The Greek word skepsis means investigation. By calling themselves skeptics, the ancient skeptics thus describe themselves as investigators. They also call themselves ‘those who suspend’ (ephektikoi), thereby signaling that their investigations lead them to suspension of judgment. They do not put forward theories, and they do not deny that knowledge can be found. At its core, ancient skepticism is a way of life devoted to inquiry. Also, it is as much concerned with belief as with knowledge. As long as knowledge has not been attained, the skeptics aim not to affirm anything. This gives rise to their most controversial ambition: a life without belief.
Ancient skepticism is, for the most part, a phenomenon of Post-Classical, Hellenistic philosophy. The Academic and Pyrrhonian skeptical movements begin roughly in the third century BCE, and end with Sextus Empiricus in the second century CE. Hellenistic philosophy is a large-scale conversation, not unlike philosophy today. The skeptics (among them Pyrrho, Timon, Arcesilaus, Carneades, Aenesidemus, and Sextus Empiricus) do engage with Pre-Socratic philosophy, Socrates, Protagorean relativism, Plato, and perhaps Aristotle. But their contemporary and principal interlocutors are Epicureans, Stoics, Cynics, and Megarian logicians (cf. Long 2006, ch. 4 and 5).
1. The Central Questions
The core concepts of ancient skepticism are belief, suspension of judgment, criterion of truth, appearances, and investigation. Important notions of modern skepticism such as knowledge, certainty, justified belief, and doubt play no or almost no role. This is not to say that the ancients would not engage with questions that figure in today’s philosophical discussions. Ancient debates address questions that today we associate with epistemology and philosophy of language, as well as with theory of action, rather than specifically with the contemporary topic of skepticism. They focus on the nature of belief, the way in which belief figures in our mental lives, and the relationship of belief to speech and action.
Pre-Socratic philosophers formulate—often in the context of revisionist metaphysical theories, which lead into epistemological discussions—such claims as “nothing is known” (Lee 2010). The firm assertion, as found in these philosophers’ works, that there is no knowledge can be turned against itself: does the proponent of “nothing is known” claim to know that nothing can be known (and if not, why assert it)? This question gives rise to a puzzle that remains at the heart of ancient debates about skepticism. Can the skeptics say anything meaningful about their philosophy without asserting anything about how things are (Bett 2013)? Skeptical writings have a peculiar format, one that comes with its own challenges: the skeptics aim to describe their philosophy tout court, as they practice it, without laying out any particular theories or doctrines. Skeptical ideas have been charged with a family of objections: they might be self-refuting, inconsistent, self-contradictory, and so on (Castagnoli 2010). Another line of objection is associated with Hume, namely that “nature is always too strong for principle.” As Hume puts it, “a Pyrrhonian cannot expect that his philosophy will have any constant influence on the mind” (part 2 of section 12, “Of the Academical or Sceptical Philosophy,” An Enquiry Concerning Human Understanding, London, 1748). It is one thing for skepticism to be coherent. It is another thing for it to be likely that anyone, no matter how much they rehearse skeptical arguments in their mind, will succeed in adhering to it (Johnson 2001), as ancient Pyrrhonist philosophers claimed to be able to do.
Like later epistemologists, the ancient skeptics start from questions about knowledge. But discussion quickly turns to beliefs (Fine 2000). The Greek term translated here as belief, doxa, can also be translated as opinion. The root of doxa is dokein, seeming. In a belief, something seems so-and-so to someone. But there is also an element of judgment or acceptance. The relevant verb, doxazein, often means ‘to judge that something is so-and-so.’ Hellenistic discussions envisage three attitudes that cognizers take to impressions (how things seem to them): assent, rejection, and suspension of judgment (epochê).
Suspension is a core element of skepticism: the skeptic suspends judgment. However, if this means that the skeptic forms no beliefs whatsoever, then skepticism may be a kind of cognitive suicide. Arguably, belief-formation is a basic feature of human cognitive activity. It is not clear whether one can lead an ordinary human life without belief, or indeed, ancient opponents of the skeptics say, whether one can even survive. Perhaps even the simplest actions, such as eating or leaving a room without running into a wall, involve beliefs (on the practical side of ancient skepticism, see Annas-Barnes 1985, 7; Burnyeat 1980). It is also hard to say whether someone who succeeded in not forming any beliefs could communicate with others, whether they could engage in philosophical investigation, or whether they could even think at all.
The ancient skeptics are well aware of these objections. The most widely discussed charge is that they cannot act without belief (Apraxia Charge). In response, the skeptics describe their actions variously as guided by the plausible, the convincing, or by appearances. The notion of appearances gains great importance in Pyrrhonian skepticism, and poses difficult interpretive questions (Barney 1992). When something appears so-and-so to someone, does this for the skeptics involve some kind of judgment on their part? Or do they have in mind a purely phenomenal kind of appearing? The skeptical proposals (that the skeptic adheres to the plausible, the convincing, or to appearances) have in common their appeal to something less than full-fledged belief about how things are, while allowing something sufficient to generate and guide action.
