Introduction

Summary

This introduction has three goals: to locate this book’s arguments in contemporary scholarship on Parmenides, to outline its methodology and structure, and to establish the stakes of the project as a whole. The first considers Parmenides’ invention of extended deductive argumentation and the practice of demonstration – the central topic of this book – as a relic of the ‘Greek Miracle’ paradigm; it also addresses discussions of Parmenides’ use of poetry and his relationship to Homer. The second addresses distinctions between actors’ and observers’ categories and between reasoning and discourse, and explores the Foucauldian discourse analysis that anchors the book’s treatment of the relationship between Parmenides and Homer. The third requires setting out what this book does not intend to do in order to specify its main contribution: providing an account of how Parmenides’ use of the image of the hodos helps him invent extended deductive argumentation and the practice of demonstration. The Introduction sketches out the three axes of the book’s argument: an exploration of the physical reality of archaic and classical Greek roads, a discussion of the semantics of the word hodos, and an articulation of the relationship between Parmenides’ and Homer’s poems.

Demonstration has been the cornerstone of claims to knowledge since at least the time of Aristotle.¹ But demonstration, and, more specifically, the extended deductive argumentation that forms its backbone, has a history. As is widely agreed, that history begins with Parmenides of Elea, in whose poem we find the first recorded extended deductive argument – and with it, the first outline of a demonstration.²

This is not the only reason why Parmenides has won acclaim, even veneration, from leading Western thinkers. Since the time of Plato³ (and – to judge from Parmenides’ influence on Zeno, Melissus, Democritus, and others – probably before), philosophers of many stripes, from Hegel⁴ to Heidegger,⁵ Russell⁶ to Popper⁷ to Anscombe,⁸ have celebrated Parmenides’ unique importance as the grandfather of their profession – though not always for the same reason. Historians of ancient philosophy and science similarly agree on the epochal importance of Parmenides’ contribution to Western thought but disagree on where, precisely, this importance lies. Some herald Parmenides for his primordial articulation of the notion of modality;⁹ others laud his groundbreaking advances in astronomy, especially his remarkable observation that the moon reflects the sun’s light (and, therefore, that the earth is spherical);¹⁰ others still foreground his seminal position in the atomic tradition.¹¹ Whatever their differences, however, nearly all acknowledge that Parmenides is the first recorded person to make an extended deductive argument, and nearly all accept that his poem shares key features with what Aristotle will later call apodeixis or ‘demonstration’. As one of the 20th century’s leading historians of ancient thought put it, ‘the aims of The Way of Truth are clear: Parmenides sets out to establish a set of inescapable conclusions by strict deductive arguments from a starting point that itself has to be accepted. Those are features it shares with later demonstrations.’¹²

Parmenides’ many other astonishing achievements do not, however, eclipse the fact that his confection of these three features – (i) proceeding from a starting point that has to be accepted (ii) by strict deductive arguments (iii) to establish an inescapable conclusion – marks a fundamental inflection point in the history of Western thought. The clarity with which we may state this is matched only by the intractable obscurity surrounding the development and fusion of these three features in Parmenides’ poem. This remains so despite agreement about Parmenides’ importance, and despite the quantity (and quality) of recent scholarship devoted to understanding Parmenides in relation to his Presocratic predecessors and successors.¹³ Exploring the origins of this complex of features (i–iii) and providing an account of their emergence, both as individual items and as a complex formed from them (viz. a ‘demonstration’), forms the central task of this book.

Two Enduring Problems: A Parmenidean Greek Miracle, and ‘Why Verse?’

There are good reasons for this intractability. The task of relocating Parmenides in his intellectual context is beset by deep, even potentially insurmountable challenges. The few ipsissima verba of Parmenides’ Milesian predecessors are embedded in settings, doxographical or otherwise, strongly marked by their pursuit of other, post-Parmenidean, agendas.¹⁴ Unless new original fragments appear, or a new understanding of the spread of people, information, and ideas can be persuasively established – or both – attempts to pin down the relationships between Parmenides and Xenophanes, or Anaximander, or Anaximenes¹⁵ (not to mention Heraclitus)¹⁶ will remain largely speculative¹⁷ (and may say more about our own conception of how ‘philosophy’ ought to work than anything else).¹⁸

