Socratic Methodology and the Problem of the Elenchus

Originally written for PHIL 1110B Plato with Prof. Justin Broackes.

The principal form of Socrates’ philosophical method is that of the elenchus, where a thesis put forward by an interlocutor of Socrates is reduced to contradiction. It seems, however, that such a method can only generate negative conclusions, despite it also seeming that Socrates wants to have warranted positive conclusions as well. Reconciling these ideas is Gregory Vlastos’ “problem of the elenchus.” Here, I will endeavor to vindicate Vlastos’ original presentation of the problem. I will discuss responses both that the elenctic form is not structurally representative of Socrates, and against responses that the problem is theoretically intractable, ultimately concluding that the problem is theoretically tractable and that the additional theoretical framework necessary to allow for positive conclusions from elenchi is independently justified.

In the Gorgias, Socrates expounds positive moral views with full force. Socrates contends that happiness consists in justice and education (470e), that the pleasurable and the good are distinct and the pursuit of the latter should be primary (494e-500a), and the central and vociferously proclaimed idea that it is better to suffer injustice than to commit it. There is, however, some tension between Socrates’ moralism and his philosophical methodology. The principal Socratic method, the elenchus, primarily consists in identifying inconsistency and hypocrisy in the interlocutor.

Socrates was so focused on the elimination of inconsistency that the Victorian classicist George Grote took the elenchus to be a solely negative method. Such a position is untenable, however, in light of the existence of Socrates’ positive moral program, unless with Grote one believes that “Socrates’ own positive convictions and his critical assaults on those of others ran on separate tracks” (Vlastos 18.) Such a radical discontinuity between the positive and negative components of Socratic philosophy seems too ugly to bear. Indeed, the fact that the putative conclusion of many elenchi are in fact Socrates’ positive moral positions (or rather, the claims that the elenchus finds inconsistency in are the negation of such positions) is clear evidence that Socrates takes his method to be establishing his positive philosophy, rather than being purely negative. Even more conclusive on this issue is a crucial quote from Socrates himself. He asks Polus, after one of the elenchi, “Hasn’t it been proved that what was said is true?” (479e) This quote implies not only that the elenchus is a positive method, but that it is a method of proof, that via the elenchus Socrates’ moral claims are established deductively.

This raises a significant question, however, first identified by Gregory Vlastos as “the problem of the elenchus.” A straightforward analysis of the elenctic method seems as though it proceeds by beginning with a claim put forth by Socrates’ interlocutor, adds to this claim some premises that both Socrates and his interlocutor agree on, and demonstrates that this body together leads to contradiction. The logical conclusion of such a structure, then, is not that the original claim itself must be false, but merely that the full body of premises together is inconsistent. That is, the claim that Socrates disagrees with deserves no special status as having been disproved by the elenchus, rather it is equally possible that one of the auxiliary premises that Socrates would endorse is at fault instead. It is also not available to Socrates to justify these premises externally. It cannot be, for example, that these premises can’t be discarded because they are self-evident, since self-evidence is mentioned nowhere in Plato and has no clear role in Socratic epistemology. Even less available is the idea that the widespread acceptance itself (the property that Socrates relies on when introducing them as premises, anticipating agreement from the interlocutor) is what justifies them. When Polus asks “Don’t you think you’ve been refuted already, Socrates, when you’re saying things the likes of which no human being would maintain?” (473e), Socrates vigorously disagrees, contending that widespread opinion is worthless and he aims for true propositions rather than popular ones.

To see the problem more clearly, let us trace through Vlastos’ example of a paradigmatic elenchus (Vlastos 19-22.) Socrates begins by establishing Polus’ thesis as p: To commit injustice is better than to suffer it. Then, Socrates gets Polus’ assent to a number of auxiliary premises, one of which Vlastos separates out as q: “To commit injustice is baser than to suffer it” and the bundle of other auxiliaries as r. Then, Socrates proceeds with the elenchus, showing that q and r together imply not-p, and so that p, q, and r are mutually inconsistent. From this, he concludes that not-p is true. Vlastos rightly notes, however, that accepting not-p is not Polus’ only possible response, and it is also available for example to reject q. Indeed, frequently in the course of an elenchus the original thesis p is as deeply held for the interlocutor as any of the auxiliary premises. This explains the typical reaction to someone argumentatively bested by Socrates in this way not of being immediately convinced of the truth of the conclusion, but rather being disgruntled and confused overall.

