Ancient Skepticism (Part 4)

4. Pyrrhonian Skepticism

4.1 Early Figures: Pyrrho and Timon

When comparing Pyrrhonian and Academic skepticism, two topics stand out: Pyrrhonism aims at tranquility; and it assigns pride of place to appearances. Anecdotes about Pyrrho’s life (365/60–275/70 BC) convey how unaffected he was (DL 9.61–69). This kind of ideal—a tranquil state of mind—is not part of Academic skepticism, but is essential to Pyrrhonism (Striker 2010; for a different view, cf. Machuca 2006; for a general account of early Pyrrhonism, cf. Castagnoli 2013; on Democritean influences on tranquility in Pyrrho and Timon, cf. Svavarsson 2013; on Sextus’ take on Pyrrhonian tranquility, cf. Svavarsson 2015). Insofar as the point of the anecdotes about Pyrrho’s life is that Pyrrho did not avoid or pursue anything with fervor, or that he did not despair about things that other people find terrible, they capture ideas that remain central to Pyrrhonism. In other respects, the anecdotes may not be trustworthy. They portray Pyrrho as a strikingly unconventional figure, unaffected not just by emotion and belief, but also by perception—to the extent that friends had to pull him off the street when a wagon approached (DL 9.62). Another anecdote makes Pyrrho appear not only unusual, but arguably a not so sympathetic character. Passing by a drowning man, he was so unmoved that he simply walked on (DL 9.63). This is in rather stark contrast with Stoic notions of unaffectedness, where the idea would be that, not being disturbed by emotions like fear and panic, the passerby is ideally equipped to help effectively. At the same time, Pyrrho seems to have said that the skeptic adheres to appearances (phainomena) (DL 9.106; Bett 2000, 84–93; on earlier notions of appearances relevant to skepticism, cf. Barney 1992). This might suggest that he would not cross the street when a wagon is approaching, and so appears to him. The biographical details are also suspicious because they mold Pyrrho’s life to fit the schema of the sage: a traveler to the East (Flintoff 1980), whose insights are conveyed in brief sayings; an enigmatic figure, exemplary and shocking at the same time.


Even though he discussed tranquility and adherence to appearances, Pyrrho was arguably no Pyrrhonian skeptic (Bett 2000, 14–62). That is, it is likely that he put forward a dogmatic position, in the sense that he had positive philosophical views about the character of reality. Pyrrho wrote nothing. Much of what we know about him is preserved through the writings of Timon, his adherent (325/20–235/30 BCE) (on Timon, cf. Burnyeat 1980b and Clayman 2009). The most important piece of testimony is a passage reporting an account by Timon:

It is necessary above all to consider our own knowledge; for if it is in our nature to know nothing, there is no need to inquire any further into other things. […] Pyrrho of Elis was also a powerful advocate of such a position. He himself has left nothing in writing; his pupil Timon, however, says that the person who is to be happy must look to these three points: first, what are things like by nature? second, in what way ought we to be disposed towards them? and finally, what will be the result for those who are so disposed? He [Timon] says that he [Pyrrho] reveals that things are equally indifferent and unstable and indeterminate (adiaphora kai astathmêta kai anepikrita); for this reason, neither our perceptions nor our beliefs tell the truth or lie (adoxastous kai aklineis kai akradantous). For this reason, then, we should not trust them, but should be without opinions and without inclinations and without wavering, saying about each single thing that it no more is than is not, or both is and is not, or neither is nor is not (ou mallon estin ê ouk estin ê kai esti kai ouk estin ê oute estin oute ouk estin). Timon says that the result for those who are so disposed will be first speechlessness (aphasia), but then freedom from worry (ataraxia); and Aenesidemus says pleasure. These, then, are the main points of what they say (Aristocles in Eusebius PE 14.18.1–5 = DC53; tr. Bett 2000 with changes)
In response to the first question, how things are in their nature, Pyrrho makes a metaphysical claim: they are indeterminate (Bett 2000, 14–29). There are no stable items, or no items with stable properties. Scholars sometimes hesitate to ascribe such a position to Pyrrho, because it is undoubtedly dogmatic. Perhaps the text can be given an epistemological reading: things are indifferentiable and unmeasurable and undecidable, because we fail in differentiating, measuring, and determining how they are (Thorsrud 2010, Svavarsson 2009). But Pyrrho’s response to the second question may only follow if we adopt the metaphysical reading (Bett 2000, 29–37). Pyrrho infers that our perceptions and beliefs are neither true nor false. They are not truth-evaluable, presumably because there are no facts which could be correctly captured. Third, if we understand these things, speechlessness (aphasia) follows, and then tranquility (ataraxia). Pyrrho does not say that we should cease to speak. He suggests that we adopt a complicated mode of speech, constructed around the expression ou mallon (“no more”), which aims to capture the indeterminate natures of things, when we attempt to say anything about anything (Bett 2000, 37–39).