Furthermore, our knowledge of the social, political, and intellectual dynamics of archaic poleis, especially in Magna Graecia, is too lacunose to identify with precision the influence of existing cultural, political, and legal institutions and practices on Parmenides.¹⁹ Vernant, responding to the connection between Hesiod and the Milesians posited by Cornford, mocked Burnet’s notion of the ‘Greek Miracle’, as if ‘[a]ll of a sudden, on the soil of Ionia, logos presumably broke free from myth, as the scales fell from the blind man’s eyes. And the light of that reason, revealed once and for all, has never ceased to guide the progress of the human mind.’²⁰ These words first appeared more than half a century ago, and in the interim an army of distinguished scholars has laboured to disassemble the Greek Miracle edifice block by block. Parmenides’ great foundation stone has escaped untouched, however: even now, we still have no detailed account that would explain just how Parmenides invented deductive argumentation, nor even one that links it to his predecessors’ modes of speaking and writing persuasively. Before Parmenides, Presocratics merely asserted;²¹ after him, they argued, and attempted to demonstrate.²² It is still as if, all of a sudden, on the soil of Elea, deductive argumentation and the practice of demonstration broke free from mere assertion, as the scales fell … In practice, the result is, as Malcolm Schofield put it, that ‘it is nowadays commonly supposed that Parmenides was a creative genius not much in debt to anybody’.²³

It is useful to juxtapose the scarcity of our knowledge of Parmenides’ social, cultural, and political setting with another quirk of the last century and a half of scholarship on Parmenides. While we often seem to be able to say too little about the tradition within, and out of, which Parmenides developed extended deductive argumentation and the skeleton of demonstration, scholars have ignored, and even lamented, aspects of his poem about which we might say much.²⁴ They have registered with dismay Parmenides’ linguistic extravagance, finding it incongruous with the triumph of austere reasoning whose birth we are supposed to witness in the ‘Route to Truth’.²⁵ How could Parmenides have elected to compose in verse?²⁶ (Especially if, as the consensus since Diels and Wilamowitz – not to mention Plutarch – has it, that verse is so defective.)²⁷ What could have motivated him to use such richly textured, imagistic language to formulate a deductive argument?²⁸ Why did he deploy the narrative mechanics and dramatic scenario of myth to stage reason’s great debut in Western thought?²⁹

The impulses animating these questions are perhaps understandable. It will always be both tempting and, at least to some extent, unavoidable to read Parmenides backwards through the prism of the formalized second-order analysis of demonstration and deductive argumentation established by Aristotle. There is no obligation, however, to read Parmenides exclusively according to the rules of this canon, even though, in many of its essential features, it continues to define the way that we think and argue.³⁰ In fact, it is precisely because the object of study here is in so many ways directly connected, and therefore immediately accessible, to our own intellectual practices, to what intuitively constitutes ‘good thinking’ today, that we must take special care.

How are we to do this? The question gives an extra bite to Geoffrey Lloyd’s insistence on the value to historians of ancient thought of the anthropologist’s distinction between ‘actors’ categories’ and ‘observers’ categories’.³¹ As a basic methodological principle, anthropologists attempt ‘to express the ideas, beliefs, [and] practices of the society in question in the terms used by members of society themselves – the actors’.³² What is at stake in doing so?

The aim of keeping as close as possible to the actors’ own categories is two-fold. Negatively, first, it helps to minimize the risks of assimilating alien ideas to our own, of assuming that the subjects studied have the same conceptual framework in mind that is suggested by the interpreter’s own (observer) categories. Positively, second, it allows an alien network of meanings to be built up in its own terms and be seen for what it is, as alien.³³

Both factors should be carefully considered in the case of Parmenides. While reading his ‘Route to Truth’ as no more and no less than the earliest attested example of an extended deductive argument helps us pinpoint one of Parmenides’ most important contributions to the history of thought, paradoxically, doing so blocks us from examining just how he accomplishes the very act – inventing extended deductive argumentation and the outline of demonstration – that we would study.

This is true on several levels. First, to characterize Parmenides’ poem as a deductive argument is implicitly to bestow upon it from the start all the qualities we today understand a deductive argument to possess; suddenly fragments 2, 6, 7, and, especially, 8, as ‘deductive arguments’, are truth-preserving, and so proceed according to a specific kind of logical necessity anchored a priori in what we would call the laws of non-contradiction and the excluded middle.