Vlastos’ interpretation of Socrates is that he is not worried about the option of discarding one of the auxiliary premises, since it will always be possible to perform the elenctic reduction to contradiction as long as the faulty thesis remains. In part, this idea is demonstrated by the elenchus Socrates performs immediately following the one just discussed, wherein the same conclusion about injustice is reached from different premises against Callicles. In this way, although Plato does not demonstrate the procedure of someone directly rejecting a particular auxiliary assumption introduced by Socrates instead of their original thesis, he comes close by replacing Polus with a different interlocutor who does. Although he does not make the claim directly, Vlastos takes Socrates’ positive moral assertion as partially consisting in the dialectical claim that any interlocutor who maintains the thesis will be defeated in this way.

As evidence, Vlastos cites the following portion of Socrates’ concluding speech to Callicles: “So, either refute her [philosophy] and show that doing what’s unjust without paying what is due for it is not the ultimate of all evils, as I just now was saying it is, or else, if you leave this unrefuted, then by the Dog, the god of the Egyptians, Callicles will not agree with you, Callicles, but will be dissonant with you all your life long” (482b.) In other words, the presence of the wrong moral belief in Callicles carries with it a contradiction—that Socrates will always be able to uncover via elenchus—that will over the course of his life cause Callicles internal dissonance. The positive moralism and negative contradiction-elimination are thereby united; a crucial component of Socrates’ object level moral claim is the meta level dialectical claim that anyone holding the wrong moral belief will always have an internal contradiction and be subject to an elenchus. Thus, Vlastos takes Socrates to hold the “meta-elenctic” claim that “whoever has a false moral belief will always have at the same time true beliefs entailing the negation of that false belief” (Vlastos 25) which he calls proposition [A].

The idea, then, is that any time someone puts forward a false moral belief as a thesis (from the perspective of Socrates, a thesis that Socrates does not agree with) it will be possible to find some true moral premises that they believe (from the perspective of Socrates, premises that Socrates does agree with) that are inconsistent with that thesis. In this way, statement [A] is equivalent to the statement that any false moral thesis can be reduced to contradiction via the method of an elenchus. Socrates does not argue for [A], but for Vlastos it is implicit in his methodology. It is the very principle necessary to unite his positive and negative philosophies. If Socrates also believes that his own moral beliefs are internally consistent, which Vlastos terms proposition [B], then he can be confident that in fact his body of moral beliefs is in fact wholly true. This is because, straightforwardly, if Socrates had some false moral belief then by [A] his whole body of moral beliefs would be inconsistent, which is exactly what [B] denies. Vlastos thinks further that [A] and [B] are supported through dialectical inductive evidence. Socrates comes to hold [A] because every time he comes upon an interlocutor with a moral thesis he disagrees with and attempts elenctic reduction, he is successful. Thus, the more times Socrates performs successful elenchi, the more confident he can be that wrong moral beliefs and susceptibility to an elenchus necessarily go hand in hand. Likewise, Socrates can come to be confident in [B] since he has an open challenge for others to reduce his own beliefs to contradiction via an elenchus. Since no one has done so, Socrates can thus be confident that it is not possible to do so, meaning that there is no contradiction among his body of moral beliefs. These justifications are shaky at best, but in establishing these meta-elenctic principles, Vlastos believes he has solved his problem of the elenchus.

Now, I will consider some attempts at dissolving or diminishing the problem of the elenchus, before attempting to vindicate Vlastos’ original formulation. In particular, I will attempt to provide theoretical, rather than inductive, justification for the meta-elenctic principles Vlastos ascribes to Socrates. First, let us consider some dissolutionary responses in which challenges are made to Vlastos’ background exegesis that frames the problem, and then we move to more straightforward responses that challenge Vlastos’ [A] and [B].

