Abstract
Parmenides marks a watershed in the history of argumentation, presenting the earliest surviving sequence of recognizably deductive reasoning in the Greek tradition. This chapter focuses on the central section of his poem (fr. 8 DK) and examines the form of its argumentation: its use of indirect proof, the articulation of its reasoning, and the role necessity plays in it.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
- 2.
Stretching back to Parmenides’ own day: see [15].
- 3.
Taking the infinitive νοῆσαι as final, not epexegetic: two roads “leading to Noesis”, not “as objects of Noesis”; see [16, pp. 63–73] at p. 71, citing [20, §1969] and Empedocles fr. 3.12 πόρος ἐστὶ νοῆσαι; as often, purpose shades into result, cf. [20, §§2004–5 and §2011a]. The goddess introduces a third road in fr. 6, so these first two cannot be the only roads “conceivable”, though they may be the only ones offering Noesis, as opposed to merely true belief; see [16, pp. 51–105]. The aorist also favors some sort of cognitive success rather than mere endeavor, hence knowledge or insight achieved rather than ongoing episodes of thinking about something.
- 4.
All translations are mine; they aim to be literal and neutral, but much remains controversial.
- 5.
- 6.
These formulations remain seriously indeterminate, especially for Road 2; but this indeterminacy is not something we need resolve here.
- 7.
Immediately, if Sextus, Adv. Math. 7.111 is accurate in citing fr. 7.2–6 continuously with 8.1–2; otherwise very shortly thereafter. The main objection to following Sextus is lexical: metrically the continuation requires μόνος (so Sextus) but Parmenides elsewhere uses epic μοῦνος (2.2 and probably 8.4); see [16, pp. 381–382]. A distinct question is whether the injunction to “decide by logos” applies to the impending arguments of fr. 8 or only the preceding critique in frs. 2–7, as “elenchus spoken by me” (aorist ῥηθέντα) has led some to conclude; so [16, pp. 108–109]. But aorist participles are often prospective (“after it has or will have been spoken”); and since any injunction is inherently prospective, and in this case a call for critical reasoning that requires meticulous analysis, listeners will reasonably expect the entire injunction, and thus the range of the elenchus, to extend in both directions, covering both frs. 2–7 and fr. 8, perhaps also the sequel extending through frs. 9–19 as well; cf. [6].
- 8.
The list is complicated by both textual and interpretive problems, including variants for every term in 8.4. I translate the eclectic text in [2], except for the last term, where I adopt “unwavering” (ἀτάλαντον, cf. Empedocles fr. 17.19 DK), over the widely preferred “complete” or “perfect” (τέλειον or τελεστόν); so [1, p. 176, n. 6], [19, p. 118], cf. [17]. Given the disputed sense of μουνογενές (uniform or unique?) I simply transliterate; cf. [3, p. 71]. In 8.3 I take ἐόν as conditional, modifying the understood subject of ἐστιν, i.e. any item on Road 1, which must adhere to GC as announced in fr. 2.3; cf. [7, p. 433] “insofar as it is”; alternatively, it could be pronominal, as in [2] and [16, p. 139, n. 3].
- 9.
The list presents a “programme” for what follows; see [13, pp. 76–77], [1, pp. 179–180], [16, pp. 137–140]. Their articulation correlates four signs in 8.3–4 with four stages of argument; but I see no need to join [5] and [16] in transposing 8.34–41 after 8.42–52. [11] assigns fourteen terms to six groups and analyzes the arguments accordingly; but his grouping disrupts the clear articulation of fr. 8’s argument into four parallel stages; see [16, pp. 143–144].
- 10.
Steps in the argument are labeled systematically both for simpler reference in what follows, and to indicate corresponding factors in the arguments for each of the four “signs”. Thus, “P” designates the initial step in each argument, each of which introduces the attribute in question for that “sign”; “Q” designates the initial rejoinder to each P (all marked by “since”); “R” indicates a step in the subsequent reasoning; and “C” designates a conclusion of that reasoning as indicated explicitly by a verbal connective. As we shall see, it is not always clear how these steps are related to one another, whether for example C is to be seen as a basis for Q, or equivalent to it or to P; or whether a particular R is to be seen as a basis for C or for Q, or for another R. But the pattern is remarkably consistent, and plainly significant; cf. [16, pp. 143–144] calling P the “demonstrandum”, Q its “subsidiary programme” (an initial “rationale” indicating the grounds for P to be supplied by R), and C the “principal conclusion” (restating P).
- 11.
Taking temporal ποτε with both verbs (obviously not with initial οὐδέ: “not ever”) and restrictively: not “was once upon a time (and may still be)” or “will be (as it always has been)” but rather “sometime was (and no longer is)” and “sometime will be (but not yet)” as in 8.20; cf. [16, pp. 141–142]. So construed the argument is framed by a fourfold disjunction that is exclusive and exhaustive: beings on Road 1 have either a period of being before not being (perishing), or one after not being (generated), or only being, or only not being; but the fourth has already been ruled out in frs. 2–7; and appeal here to GC (the third option) now rules out the first two, thereby removing “is not” for all time and leaving the third option, only being.
- 12.
Use of ἐπεί marks the clause as affirming the antecedent of an implied conditional. Stoic logic later called this a “paraconditional” (Diog. Laert. 7.71), presumably because it employs a conditional as an abbreviated argument (or strictly speaking, conflates the two, hence “para”), viz. their first indemonstrable, modus ponens.
- 13.
Formally, a rhetorical question Q challenges a claim P by assuming “if P, then Q” and inviting the rejection of Q; cf. Aristotle, Rhet. 3.18. So here the question R1 “what birth will you find?” assumes that anything generated (P) has a birth (Q) and invites the denial of any birth.
- 14.
Or taking the several terms of R4 in turn: if somehow, then “later or sooner” (8.10a); and if so, then “some need roused” it to do so then (8.9b); and if so, then it “starts from nothing” (8.10b), which contravenes R3.
- 15.
Adverbial ὅπως (8.9) is commonly read as a conjunction introducing indirect discourse: “it is not to be declared that (it) is not”; cf. the formulation in fr. 2.3. But as an adverb of manner it yields GC: “not to be declared is any way in which (it) is not.”
- 16.
Demonstrative οὕτως, commonly assigned inferential force here (“thus”), is better construed adverbially, modifying “is entirely” to specify the way in which any item on Road 1 “is entirely”—namely for all prior time, since ungenerated; so [1, p. 188], cf. [11, pp. 194–195]. Then modal χρεών serves to flag the conclusion as logically necessary; on this key term, see [12, pp. 277–278], and for its logical role here, sec. 1.3 below. My translation reflects its association in the poem with divine norms (esp. fr. 1.28, cf. 1.32, 2.5, 8.54); how this relates specifically to logical necessity is a complex question I address elsewhere.
- 17.
- 18.
The privative “undivided” does not occur in extant lines; the initial list apparently uses two terms to specify it (8.3): both “whole” (cf. 8.38) and “monogenous” (cf. n. 8 above).
- 19.
For the text of fr. 8.33, excising μὴ before ἐὸν δ᾽, see [16, p. 384].
- 20.
Initial οὐδέ in 8.23 (repeated in 8.24) supplies the requisite syntactic connective but leaves unstated any logical connection R1-2 has to P or Q in 8.22.
- 21.
Why χειρότερον rather than metrically equivalent βαιότερον (same position in 8.45) or the more common ἥσσων (with μᾶλλον in 8.48)? Perhaps to signal a wider scope for this sign, including qualitative “differentiation” as well as spatial “division”; cf. later use of the term “division” for methods of conceptual analysis, notably in Plato’s Eleatic dialogues.
- 22.
The goddess follows the same strategy in frs. 2–7: fr. 2 announces a pair of contrary options in parallel phrasing (Roads 1 and 2), one of which it promptly rejects; fr. 6 then introduces a third option (Road 3) that occupies the space between the initial pair; its elimination then leaves only the first, if the three alternatives are exhaustive—as they seem to be: being in every respect, not at all, and in only some respects.
- 23.
Whether literally, for example, by spatial continuity or volumetric repletion, or by some more abstract sort of unity or uniformity (cf. n. 21 above).
- 24.
Present πελάζει is commonly taken transiently as “draws near”; but the stative sense of its cognate adverb πέλας is plainly more apt for Road 1; cf. “neighbors” in [1, p. 211]. In epic usage, the verb indicates close or immediate proximity and appears almost always in the aorist, implying arrival; the only present indicates abstract inherence (Athena inflicting pains on Ares in Iliad 5.766). If the present has similar force here, where an aorist would undercut the very argument against change in which it appears, then the conditional “if R4, C” has solid footing.
- 25.
Proclus, On Euclid Elements 1.5 citing Eudemus (fr. 140 Stork) on Thales (A20 DK); fr. 8 has both ἰσοπαλές (8.44) and ἶσον (8.49) alongside ὁμόν (8.47).
- 26.
The single exception proves the rule: in 2.7, in what [23] calls the “governing deduction” that eliminates Road 2 and thereby underpins Road 1 and GC.
- 27.
- 28.
Likewise for its correlative ἀνώλεθρον (8.3): present infinitive (8.14), and noun (8.21, cf. 27). I count eleven verbal adjectives in the extant lines: six are plainly descriptive (πλαγκτὸν 6.6, ἀγένητον 8.3, ἄπαυστον alongside ἄναρχον 8.27, ἀτελεύτητον 8.32, ἀκρήτοιο 12.1, πολυπλάγκτων 16.1), one probably so (ἄπυστος 8.21, alongside ἀπέσβεσται), and four probably modal (ἀνυστόν 2.7, φατὸν and νοητόν 8.8, ἀνόητον 8.17). The first seven all refer to items on Road 1, and the four modal instances all occur in epistemic contexts describing what mortals can know or tell. The modality in question thus appears to be narrow and precise: twice deontic (in a single prohibition: “not to be said or pondered”, following “I won’t allow” in 8.7) and twice predictive (failure on Road 2).
- 29.
“Denial” here requires external negation like 8.23 (“it’s not the case it might”), not internal negation like “it might not occur.”
- 30.
Contrast the narrow scope of “contrary-to-fact” claims about particular situations present or past, expressed in the indicative with (ἄν or κε), and the broad scope and counterfactual force of generalizations expressed with subjunctive antecedents (and ἄν or κε; or optative with past tenses). In this light, the total absence of subjunctive conditionals (the only extant subjunctive, in 8.61, is purposive) may itself reflect an extensional focus.
- 31.
Personified in the first instance as dramatic figures or verbal agents, as Necessity here, and as the goddess Themis or “Right” in fr. 1.28; and even where “Right” is syntactically impersonal, as here, the name has special resonance, evoking the distinctive authority of Themis.
- 32.
- 33.
A key source of obscurity is the scope of the connectives ἐπεί and γάρ in R3-4: does R3 support all of R2, or only step 4, or only one of its contrary options; and does R4 support all of R3, or only step 4, or simply R2 or R1 by recapitulating intervening results? The ramification of R2 into three terms also obscures how its pair of negated conjuncts (steps 3 and 4) is meant to support R1, singly or jointly.
- 34.
- 35.
I thank the participants and organizers of the Berlin conference for lively discussion, likewise everyone at the 12th UNAM-UT Encuentro, Raymundo Morado for an illuminating response, Jonathan Beere for helpful comments, and especially Matt Evans for invigorating criticism and the stimulus of his own work on Parmenides.
Works Cited
Barnes, J. (1982). The presocratic philosophers. Routledge and Kegan Paul.
Coxon, A. H. (2009). The fragments of Parmenides: A critical text with introduction and translation, the ancient testimonia and a commentary, revised ed. R. McKirahan. Las Vegas: Parmenides Publishing (original edition, Van Gorcum, 1986).
Curd, P. (1998). The legacy of Parmenides. Princeton University Press.
Curd, P., & Graham, D. (Eds.). (2008). Oxford handbook of presocratic philosophy. Oxford University Press.
Ebert, T. (1989). Wo beginnt der Weg der Doxa? Eine Textumstellung im Fragment 8 des Parmenides. Phronesis, 34, 121–138.
Lesher, J. (1984). Parmenides’ critique of thinking: The poludêris elenchos of fragment 7. Oxford Studies in Ancient Philosophy, 2, 1–30.
Lesher, J. (2008). The humanizing of knowledge in presocratic thought. In [4], 458–84.
Lloyd, G. E. R. (1990). Demystifying mentalities. Cambridge University Press.
Lloyd, G. E. R. (2003). Demonstration and the idea of science. In J. Brunschwig, G. E. R. Lloyd, & P. Pellegrin (Eds.), The Greek pursuit of knowledge (C. Porter et al., Trans.), 92–117. Harvard University Press.
Long, A. A. (Ed.). (1999). Cambridge companion to early greek philosophy. Cambridge University Press.
McKirahan, R. (2008). Signs and arguments in Parmenides B8. In [4], 189–229.
Mourelatos, A. P. D. (1970). The route of Parmenides. Yale University Press; reprinted with additions Parmenides Publishing, 2008.
Owen, G. E. L. (1960). Eleatic questions. Classical Quarterly, 10: 84–102; reprinted in Owen, Logic, science and dialectic: Collected papers in Greek philosophy (Cornell University Press, 1986), 3–26.
Palmer, J. (1999). Plato’s reception of parmenides. Oxford University Press.
Palmer, J. (2008). Classical representations and uses of the presocratics. In [4], 530–554.
Palmer, J. (2009). Parmenides and presocratic philosophy. Oxford University Press.
Reilly, T. J. (1976). Parmenides, fragment 8.4: A correction. Archiv für Geschichte der Philosophie, 58, 57.
Schofield, M. (1983). Parmenides of Elea. In G. S. Kirk, J. E. Raven, & M. Schofield (Eds.), The presocratic philosophers, 239–262. Cambridge University Press.
Sedley, D. S. (1999). Parmenides and Melissus. In [10], 113–133.
Smyth, H. W. (1956). Greek grammar, revised ed. G. Messing. Harvard University Press.
Stork, P. (forthcoming). Eudemus of Rhodes: The sources, text and translation. In T. Dorandi & E. Spinelli (Eds.), Eudemus of Rhodes: Text, translation, and discussion. Routledge.
Vlastos, G. (1983). The Socratic Elenchus. Oxford studies in ancient philosophy 1: 27–58; revised in Vlastos, Socratic studies (Cambridge University Press, 1994), 1–37.
Wedin, M. (2014). Parmenides’ grand deduction: A logical reconstruction of the way of truth. Oxford University Press.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this chapter
Cite this chapter
White, S. (2021). Truth Attending Persuasion: Forms of Argumentation in Parmenides. In: Bjelde, J.A., Merry, D., Roser, C. (eds) Essays on Argumentation in Antiquity. Argumentation Library, vol 39. Springer, Cham. https://doi.org/10.1007/978-3-030-70817-7_1
Download citation
DOI: https://doi.org/10.1007/978-3-030-70817-7_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-70816-0
Online ISBN: 978-3-030-70817-7
eBook Packages: Religion and PhilosophyPhilosophy and Religion (R0)