Or at least they ought to. For, second, labelling the poem a deductive argument has the consequence of establishing a distinctive interpretive frame and corresponding set of hermeneutic expectations.³⁴ Understanding it as a deductive argument first and last, one reads the poem against such criteria as validity and soundness, guards against such things as illicit modal upgrades³⁵ or confusions of necessitas consequentiae and necessitas consequentiis,³⁶ discusses its language and structure in the philosopher’s idiom of quantification and predication,³⁷ claims made de dicto and de re.³⁸ Appropriate intertexts become the Discourse on Method³⁹ or the Critique of Pure Reason,⁴⁰ ‘On Denoting’⁴¹ or the Tractatus.⁴² This has consequences. Judged by rules unformed and standards yet unknown for hundreds or thousands of years, Parmenides is perpetually – but also, given his nonpareil innovation as a practising logician, inexplicably – on the verge of suffering amateurish lapses or committing schoolboy blunders.⁴³

Even more significantly for the present discussion, such a stance excludes from analysis – because by definition they should have no bearing on the deductive validity of the argument itself – the imagery that shapes, guides, and inflects the language and structure of Parmenides’ argument; the argument’s dramatic framing; its intertextual relations (except insofar as these intertexts are other deductive arguments); and its relationship to its sociocultural and historical context. In fact, such a hermeneutic stance not only prevents these dimensions from being considered, but configures basic features of the text as problems. Why verse for a deductive argument?⁴⁴ Why the dramatic encounter between kouros and goddess in a proof about the nature of what-is?⁴⁵ Why so many images, such figurative language?

Similarly, referring to the poem as (simply) a deductive argument makes it hard to avoid retrojecting onto the poem’s earliest audiences a sense of the privileged status deductive argumentation today enjoys as the authoritative means by which to prove the validity of a claim. But why should a contemporary of Parmenides have found the sequence into which he ordered his claims compelling in and of itself?⁴⁶

Third, to approach the ‘Route to Truth’ from the presumption that one is reading a deductive argument is to accept as a fait accompli the very achievement one wishes to examine as the product of a complex process. The notion of a systematic argument of interlinked claims which begins from a necessary point of departure, proceeds from one claim to the next according to some kind of necessity, and ultimately arrives at a final destination, is all taken for granted of a demonstration (not least since these are among its defining features). But these are precisely the new elements that Parmenides introduces onto the discursive scene. To refer to Parmenides’ argumentative style as ‘deductive’ (and leave the matter there) is therefore to accept as a finished article that which we are in fact seeing fashioned before our eyes.

And this in turn, fourth, short-circuits from the start any attempt to examine the specific strategies and techniques by which Parmenides develops these new elements – precisely what we are interested in here. Calling this portion of his poem no less and no more than a deductive argument makes it seem as if this specific manner of advancing a claim (obviously and inherently superior, on this view, to its predecessors) had always been sitting around waiting to be discovered. To refer to Parmenides’ fragments 2–8 as a ‘deductive argument’ or a ‘demonstration’, with no further elaboration, thus threatens ipso facto to prevent us from gaining fundamental insights into the process by which deductive argumentation emerges, the very techniques and strategies Parmenides used to make this manner of expressing claims about the nature of what-is seem plausible, or even intelligible.

The Two Problems Resolve Each Other

Against this backdrop, Lloyd’s remark concerning the benefits of allowing ‘an alien network of meanings to be built up’ could hardly be more salutary. It is true that ‘the terminology in which [Parmenides] describes what he is doing is a very limited one’ and that ‘[h]e has no word for deduction’.⁴⁷ (Indeed, why would he?) But Parmenides does have language to describe the arguments that span fragments 2, 6, 7, and 8: and this centres on the programmatically repeated notion of what he calls a hodos dizēsios or ‘route of inquiry’.⁴⁸ What is more, if this ‘terminology’ is indeed ‘limited’ insofar as it is not part of a larger system of technical vocabulary coined for special purposes, it is in other ways far richer, deeper, and of more subtle texture for precisely the same reason. These terms, not being ‘technical’, remain the more powerfully charged by the currents of polysemy, ambiguity, intertextuality, and the play of signifier and signified, for remaining enmeshed in the web of language.

Or, rather, network. For in light of Lloyd’s call to use actors’ categories (and not – or not only⁴⁹ – our own), the gap (between Parmenides and his predecessors) and the excess (in Parmenides’ use of language and imagery) discussed above can be seen to form two sides of the very same Parmenidean coin. More: these two mysteries (where did Parmenides’ extended deductive argument and move towards demonstration come from? Why the poetry, the polyvalent language, and myth’s mosaic of imagery?), once viewed together, cease to be mysteries at all. Rather, each can be seen to provide the key that unlocks the other. To address the question of how Parmenides invented extended deductive argumentation, that is, we must return to his poem prepared to read it as a poem: to attend to the densely imbricated richness of his language and the many layers of resonance compressed in, and radiating out from, key words; to trace with care the imagery that Parmenides puts into circulation and mobilizes, activates, and exploits; to read and hear this poem alongside its major predecessors in dactylic hexameter, with ears sharply attuned to echoes in linguistic and imagistic detail, dramatic setting, plot mechanics, and formal organization and structure; and to relocate this poem in the physical and social reality of its time and place.