However, my claim that ancient skepticism is about belief, while modern skepticism is about knowledge, needs to be qualified. Ancient skepticism is not alone in being concerned with belief. Recall that Descartes speaks repeatedly of demolishing his opinions (for example, Med 2:12, AT 7:18; cf. Broughton, 2002, 33–61). Contemporary epistemology often pays equal attention to the notions of both justified belief and knowledge. The distinctive focus of ancient skepticism on belief becomes clearer once we consider a third concept that figures centrally in ancient discussions: criterion of truth. It is a core ancient intuition that, if we cannot identify an impression as true, we should hold back from taking it to be true and from believing anything on the basis of it. The skeptics and their opponents discuss how one recognizes a true impression as true. Is there anything about impressions of truths that marks them as true? Are there some evident things (some kind of impressions), which can be used as standards or criteria, so that nothing is to be accepted as true if it is not in agreement with these evident things? The Stoics and Epicureans formulate theories that conceive of such criteria. The skeptics respond critically to their proposals. Accordingly, the conception of a criterion of truth assumes as central a role in ancient debates as does the notion of knowledge in modern discussions. This debate includes in-depth analysis of sense perception and its relation to belief. According to Epicurus, all sense perceptions are true, but judgments based on these perceptions are true or false (Striker 1977, Vogt 2016). The Stoics explore differences between sense perception, illusion, and hallucination (Vasiliou forthcoming). Their account of the criterion of truth starts from perceptual impressions that qualify, or fail to qualify, as cognitive (Shogry forthcoming-b). The Stoics propose that we should accept only cognitive impressions, and accordingly we should only form beliefs based on a subset of true perceptual impressions.
Discussion of the criterion of truth arguably also covers some of the ground that is later discussed in terms of certainty. The Stoics say that a particular kind of impression is the criterion of truth: the cognitive impression. Cognitive impressions make it clear through themselves that they reveal things precisely as they are. This notion is an ancestor to the later conception of clear and distinct impressions, and thus, to discussions of certainty.
Consider next the notion of doubt. Doubt is often considered the hallmark of skepticism. So how can it be that ancient skepticism is not about doubt (Corti 2010, Vogt 2014a)? Insofar as ‘to doubt’ means no more than ‘to call into question,’ the ancient skeptics might be described as doubting things. However, skeptical investigation as Sextus Empiricus describes it does not involve doubt (I shall focus here on Pyrrhonism; on Cicero’s use of dubitari, see Section 3.3). Skeptics find themselves struck by the discrepancies among impressions. This experience is described as turmoil. They aim to resolve this disturbance by settling what is true and what is false among them. But investigation leads them to suspension of judgment, which brings its own peace of mind (Outlines of Skepticism [= PH] 1.25–30). Where in this account should we locate doubt? Is the initial turmoil the ancient skeptic experiences a kind of doubt? Are the ancient skeptic’s investigations a kind of doubting? Should we describe suspension of judgment as a kind of doubt? All three stages may resemble doubt, at least insofar as the ancient skeptics have not settled on answers to the questions they investigate. But all three stages are also different from doubt as it is conceived in later epistemology. The ancient skeptics do not describe themselves as making an active effort at doubting what ordinarily they would believe, as some philosophers in the Cartesian tradition have it. Instead the ancient skeptics find themselves in turmoil because of discrepancies in how things strike them. Moreover, the progression that ancient skeptics describe differs from the doubt-belief model that later thinkers tend to employ. The ancient skeptics improve their psychological condition by moving from turmoil to suspension of judgment, not by removing doubt. It seems best, then, to refrain from invoking the modern conception of doubt as at all fundamental in the reconstruction of ancient Greek skepticism.
Some of the distinctness of ancient skepticism lies in the fact that it is developed by philosophers who genuinely think of themselves as skeptics. In later epistemology, skepticism is largely construed from the outside. In particular, early modern skepticism is, for the most part, conceived by philosophers who aim to refute it. But ancient skepticism is explored by skeptics, and that is, by philosophers who intend their lives to be reflective of their philosophy (Cooper 2012 ch.5.5–7; Bett 2013b). Socrates raises the challenge that it might be truly bad (for one’s life, for the state of one’s soul, and so on) to base one’s actions on unexamined beliefs. For all one knows, these beliefs could be false, and without investigation, one does not even aim to rid oneself of false belief, which is admittedly a bad thing for one’s soul. Only an examined life is worth living (Cooper 2007). Once we take this challenge seriously, as the ancient skeptics do, we embark on a kind of investigation that is seen as directly relevant to our lives. Our beliefs are assumed, at this pre-skeptical phase, to be guiding our actions. Confidence in unexamined views seems misplaced. Others regularly disagree with us. With respect to even the most basic questions, such as whether there is movement, or whether there are good and bad things, we face conflicting views. In favor of each view, some arguments can be adduced, some practices invoked, some experiences cited. These conflicting arguments, practices and experiences need to be examined. But that just raises further views that are in conflict. As a consequence, suspension of judgment on every such question looks rationally mandatory. But it is also rational to persist in investigation. The skeptic is committed to a search for the truth, on virtually all questions, even if this search repeatedly and predictably leads to suspension of judgment (Cooper 2012).
(End of Part 1)