4.2 Aenesidemus, the Ten Modes, and Appearances

Aenesidemus (first century BCE) was discontented with the views discussed in the Academy at his time, of which he began as an adherent. Philo’s proto-externalism as well as a counterposition formulated by Antiochus both appeared to him dogmatic. Aenesidemus aimed to revive a more radical skepticism, and left the Academy for this purpose. Arguably, he is the first Pyrrhonian skeptic. Aenesidemus wrote a treatise, the Pyrrhonian Discourses, probably similar in structure to Sextus’ Outlines of Pyrrhonism: a general account of skepticism, followed by books on particular philosophical questions (partially preserved in a summary by Photius; Hankinson 2010). The basic elements of Aenesidemus’ skepticism are: the skeptic puts appearances and thoughts into opposition; this generates equipollence (isostheneia, lit. “of equal weight”) between several appearances and/or thoughts; suspension of judgment follows; with it comes tranquility; and the skeptic leads a life according to appearances (DL 9.62, 78, 106–7). However, we do not know much detail of his views on these matters. Instead, Aenesidemus is famous for having developed Ten Modes or Tropes—forms of argument by which the sceptic puts appearances and thoughts into opposition. Key questions about Aenesidemus’ skepticism concern (i) the interpretation of his Modes, (ii) the relationship of his philosophy to competing theories, (iii) the scope of the Ten Modes, and (iv) the skeptic’s mode of speech.

(i) Conflicting Appearances or Causal Invariance. The Ten Modes are preserved in Diogenes Laertius (9.78–88), Philo of Alexandria (On Drunkenness 169–202), and Sextus. Diogenes’ account of the Ten Modes may postdate Sextus’s (Sedley 2015). Sextus gives extensive illustrations, and integrates the Ten Modes into his general account of Pyrrhonism (PH 1.36–163; cf. M 7.345 for ascription of the Ten Modes to Aenesidemus; cf. Annas-Barnes 1985 and Hankinson 1995, 268; the sequence below follows Sextus). Here is the first of the Ten Modes, interpreted in two ways.

Arguments concerning oppositions based on the differences between kinds of animals.

Conflicting Appearances Interpretation:
X appears F to animal of kind A (e.g., humans) and F* to animal of kind B [where F and F* are opposite or otherwise incompatible properties]. We cannot judge how X really is, because we are a party to the dispute.

Causal Invariance Interpretation:
For something to be ‘really’ F, it would have to consistently affect different perceivers as F. But different constitutions of different animals cause different impressions of the same thing. For different animals, something is F and F* (where F and F* are opposite or otherwise incompatible properties). Therefore, things do not seem to really be F or F*.

The Conflicting Appearances Interpretation is based on Sextus’ account (Annas-Barnes 1985). The focus here is on the idea that every kind of animal, perceiver, sensory faculty, thinker, or judger (depending on which mode we consider) is only one of several animals, perceivers, sensory faculties, thinkers, or judgers. The object is perceived or considered from a particular point of view. Everyone is a party to the dispute, and there is no ‘view from nowhere.’ Accordingly, the dispute cannot be decided. The Causal Invariance Interpretation, on the other hand, suggests that the focus on decidability is introduced by Sextus. Aenesidemus may (implicitly or explicitly) have endorsed the following idea: if X were F by nature, X would affect everyone as F. If X affects different people (living beings, sensory faculties, etc.) as F and F*, X is by nature neither F nor F*. For example, if X is harmful to A and beneficial to B, it is neither harmful nor beneficial in its nature (Woodruff 2010, Bett 2000). The Ten Modes can generally be construed as engaging either with conflicts between appearances or with causal invariance:

Arguments based on the differences among human beings (differences in body and in soul).