In the ‘network of meanings’ Parmenides builds up in his poem, no nexus of language and imagery bears a greater symbolic charge, or is asked to do more work, than the figure of the hodos just cited and its related language of roads, travel, and journeying.⁵⁰ My core claims are premised on the idea that providing an account of Parmenides’ invention of extended deductive argumentation requires that we examine the network emanating from, and compressed into, the phrase hodos dizēsios along three axes: the relationship between word and world, the relationship between signifier and signified, and the relationship between text and intertext. I shall address these points in turn.

The Agenda: A General Outline of the Book

First: archaic Greek roads were not at all like our own. The physical nature and social function of archaic Greek roads (to be discussed in Chapter 1.1) have been neglected by analysts of Parmenides, but have a crucial bearing on our understanding of Fragment 8. One of the most striking features of Parmenides’ text is the notion of necessity that defines the claims he advances and, depending on one’s interpretation, the sequence in which these claims are advanced.⁵¹ The multifarious techniques he uses to express this notion – including the invocation of personified forms of dikē, anankē, moira, and (possibly) themis; the deployment of images of binding or fettering (frs. 8.14, 8.26, 8.31, 8.37) and reference to ‘bounds’ or ‘limits’ (Fr. 8.26); and, most arrestingly, the repeated use of the words chrē and chreōn with a sense that is still hotly disputed⁵² – have been much discussed, with one exception: the physical nature of ancient Greek roads.

Second: the semantics of the word hodos and its neighbours in the Homeric semantic field impose a distinctive shape upon the overarching contours of Parmenides’ hodos dizēsios.⁵³ The semantic analysis conducted in Chapter 1.2 will suggest a conceptual footprint whose outlines are defined by the fact that in the Homeric semantic field, a hodos is always a hodos to somewhere, a journey oriented towards, and undertaken with reference to, a fixed, stable final destination, to an end.⁵⁴ The thematic use of the word hodos thus inscribes the endeavour denoted by the phrase hodos dizēsios within a distinctively teleological framework.

Finally, Parmenides’ use of the figure of the hodos orchestrates a complex web of associations with the use of the word and image in the Odyssey, and Odyssey 10–12 (and especially 12) more precisely. Here we are fortunate to be able to draw on two important studies of this relationship. Nearly six decades ago Eric Havelock first made the case for a Parmenides inspired by Odyssey 10–12:

We suggest … that he composed a philosophical poem partly in the mood of an Odysseus, voyaging successively to Hades and past the Planctae and Scylla and Charybdis to Thrinacia’s isle … Once books ten to twelve of the Odyssey (or a section approximating thereto) are accepted as his central frame of reference, the patterning of his poem becomes clearer and some of his symbols become easier to interpret.⁵⁵

Another of Havelock’s major insights was to reject the commonplace – still evident even in many sophisticated contemporary accounts – that one should draw a clear distinction between Fragment 1, with its symbolism, imagery, and narrative mode of organization, and the remaining fragments, particularly 2–8, where the ‘real philosophizing’ is thought to occur; his insistence that the influence of the ‘Homeric echoes’ in Parmenides ‘is not confined to the “proem” but affects also the general structure of Parmenides’ philosophical argument’ is of decisive importance.⁵⁶

Alexander Mourelatos’s influential 1970 study, The Route of Parmenides, developed this fundamental insight in a number of essential ways.⁵⁷ One important step forward was his elaboration of Havelock’s vague parallels between the itineraries Circe narrates to Odysseus and those Parmenides’ goddess narrates to the kouros.⁵⁸ Perhaps even more importantly, Mourelatos explicitly theorizes the relationship between these two texts, pairing the distinction between ‘motifs’ and ‘themes’ with a theory of metaphor according to which a metaphor sometimes ‘fashions a new outlook, a new concept’.⁵⁹ Just as when metaphors of this type are used, ‘old words, old motifs, old images are appropriated and extended towards the expression of ideas and concepts which are still in the process of development and formation’, so Mourelatos claims that ‘Parmenides uses old words, old motifs, old themes, and old images precisely in order to think new thoughts in them and through them.’⁶⁰ Specifically, ‘the image of the route mediates a new concept of the nature of thinking and knowing’.⁶¹