For Benson, the problem of the elenchus is intractable. In other words, “if Vlastos is right concerning ‘the problem of the elenchus’ , then it cannot be solved” (Benson 1987 67.) I will return to his justification for this position momentarily, but first one must note that it goes hand in hand with a denial of the antecedent. That is, given his position on the intractability of the problem it should be no surprise that Benson thinks Vlastos’ formulation of the problem is incorrect. To establish this claim, Benson provides bountiful textual evidence from dialogues that he thinks are more paradigmatic of the Socratic method than the Gorgias (namely, the Euthyphro, the Charmides, the Laches, and the Apology.) For Benson, the elenchus as demonstrated in these texts is purely negative after all, and Socrates does not take himself as establishing his positive moral positions using it. Insofar as he does have positive moral positions, Benson takes them as being established by other means, and as secondary to the negative component of his philosophy. Benson notes that “Socrates believes that his wisdom lies in his awareness of his ignorance, not in any knowledge of positive moral doctrines his methods have permitted him to establish” (Benson 1995 111.)

There are a few important things to note against such a line of reasoning. Firstly, it is unsurprising that the paradigmatic elenchus is not taken to establish positive moral principles. Even on Vlastos’ view, the straightforward and primary purpose of the elenchus is negative—only after the introduction of his dialectical principles can this negative reduction be seen also to provide support for Socrates’ positive views. The idea that a separate methodological approach is taken between the positive and negative methods resembles the early attitude of Grote we saw before. Such a position, rather than putting the negative component of Socrates’ philosophy first, as the quote about wisdom seems to indicate, instead severs the positive moral positions from this source of wisdom. In other words, casting Socrates as having a unified method with a principally negative character (that is, the elenchus) in fact allows him to generate positive moral views without sacrificing the negative character of his wisdom. There would be little reason to trust positive views generated by some secondary philosophical method if they were severed from the principal negative elenchus.

Benson’s response resembles that of Brickhouse and Smith, which denies the existence of the elenchus altogether. For Brickhouse and Smith, Socratic thought is characterized by examination and reasoning applied to all corners of life. The principal element, on this view, is not structural or methodological but rather motivational. They elect not “to suppose that “the examined life” has to follow some ‘unique form of argument’ to be worth living, nor do we have to suppose that the reasons we should follow when we examine our lives must always flow from a single form of reasoning” (Brickhouse 155.) Socrates’ arguments have certain common flavors, often refutative, hortative, ethical, and so on, but they do not follow a single structure. In this way, Socrates has no “special advantage” (Brickhouse 156) when it comes to ethical reasoning, but merely a dedicated commitment to it.

A response along these lines significantly discredits Socrates. Firstly, Socrates obviously does not need every work he speaks in the dialogue to find some place in an elenchus in order for the elenchus to be his principal structure of philosophical engagement. The topics he discusses can be various, and his commentary heterogeneous and rich, even when the underlying guiding form is that of the elenchus. Philosophy that conforms to a unified structure is more beautiful and coherent, not more lifeless and robotic, in virtue of that fact. Secondly, the same error is committed here as in Benson’s account, where positive moral positions are put in doubt as coming from some place of “special advantage” when they flow from the elenchus. The idea, I think, is that by giving Socrates’ principal argumentative form the power to generate positive moral commitments, its negative character is thereby poisoned and he becomes at heart, a sort of supreme moralist. The reverse, in fact, is true. Positive moral commitments that ultimately flow from the examination and elimination of contradiction allows crude moralism to be replaced by sophisticated method. This is to say nothing, of course, of the theoretical beauty and coherence Socratic philosophy overall takes on when it principally inhabits a single form. Rather than merely recommending a dedication to reason and self-examination, as Brickhouse and Smith suggest, Socratic philosophy also provides a template through which to pursue that goal. Replacing the Socratic method with a mere Socratic dedication transforms the thought from the building block of all philosophical inquiry into a vague recommendation. We have good reason, therefore, to stick with an interpretational framework that ascribes unity to Socrates’ philosophical method, and therefore places a great deal of importance on the elenchus.