Arguments based on the differences between the senses and on the complexity of perceived objects.

Arguments based on states (dispositions and conditions of a human being, such as age, motion versus rest, emotions, etc.).

Arguments based on positions, distances, and places.

Arguments based on mixtures (objects in conjunction with external things like air and humidity; physical constituents of sense organs; physiology of thought).

Arguments based on the composition of the perceived object.

Arguments based on relativity (to the judging subject, to circumstances, etc.). 10-8 comprises at least 10-1 to 10-7, or all Ten Modes.

Arguments based on constancy or rarity of occurrence.

Arguments concerned with ways of life, customs, laws, mythical beliefs, and dogmatic assumptions, all of which can be put into opposition to each other.

(ii) Skepticism, Relativism, Epicureanism. Consider first the relationship between skepticism and relativism (cf. Bett 2000; Vogt 2012a, ch. 4). Relativism, as envisaged in Plato’s Theaetetus, looks at a similar range of phenomena. Things appear different to different kinds of animals; to different people; and so on. Relativism embraces the intuition that there is (as we would say today) faultless disagreement. That is, you and I disagree, but neither of us is wrong. Accordingly, metaphysical relativism claims we must give up the intuition that we both refer to the same thing. In the Theaetetus, the world dissolves into radical flux: there are no stable items with stable properties that we both refer to.

The Ten Modes, according to Conflicting Appearances, differ from relativism by turning precisely the other way (Annas-Barnes 1985, 97–8; Pellegrin 1997, 552–3). They implicitly rely on the intuition that there are stable items with stable properties. Of course, the skeptic is not committed to the thesis that opposites cannot hold of the same thing, and that therefore no two conflicting appearances can be true. However, the modes presuppose a common sense metaphysics that does not accommodate faultless disagreement. In all cases of disagreement, at best one of us can be right. If we cannot figure out which view is right, we should suspend. This does not mean that Pyrrhonians are committed to a common-sense metaphysics. The Ten Modes are only one of several tools that skeptics have at their disposal. They may thus imply a metaphysics that, at other points, skeptics would call into question (cf. Fine 2003b, 352).

Causal Invariance differs from Conflicting Appearances precisely with respect to the metaphysics that is, even if only dialectically, invoked. Aenesidemus seems to have explored the relationship between skepticism and flux. He remarks that skepticism leads to Heraclitean philosophy. The idea that one thing appears to have contrary properties (the ones it appears to different animals/persons/senses to have) leads to the idea that one thing actually has contrary properties (PH 1.210; cf. Schofield 2007 on the role of Heraclitus and causal invariance). This remark can be taken as an expression both of moderate flux and of relativism (Aenesidemus does not seem to think of radical flux, where it is no longer even possible to refer to anything). There is no stable reality of how things are (moderate flux); X is F and F* insofar as, if X seems F to A, this is true for A, and F* to B, this is true for B (relativism). This proposal differs from the Causal Invariance interpretation of the Modes presented above. There, Aenesidemus seems to argue that things do not really have stable properties (they are neither F nor F* by nature); he does not say that they are F and F* (as relativism says).

A third approach, competing with skepticism and relativism, is Epicurean epistemology. Again, the set of phenomena to be accounted for is the same. But it is described differently. Epicurus insists that we should not even speak of conflicting appearances. Rather, we should speak of different perceptions. Perceptions cannot refute each other, because they are of the same weight (Epicurus here uses the term that is central to Pyrrhonism: equal weight, isostheneia) (DL.10.31–2). The fact that perceptions differ has perfectly reasonable explanations: I look from a distance, you look from nearby; I have a cold, you are healthy; I am a human being, another cognizer is a dog; and so on. These facts figure in the explanations of how our perceptions are constituted. Accordingly, Epicurus argues, all perceptions, even though they differ, are true. They all have a causal history that physics can explain. The precise interpretation of this proposal is controversial. One might object that the notion of truth employed here is deeply puzzling. It is not clear what it means to describe all perceptions as true if they cannot be true or false.