By pointing the way towards a reading of Parmenides that identifies the profound influence of Homer on his poem, Havelock and Mourelatos have each taken us forward a long way. Even so, their analyses leave several fundamental questions unanswered. Just how does Parmenides actually accomplish his mediation of a new concept of thinking and knowing? What specific role does the figure of the hodos actually play? In other words, how does the surface level of language (words, motifs, images) examined by Mourelatos relate to the ‘general structure of argument’ that Havelock invokes? And how do the individual words and images that Mourelatos studies achieve the revolutionary outcome – an ‘entirely new mode of thinking and knowing’ – that he identifies? Between individual words and general structure lies the entire domain of argumentation – its principles of construction, its architecture, its patterns of formation. And between the whence of the image and the whither of a new mode of thinking⁶² lies the entirety of the (met)hodos.⁶³ The terrain that forms these ‘betweens’ is what we shall explore in chapters 3 and 4 (on Homer, and Odyssey 12 in particular) and chapters 5 and 6 (on Parmenides’ ‘Route to Truth’, and especially fragments 2 and 8, respectively).

(Met)hodology

But how? We began with Geoffrey Lloyd’s observation that it was Parmenides ‘who was – as all recognize – the first to produce a sustained deductive argument’.⁶⁴ Note Lloyd’s use of the word ‘argument’ rather than ‘reasoning’. Though the relationship between argumentation and reasoning is theorized differently by different thinkers, Lloyd’s use of ‘argument’ undoubtedly refers to a discursive undertaking, as opposed to the mental activity often captured by the term ‘reasoning’.⁶⁵ It is thus the domain of discourse that Lloyd identifies as the decisive locus of innovation of Parmenides’ contribution to early Greek thought in this case.

This is a crucial insight. The distinction between reasoning and argument allows us to formulate a much more precise account of Parmenides’ place in the history of thought. If it would be absurd to say that Parmenides was the first person to reason deductively, it is of the utmost importance that he is the first person we have any record of attempting to articulate his deductive reasoning in the form of an explicit (and extended) discursive framework. Accordingly, any attempt to examine the origins and early evolution of deductive argumentation, or to examine the strategies by which Parmenides develops it, must be located at the level of formal discursive organization. My claim will be that in its formal organization – in the articulation of its arguments and in the manner in which these arguments are connected to each other – Parmenides’ revolutionary sequence of deductive arguments is deeply influenced by the Homeric strategies of narration deployed in Odyssey 12. These, I shall contend, form the basic underlying architecture of Parmenides’ epoch-making arguments.

To tie all these threads together: if Parmenides’ main achievement occurs at the level of discourse (not reasoning), and if his indebtedness to Homer can be found not only at the level of language or motif (as Mourelatos has it) but at the level of the poem’s structure and organization (as intimated by Havelock), what we need is a theoretical apparatus that allows us to identify, at the level of discourse (i.e. spanning the levels of both the individual word and, especially, ‘general structure’), the structural continuities that link Parmenides’ fragments 2, 6, 7, and 8 to Odyssey 12.

Michel Foucault’s analysis of discursive regularities, undertaken in his Archaeology of Knowledge, offers just such an apparatus. Although this neglected masterpiece has been criticized for presupposing too static a view of discursive regularities (and therefore having difficulty accommodating, let alone explaining, change), this quality is precisely what makes it so valuable in this setting:⁶⁶ for all that Parmenides’ deductive argumentation has traditionally been presented as a radical rupture with the past, one of my main goals in this book is to emphasize its fundamental similarity to the mode of narration that structures Odyssey 12.

Explaining how Foucault’s notion of ‘discursive regularities’ can help us identify more precisely the level at which Parmenides most relies on – and best analyse the specific ways he refashions – the Homeric poem he inherits requires a brief discussion of Archaeology of Knowledge.⁶⁷ It is helpful to understand the Archaeology of Knowledge as expressing a kind of methodological manifesto for a programme of an Annaliste epistemological history;⁶⁸ this is so insofar as it fuses the French Annales School’s interest in the formation of series, viewed from the perspective of the longue durée, with a focus on the processes of knowledge production and a fine-grained concern for distinctive layers or strata of continuity and discontinuity that define the relationships between these different processes.⁶⁹

One of the fundamental units of analysis produced by this fusion is the discursive regularity. For the Annales School so closely associated with it, investigating the longue durée involved looking at regular patterns or ‘structures’ formed by the relationship between such things as, for example, ‘geographical frameworks, certain biological realities, certain limits to productivity’ and specific patterns of human activity – such as, for example, ‘the persistence of certain sectors of marine life, the endurance of roads and trade routes, and the surprising unchangeability of the geographical boundaries of civilizations’ – that they shape.⁷⁰ For Foucault, the patterns of human activity to be investigated are made of words: Foucault’s structures are formed by series of utterances, inscriptions, texts – of discursive events.

The ‘event’ in ‘discursive event’ is important. Foucault sets his sights not merely on what might (according to the rules of grammar or logic) have been written or said, but rather on what was actually written or said – at a particular moment, by a particular historical actor using a particular conceptual vocabulary, in a particular format, and via a particular form of publication. As suggested, however, it is not single events but rather series of them that are of interest. And just as any historical set of events can form a series, so discursive events, in the fact of their being said or written (when other linguistic sequences could have been produced, but were not), can form a series, too. Likewise, just as the series that members of the Annales School investigated have their own underlying patterns and rules of production and accumulation, so, too, will the category of series formed by discursive events: namely, a discursive regularity.⁷¹

What Foucault’s notion of discursive regularities provides historians of thought, then, is an excellent set of tools to examine discursive landscapes from the perspective of the longue durée. It is precisely in this landscape that, as we saw, Parmenides’ great innovation is located – and also where his relationship to Homer’s Odyssey must be excavated. We can therefore restate Mourelatos’s premise – ‘the image of the route mediates a new concept of the nature of thinking and knowing’⁷² – with a new level of specificity and insight: the discursive regularities (explored in chapters 3, 4, 5, and 6) that link Parmenides to Homer mediate the transition from the Odyssey’s narration of human movement through physical space in time to Parmenides’ path-breaking deductive argumentation (movement through logical space in discursive sequence) and move towards demonstration. Even more specifically, and to preview one of my primary claims here in full: Circe’s hodos lays before our eyes a blueprint of the discursive architecture that Parmenides used to build the first attested sequence of extended deductive argumentation in Western thought.

Aims: What Is and Is Not at Stake

Above, I emphasized the importance of reading Parmenides’ poem as a poem, not merely an argument; this is particularly important, I suggested, where the relationship between Parmenides’ poem and Homer’s Odyssey, particularly book 12, is considered. This might imply that I intend to proceed according to the rules of intertextuality as normally understood: namely, line up two bits of text; show, via distinctive features common to both, that there is a high probability that the later text interacts with the former; and then tell a good story about how part of the second text’s meaning is generated as a result of this interaction.⁷³ Inevitably, some version of ‘lining up the texts’ will indeed occupy much of what follows, and I shall discuss in a number of places the points of overlap between Odyssey 12 and Parmenides’ poem that are sufficiently marked to justify the exercise.⁷⁴ This procedure remains an invaluable component of sound textual analysis in my view; indeed, it is worth emphasizing that the fundamental observation that prompted the current study is the deep but hitherto unobserved set of similarities between Odyssey 12 and Parmenides’ ‘Route to Truth’, and that these similarities remain the starting point, and the anchor, for all that follows below.

Intertextuality takes many forms, however, and can be evaluated from many perspectives. The crucial difference between this endeavour and most literary criticism now practised in Classics is that what the two poems under consideration here share most of all is a discursive architecture, a similar manner of structuring different units of text. That is to say that the intertextuality between Parmenides’ poem and Odyssey 12 does not so much generate meaning in the former text (though it may also do this at times) as provide a framework or structure for its shape at a variety of different levels. It is for this reason that I referred above to the ‘discursive blueprint’ that Odyssey 12 offers Parmenides, and it is for this reason that the toolkit offered us by Foucault’s Archaeology of Knowledge is so valuable.⁷⁵ It is because my aim is to confront this last relic of the Greek Miracle – the genius Parmenides indebted to no one for his invention of extended deductive argumentation – that I have given such prominence to Foucauldian archaeology.⁷⁶

In Chapter 2 I argue that, since Parmenides is operating within the same cultural and poetic milieu as his late archaic comrades in verse, we should approach his poem with the same general assumptions about late archaic receptions of Homer that we bring to bear on his fellow poets. I therefore assume that Parmenides is interacting directly with a Homeric text that is relatively fixed, and that this text is largely similar to the one that has come down to us. I have adopted this position partly for convenience, since doing so enables me to ‘line up the texts’ and compare their discursive architecture and other features in the most concrete fashion. Incidentally, I also take the view that this assumption is in fact correct, a point I shall touch on again at the beginning of Chapter 2, where I discuss late archaic receptions of Homer in greater detail. It does not seem to me, however, that the core thesis for which I argue below would be much damaged should one adopt a different perspective on any number of Homeric questions. Provided that one’s view of the process of Homeric textualization or canonization still allows one coherently to discuss, for example, the A-B-C pattern, or the notion of catalogic discourse in Homer, there is ample scope to discuss the possibility of a similar discursive phenomenon associated with narrating the itinerary of a hodos.

