and Pelling (2014).

38. On women as the instrument through which power is sought and symbolised, see Dewald (1981: 113). On the resonance with the similar display of a woman (from the deme of Paeania, 1.60) as part of a ruse for gaining power, see Irwin (2007: 51).

39. See Christ (1994).

40. Rood (2006: 294), cf. Payen (1997).

41. See Barker (2009: 193-198).

42. When they are in Chios, the Persians send a message telling them ‘to go away back’ (5.98.4), which the Paeonians ignore and return to Paeonia under their own steam. Does ‘go away back’ mean ‘go back to Phrygia’? Or is the instruction ‘to go back to Paeonia’, in which case, are the Paeonians refusing to return under Persian escort rather than independently? (Thus Hornblower (forthcoming) ad loc, with the suggestion ‘because they feared a trick?’) Spatial categories are becoming so confused that finding a way ‘back’ may mean different things for different people.