Let us turn, then, to responses that take the problem of the elenchus on its own interpretational terms and challenge it on theoretical grounds. First, Benson presents a highly formalized version of the argument quickly sketched above that the elenchus can only demonstrate inconsistency (what Benson calls the “non-constructivist” position) rather than conclusively show the negation of the interlocutor’s central moral thesis (the “constructivist” position.) The argument, roughly, notes that if the argument’s conclusion is going to be the negation of the central thesis p, then p must enjoy some special alethic status among all the premises p, q, r, and so on. In other words, there must be some reason to preserve the auxiliary premises that, combined with their mutual inconsistency with the moral thesis, thereby allows one to draw the negation of the thesis as a conclusion. Even more demandingly, that special status has to be available to the interlocutors in order for the conclusion to properly be drawn, named by Benson the “availability constraint” (Benson 1987 70.) In other words, then, it is not possible simply to let the alethic property be “truth” (which, on Socrates’ view, is indeed a special property that the auxiliary propositions and not the central thesis share) since the truth of the relevant propositions cannot reasonably be expected to be available to Socrates and the interlocutor independently of the argument. As mentioned before, obvious candidates for such a special status such as self-evidence or widespread belief do not seem to be available to Socrates. Benson’s argument rules out any possible candidates by contending that the only thing that must be available to all participants in the elenchus in order for it to go through is that all the interlocutor believes all the premises, what he calls the “doxastic constraint” (Benson 1987 71.) Since that belief is shared among all the premises and not specially granted just to the auxiliaries, it cannot be the special alethic status necessary. Thus, nothing can.

The complete structure of Benson’s argument is not worth going through, since he essentially misses the point. In Benson’s portrayal of the constructivist position, the elenchus is meant to directly establish the positive conclusion consisting in the negation of the central thesis. This, however, is not the correct picture. The elenchus, even for Vlastos, is still principally a negative tool. The only direct result of the elenchus is the establishment of inconsistency in the positions of the interlocutor. However, when coupled with the theoretical background linking inconsistency and moral incorrectness (that is, Vlastos’ [A]), this indirectly demonstrates the positive conclusion. The special property held by the central thesis is that it is not part of Socrates’ own belief set, which is relevant because of the consistency guaranteed by principle [B]. Thus, the principal theoretical task remains vindicating [A] and [B], rather than countering Benson by establishing a direct positive function for the elenchus.

In Kraut’s original response to Vlastos, which appeared in the same publication issue as the original article, he puts some pressure on these crucial meta-elenctic propositions from a different direction. For Kraut, the elenchus can in fact produce positive moral results without the additional theoretical infrastructure, that is, “the elenchus can yield proofs without the benefit of assumptions A and B, and I doubt that Socrates relies on these two principles to reach the conclusion that all of his moral beliefs” (Kraut 59.) Two major issues crop up in Kraut’s response. Firstly, unlike Vlastos, Kraut’s attribution of positive power to the naked elenchus is susceptible to the refutation given by Benson. There will be a serious onus on him to come up with a special property shared only by the auxiliary premises that allows the conclusion to reject only the central moral thesis. In effect, he tries to get around this problem by noting that “one can’t always give a reason for everything one believes, and this fact does not deprive one of proof” (Kraut 62.) In other words, then, the elenchus is positive insofar as it is expansionary; it can build up rejections of specific theses from a body of already established moral grounds. Still, as Kraut admits, this line of thought applies equally well to question-begging arguments. Moreover, it does not overturn Benson’s objection.

Secondly, and more germanely, the particular comments Kraut makes about Vlastos’ theoretical principles largely mistakes them. Kraut thinks that Vlastos’ interpretation of Socrates’ claim to always be able to elenctically reduce Callicles to inconsistency as long as he holds on to his central moral thesis is a “wildly overconfident claim” and wants to backtrack to “the less arrogant claim that, given a certain fixity in human beliefs, he can always find contradictions in his opponents” (Kraut 67.) This misunderstands the claim as being about Socrates’ own abilities, rather than the encapsulation of principle [A] that it is. In other words, the idea is not that Socrates could always find the proper elenchus to discover the contradiction, but rather that the contradiction must always simply exist. As another counterpoint to principle [A], Kraut notes that the inductive justification endowed on it by successful past elenchi “presupposes that Sen-rates has sorted out the true from the false: whatever confidence he has in his ability to recognize which beliefs are true is therefore independent of and prior to his belief in A” (Kraut 68.) This, too, subtly mistakes the theoretical role of [A]. Socrates is not explicitly relying on [A] before embarking on any elenchi. In order to motivate an elenchus in practice, Socrates of course must rely on his moral intuitions. However, if [A] fails in some instance, then so too will the elenchus associated with that instance. In this way, [A] is not necessary to get the elenctic project up and running from the beginning (although at first it will have a purely negative character), but as it develops one gains more and more confidence in [A], and the positive character thereby develops. Finally, against [B] Vlastos provides some textual examples of Socrates being worried about inconsistencies among his own moral beliefs. This, too, is a non-issue. Socrates may well be worried about making logical errors, incorrectly performing an elenchus by claiming logical consequence between propositions when none holds. Socrates is, after all, a human. This does not undermine the philosophical structure of the project. One does not have to be absolutely certain about the principles in order for the elenchus to be performed successfully. Knowledge is not completely required. Instead, it simply has to be the case that the full theoretical framework guarantees a successful elenchus in ideal conditions. In this way, any mistakes are simple human errors rather than deep philosophical mistakes.