If the markedly close correspondences between Odyssey 12 and Parmenides’ proem and the ‘Route to Truth’ allow us to posit an intertextual relationship between the two texts, there is no need to commit to a more specific characterization of this intertextuality. Whether this intertextuality is ‘deliberate’, whether Parmenides’ ‘Route to Truth’ is part of a larger discursive regularity involving not only Homer, but an entire body of now-vanished poems portions of which were organized by the figure of the hodos (or whether both are or can be true!), are questions about which I remain agnostic.⁷⁷ What matters is that the texts are so similar in the way intimated above and analysed below. I submit that the primary discussion that follows in chapters 3–6 stands up just as well whether one chooses to see these similarities as emerging organically out of a thought culture for which Homer is our best witness or as the product of deliberate invocation of Homer – or indeed to see them as anything in between.⁷⁸ In every case, what remains true is that, once one accepts the discursive similarities between Homer and Parmenides, the latter is no longer a ‘creative genius in debt to nobody’.

These questions about the relationship between Homer and Parmenides having been addressed, it is important to take a step back. By staking out this field (Parmenides’ poem, along with the necessary context: physical, linguistic, cultural, and, above all, poetic and discursive), this method (Foucauldian archaeology, supplemented both by more traditional literary criticism and by attending to discussions of Parmenides’ arguments), and, most of all, this strictly delimited aim (explaining Parmenides’ invention of the outline of demonstration and the practice of extended deductive argumentation), my intention is to avoid a number of other possible issues. Despite my insistence on the importance of reading Parmenides’ poem as a poem, it is not my goal to examine Parmenides’ relationship to the larger hexameter tradition or the rich world of archaic poetry as a whole.⁷⁹ While I shall indeed conduct a strategically targeted survey of these topics in Chapter 2, because my principal goal is to provide an account of Parmenides’ invention of extended deductive argumentation and the practice of demonstration, the main task is to identify and articulate the ties that bind the extended deductive argument and characteristic moves of demonstration that Parmenides makes in fragments 2, 6, 7, and especially 8, specifically to his time, place, and linguistic and poetic milieu. This is not, of course, to deny or devalue the connections between Parmenides and other predecessors in hexameter verse, notably Hesiod and notably in the proem;⁸⁰ rather, these simply do not have a great bearing on a discussion of Parmenides’ invention of extended deductive argumentation. Similarly, my interest in siting Parmenides within the world in which he lived and, especially, in relation to his poetic predecessors, means that, while I shall make some strategic comparisons between Parmenides and his poetic contemporaries – especially Pindar⁸¹ – in Chapter 2, I shall not attempt to examine these relationships in a comprehensive way. Illuminating and valuable though such a project would be, it is not clear this would shed much light on Parmenides’ use of extended deductive argumentation.

A similar point may also be made regarding the tradition of reading Parmenides’ poem against the backdrop of ritual, mystic, mantic, or other religious texts and contexts. Attempts to reconsider Parmenides in his sociocultural context or to attend to the poetic texture of his language have often come from scholars who have searched for evidence to support readings in this vein.⁸² However stimulating these discussions may be in their own right, however, they too have little bearing on the task of accounting for Parmenides’ invention of extended deductive argumentation. This is partly because much of this branch of scholarship focuses on the proem, and is much less convincing when it moves beyond this, especially to fragments 2–8, the main focus of my analysis.⁸³

While this line of thinking does little to illuminate Parmenides’ invention of extended deductive argumentation in its own right, the two strands of scholarship are neither necessarily hostile nor incompatible. As has recently been argued, accepting the notion that Parmenides’ poem represents, or is the product of, a divine revelation, or is otherwise tied to mystic rituals, does not preclude an interest in the rigour or origins of his argumentation.⁸⁴ In short, however rich this vein of research is, it operates at a tangent to the current inquiry into the emergence of extended deductive argumentation and the practice of demonstration.⁸⁵

On another note, despite my insistence on the value of Foucauldian archaeology to the endeavour at hand, I do not claim to have delineated any kind of larger archaic Greek discursive regularity or regularities per se. It is tempting, of course, to consider how the topics discussed below might constitute some part of such a thing, and the discussion in Chapter 3 of A-B-C patterns and catalogic discourse, for example, gestures towards what part of a hypothetical discursive regularity of this sort might look like; likewise, the common features shared by the two hodoi described in Odyssey 10 and 12 offer us enticing grounds for speculation. The overwhelming absence of other texts from this period, however, prohibits us from going further.