(iii) Scope. For the greatest part, the Ten Modes seem to be concerned with perception in a broad sense, so that it includes pleasure and pain, harm and benefit, as well as pursuit and avoidance. To perceive something as pleasant or beneficial is to pursue it. Perception and evaluation are also mixed in another way: depending on the frequency with which we perceive something, it seems more or less amazing and precious to us. 10-10 envisages oppositions that can be construed with the help of dogmatic theses. The Ten Modes thus fit Sextus’ description of what skepticism is: the ability to put appearances and thoughts (phainomena and nooumena) into opposition (PH 1.8, 1.31–33; cf. DL 9.78).

Another issue concerning the scope of the Ten Modes is whether they address general or particular matters. Compare the example of whether the tower is round or square (T) to the example of whether honey is sweet or bitter (H). (T) is a particular; the question is whether this tower is round or square. (H) can be construed as a particular (“is this bit of honey sweet?”), or as a general issue (“is honey sweet?”). The Ten Modes offer strategies for suspension of judgment on both kinds of questions.

Scholars have asked whether it is a problem for skepticism if the Ten Modes appear ‘systematic’ (Sedley 2015). A set of arguments that aims to be complete, covering domains according to a standardized pattern, may appear to be out of tune with the skeptic’s presumed mode of investigation. Purportedly, skeptics think through given questions as they arise, arriving at suspension of judgment in a piecemeal fashion. If this self-description is to be taken at face value, then modes of generating suspension of judgment across the board may appear problematic. In this respect, Diogenes Laertius’ report of the Ten Modes may be superior to Sextus’ account. Diogenes begins with a remark that suggests that skeptics pick up where other philosophers have already begun to make an argument (9.78–9). Other philosophers have collected, presumably, ways in which ‘we are persuaded’, say, because things regularly appear the same way. Now the skeptics, as it were in response to this, add a collection of further cases, where things do not appear the same way to different cognizers. If this is the dialectical set-up, the Ten Modes may not be ‘systematic’ in ways that harm skepticism. They may co-opt the patterns of dogmatic reports about cases where appearances are stable (Sedley 2015).

(iv) Language. Aenesidemus contributes an interesting move to the question of how the skeptic can speak. Consider the relationship between a skeptic’s state of mind and their utterances. One way to construe this relationship is that an utterance reflects a state of mind. This is a background assumption to the idea that, if skeptics use assertoric language, they hold beliefs. Another option is to assume that language does not have the means to capture the skeptic’s state of mind. On this premise, a skeptic might flag their utterances as falling short of doing so. This is Aenesidemus’ strategy. He says that the Pyrrhonian determines nothing, and not even this fact that he determines nothing. The Pyrrhonian puts matters in such terms, he says, because he has no way to express the actual thought of the sceptic in determining nothing (Photius, Bibl. 169b40–170a14, = 71C(6)-(8) LS).

4.3 Agrippa, and the Five Modes

Almost nothing is known about Agrippa (1st to 2nd century CE; SE PH 1.164–177; DL 9.88–89). However, the modes of argument that Sextus calls the Five Modes are attributed to him. These modes are among the most famous arguments of ancient skepticism (Barnes 1990, Hankinson 2010).

5-1 Diaphônia:
The mode that argues from disagreement. With respect to some matter that presents itself, there is undecided (anepikriton) conflict, both among the views of ordinary life and the views held by philosophers. Due to this, we are unable to choose or reject one thing, and must fall back on suspension.

5-2 Eis apeiron ekballonta:
Arguments that throw one into an infinite regress. That which is brought forward to make a given matter credible needs yet something else to make it credible, and so on ad infinitum. Since we thus have no starting point for our argument, suspension of judgment follows.