Having dissipated the theoretical concerns in the literature, then, I would like to provide some positive reason for believing in Vlastos’ theoretical framework consisting in [A] and [B]. First, to be clear, these are not exegetical reasons for reasoning out why Socrates implicitly accepts these principles. Indeed, it is not clear to me to what degree Socrates was aware of the full theoretical framework his methodology implicitly calls upon. That is irrelevant, however, to a vindication of the elenchus as a general method. All we need are good philosophical reasons to be satisfied with the framework. First, Vlastos’ original inductive justification does hold up. Surviving more and more negative challenges should begin to grow our confidence that a body is impervious to them. Although this does not give us knowledge or certainty, as discussed above that is not a necessary criterion. There is also, however, independent theoretical reason to think that moral beliefs work in the way described by [A]. Ultimately, our moral beliefs are profoundly interconnected and justified by deep moral intuitions. We do not come to have particular individual moral beliefs in isolation. That is, since particular moral beliefs flow from conceptions of moral values they will be networked and interdependent. Any particular moral belief will have many siblings and cousins. In order for the full network, then, to be wrong but consistent, it will need to be profoundly wrong all over the place. One would have to have a badly broken conception of moral value that generates many moral beliefs that are all wrong in the same way. This would need to continue to the core—there is good reason to think that a consistent but wrong moral network of belief will need to also commit moral errors about fundamental, simple questions. There is good practical reason to think that an interlocutor is not like this, then, since from this reasoning we can see that a wrong but consistent moral thinker would need to be a near-alien. This thought gives theoretical reason to doubt its possibility as well, however, as long as intuition plays a role in moral epistemology. A moral alien of this ilk would have to have the problem infect his intuition, and thus his core rational capabilities, as well. This is implausible. The overall picture of our body of moral knowledge building out from easy, well-agreed upon base claims to complicated, nuanced issues makes it very unlikely that an error could possible infect the entire network in a way contradictory to [A].

In sum, then, I hope to have shown that Vlastos’ original formulation of the problem of the elenchus was correct, both in the problem and in the solution. The elenchus is indeed the principal structural tool for Socratic philosophy and it is indeed principally negative in character. Approaches which attempt to dissolve the elenchus itself or relegate its importance to only some section of Socratic philosophy are theoretically disuniting and do not do the Socratic approach justice. Thus, the burden of reconciling positive Socratic moralism with these two facts is real. Moreover, the burden disappears in light of a connection between coherence and truth encapsulated in Vlastos’ principle [A] as well as a confidence in the coherence of Socrates’ own positions encapsulated in Vlastos’ principle [B]. There are independent reasons to be satisfied with these principles and good reason to disbelieve theoretical challenges to them, and so the problem of the elenchus is thereby also solved.

Works Cited

Benson, Hugh H. (1987). “The Problem of the Elenchus Reconsidered.” Ancient Philosophy 7: 67–85.

Benson, Hugh H. (1995). “The Dissolution of the Problem of the Elenchus.” Oxford Studies in Ancient Philosophy 13: 45–112.

Brickhouse, T., and N. Smith (2002). “The Socratic Elenchos?” In Does Socrates Have a Method? Rethinking the Elenchus in Plato’s Dialogues and Beyond, ed. G. Scott, pp. 145–158. University Park, PA: Penn State University Press. https://doi.org/10.1515/9780271032214-012

Kraut, Richard (1983). “Comments on Gregory Vlastos, ‘The Socratic Elenchus.’” Oxford Studies in Ancient Philosophy 1: 59–70.

Plato (1987). Gorgias. Trans. Donald J. Zeyl. Indianapolis: Hackett.

Vlastos, Gregory (1993). “The Socratic Elenchus: Method Is All.” In Socratic Studies, ed. M. Burnyeat, pp. 1–38. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511518515.002