Comments of a similar sort might also be made regarding the so-called Doxa portion of Parmenides’ poem. Much of the most exciting recent scholarship on Parmenides has involved reconsidering the old, vexed question about the relationship between Doxa and the ‘Route to Truth’.⁸⁶ These discussions of Doxa have certainly given us a more robust understanding of what Parmenides hoped to accomplish in his poem, and they are an important step forward. Be that as it may, the question of Parmenides’ invention of extended deductive argumentation and the outline of demonstration are not, so far as we can tell, immediately connected to the Doxa section of his poem. As a result, the only occasion to discuss it will come in the final section of this book (Part III: Doxai), a deliberate non-conclusion that offers more general reflections on the Doxa section’s relationship to the ‘Route to Truth’, particularly in the light of the Homeric analysis developed here.

If it is not my goal to provide an exhaustive view of Parmenides in relation to his poetic or religious context, neither will it be my concern to advance my own specific interpretation of Parmenides’ arguments,⁸⁷ less still to stake out a view on what precisely Parmenides’ larger philosophical positions are. (Though in Chapter 6 I shall examine how the view advanced in the pages below might square with various interpretations of Parmenides’ arguments presented by others, and what new light the account offered here can shed on these interpretations.) By the same token, however, I do claim that those who in the future wish to offer specific interpretations of Parmenides’ arguments will need to explain how their interpretations can be reconciled with the analysis undertaken in this project. The point is not categorically to deny that a given thinker, on account of thinking from within a specific tradition, is able to argue in a specific way or to make specific arguments (especially when that thinker is as radical and innovative as Parmenides).⁸⁸ But no such categorical denial need be presumed here; if some readers will insist that form cannot determine content, we must equally insist that form does necessarily shape the matrix of possibilities for content in a distinctive way. To conclude: if the domain explored in this project is not deemed prior to philosophical analysis of Parmenides’ arguments, neither should philosophical analyses of Parmenides’ arguments take rigorous priority over considering the argumentative form in which they are expressed. That is, should the claims advanced in this book be found persuasive, they would need to be borne in mind as a crucial set of factors for scholars to use in formulating their understanding of Parmenides’ arguments. From this, it also follows that the findings presented here ought to serve as one of the main criteria by which the strengths and weaknesses of interpretations of Parmenides can be assessed.

One final observation: I do not actually get down to the nuts and bolts – the particles, the modally charged negations, the aspects and tenses – of Parmenides’ text until Chapter 5, halfway through the ‘Routes’ portion of the book. In structuring my overarching argument this way, and in the manner in which I have elected to style the book’s larger programme and Table of Contents, I have assumed a relatively high degree of familiarity with Parmenides’ poem on the part of the reader; without this, the relevance and importance of the material discussed in chapters 1, 2, and especially 3 and 4, to the problem at hand will be less clear. This strategy is not without its risks. Parmenides is hardly a ubiquitous presence in the contemporary Classics curriculum, and proceeding on this assumption may induce some frustration in a portion of my potential audience. Nevertheless, I hope that scholars of the archaic reception of Homer, and of Homer himself, will find material of value in Chapter 2, and in chapters 3 and 4, respectively; likewise, I hope that all who have occasion to consider ancient Greek roads and their associated lexicon will find something useful in Chapter 1. On the other hand, I have faced a challenge of the reverse nature in writing Chapter 2, where my goal is to bring the discussion of Parmenides into contact with recent advances in the study of the archaic reception of Homer; here I have tried to make the discussion rich enough to be fruitful for scholars of Parmenides without being tiresome for scholars in the field of literature. This proved a delicate balancing act; in view of the risks and rewards of writing for different – and sometimes rather distant – subfields of the discipline, I ask forbearance from readers who would have charted the hodos of argument otherwise.

***

These, then, are the stakes. From one perspective, the scale of this project might be deemed enormously ambitious: to trace the origins and early evolution of extended deductive argumentation and the practice of demonstration, thereby delineating a key portion of the genealogy of the Western conception of knowledge. From another, however, the domain of inquiry is narrow and its epistemic stance humble: this is simply an attempt to read a poem with attention to the richness of its language and imagery, in relation to its cultural context, and alongside its poetic predecessors – no more and no less than what any poem deserves. To perform an archaeological excavation of this buried hodos and recover the first instalment of this invention of the concept of method – a μῦθος ὁδοῖο, if ever there was one – we must rethink and re-examine the methods of our hodos and the hodos of method.