5-3 Pros ti:
Arguments from relativity. X only ever appears such-and-such in relation to the subject judging and to the things observed together with it. Suspension on how X really is follows.

5-4 Hypothesis:
Someone makes an assumption without providing argument. A dogmatist, if thrown back into an infinite regress of arguments, just assumes something as a starting-point, without providing an argument (anapodeiktôs). We suspend over mere hypotheses—they could be false, opposite hypotheses could be formulated, and so on.

5-5 Ton diallêlon:
Arguments that disclose a circularity. This mode is used when that which ought to confirm a given investigated matter requires confirmation (pistis—credibility) from that matter. We are unable to assume either in order to establish the other. We suspend judgment on both.

It is a commonplace to say that, while the Ten Modes, as presented in Sextus, are concerned with conflicting appearances, the Five Modes are about argument or proof. In these modes, the skeptics develop strategies by which to attack theories that the dogmatists defend. If this is how we characterize the Modes, Aristotle’s objection (section 2.3) immediately comes to mind. Do the Five Modes reveal the skeptic’s lack of understanding because they presuppose that everything is subject to proof? (Barnes 1990; Hankinson 1995, 182–92; Long 2006, ch.3) The three so-called formal modes—Regress, Hypothesis, and Circularity—can be construed in this fashion: when employing them, the skeptic can argue that every premise must be supported by argument; if it is not so supported, the theory begins from a mere hypothesis (on earlier uses of the term hupothesis in Plato and in medicine, cf. Cooper 2002 [2004a]); or it is ultimately circular.

However, the skeptic might not be vulnerable to this objection. First, the Five Modes can be construed as dialectical, invoking dogmatic theories of justification (Striker 2004). Second, they might be broader in scope. 5-1 and 5-3 explore disagreement and relativity. Skeptical examination often begins with the Mode of Disagreement: different answers to a given question are surveyed, and the conflict between them is observed. The interpretation of 5-1 hangs, for the most part, on the question of whether anepikriton should be translated as ‘undecided’ or ‘undecidable’ (Barnes 1990). It would be dogmatic to claim that matters are undecidable. The Pyrrhonist must prefer the idea that, up to now, matters have not been decided. This leads to the question of whether something can be found that would decide matters, and thus to the application of further modes. Scholars have observed that 5-3, the Mode of Relativity, does not really fit into the Five Modes. However, the Five Modes could be designed to supersede and include the Ten Modes, and 5-3 might be viewed as capturing the common thread of the Ten Modes. With the help of 5-3, the skeptic can argue that the premises that theorists employ are formulated from particular points of view, in particular contexts, and so on.

Third, even the so-called formal modes (5-2, 5-4, 5-5) might not be narrowly concerned with proof, but rather with everything that can lend credibility to something else. Consider Regress (5-2), the first of the formal modes. The text does not actually speak of proof (apodeixis) (this is obscured by Bury’s Loeb translation; 5-4 is the only place in Sextus’ report of the Five Modes that uses a cognate of apodeixis). Sextus’ language is wider: the mode deals with everything that can make something else credible. We might read this in the context of the Hellenistic view that proof is a species of sign (PH 2.122). A sign reveals something non-evident. Smoke that reveals fire has, from this point of view, a function and structure that is similar to a proof. The target of the Five Modes might be sign-inferences in general. If this is so, then their target might include what we would call inductive reasoning and causal explanations (when Sextus introduces a further set of modes, the Causal Modes, he says that they are not really needed, because the relevant work can be done by the Five Modes; PH 1.180–86). Taken together, the Five Modes deny all “proof, criterion, sign, cause, movement, learning, coming into being, and that there is anything by nature good or bad.” (DL 9.90). This is notably more than just proof. Indeed, it is an excellent summary of the key topics in Sextus’ discussions of logic, physics, and ethics.

(End of Part 4)


Leave a Reply

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

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

Twitter picture

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

Facebook photo

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

Connecting to %s

Blog at

Up ↑

%d bloggers like this: