there was a candle there or not, so a complete description of my evidence doesn't provide us with evidential certainty for believing that there's a candle there. If my evidence really is the same in both cases, experience alone doesn't provide evidential certainty.
2.73 Recall, however, that not everyone believes that the experiences we have when we see a candle are conscious occurrences that involving nothing outside the head. Recall that the disjunctivist about experience thinks that while hallucinations might involve nothing outside of the head, the experience you have when you see something consists of a relation of direct awareness that involves you and the candle. When you see that there is a lit candle on the table, you have excellent reason to believe that there is a candle nearby – you see that there is! When you suffer from some hallucination, it's false that you see that there is, so this couldn't be a good reason for you to believe that there is one nearby. Perhaps the disjunctivist gives the Cartesian foundationalist the tools they need to understand how it's possible to have non‐inferential justification for our beliefs about the external world.
2.9.2 On Modest Foundationalism
2.74 Modest foundationalists might respond to the Epistemological Argument from Error with a shrug. Let's suppose that experience doesn't give us evidential certainty? It still might give us knowledge and still might justify our beliefs. Neither knowledge nor justification requires evidential certainty, they would say.
2.75 We saw above why someone might be cool to the idea that justification requires evidential certainty. The modest foundationalist might say that once we're open to the idea that experience provides justification without providing us with certainty, there is little that would prevent it from providing us with knowledge. Suppose I have the experience as of a red tomato. If I'm having this experience because I'm hallucinating, I wouldn't have knowledge. However, if I don't have any reason to suspect that I'm hallucinating, it seems that I would have justification. And if I'm not hallucinating, it's not plausible that my experience would do less well when it comes to justifying belief. In that case, I would have a justified belief about a tomato that's true. We're getting pretty close to knowledge at this point.
2.76 The interesting challenges for modest foundationalists arise when we try to fill out the details of the view. The claim that certainty isn't needed for non‐inferential justification is a negative claim. It tells us nothing about what is needed and it tells us nothing about how experience provides us with the justification we need for our perceptual beliefs. As we shall see in the chapters that follow, there are a number of different ways to fill in the details of the view. Perhaps experiences justify because they're inputs to reliable processes. Perhaps they justify because they provide us with evidence for believing things about the external environment. As you develop a better idea of what it takes to acquire knowledge and to justify belief, you should use this to help you decide between proposals about how precisely experiences justify.
Free Internet Resources
The Illusions Index, https://www.illusionsindex.org.
Internet Encyclopedia of Philosophy, http://www.iep.utm.edu. See entries for The Epistemology of Perception (Daniel O'Brien), Objects of Perception (Daniel O'Brien), Sense‐Data (Paul Coates).
Stanford Encyclopedia of Philosophy, http://plato.stanford.edu. See entries for The Contents of Perception (Susanna Siegel), The Disjunctive Theory of Perception (Matthew Soteriou), Epistemological Problems of Perception (Jack Lyons), Foundationalist Theories of Justification (Ali Hasan and Richard Fumerton), The Problem of Perception (Tim Crane and Craig French), and Sense‐Data (Michael Huemer).
Wireless Philosophy, http://www.wi‐phi.com. See videos for Paradoxes of Perception (Eugen Fischer).
Notes
1 1 For discussions and defenses of liberal foundationalism, see Huemer (2001), Pryor (2000), and Schellenberg (2013).
2 2 See, however, Sosa (2017, chapter 1) for a reading of Descartes according to which this prescription should be qualified as one concerning judgments, as opposed to what Sosa calls functional beliefs. For more on this distinction, see Sosa (2015).
3 3 James (1956) tells us that we have two goals as “would‐be knowers,” which is to seek out truth and avoid error. To think about the kind of risk that is acceptable or unacceptable, we could imagine attaching “weights” to accuracy and inaccuracy and that this can help us determine the degree of risk that is acceptable. This idea has been developed recently by Easwaran (2016) and Dorst (forthcoming). On their view, it is rational to believe what is sufficiently probable because believing what is sufficiently probable (even if it is not certain) can maximize expected epistemic value. For a helpful introduction to rationality and norms for coping with uncertainty, see Peterson (2009).
4 4 For a detailed discussion of perceptual recognitional abilities, see Millar (2008).
5 5 Of course, in some rare cases what one hallucinates turns out to be so – these cases are called “matching hallucinations.”
6 6 Naïve realism is also known as “direct” realism. For sympathetic discussions of naïve realism, see Brewer (2011), Fish (2009), McDowell (1994), and Logue (2012). For a helpful overview of the philosophical problems surrounding perception, see Crane and French (2015).
7 7 See, in particular, Berkeley's Principles of Human Knowledge (1710), in which he defends a form of what is called “empirical idealism.”
8 8 See Robinson (1994, p. 32).
9 9 See Goldman (1986). This view is discussed in detail in Chapter 9, which focuses on the wider debate between internalism and externalism about epistemic justification.
10 10 Barnes (1944, p. 98).
11 11 In fact, an appreciation of this very point about color exclusion arguably marked an important shift in Wittgenstein's thinking between his early and later works. See, for example, Wittgenstein (1929, p. 167). For helpful discussion on this point, see Sievert (1989).
12 12 Price (1950, p. 246).
13 13 Huemer (2001, p. 154).
14 14 See McDowell (1994) for a defense of the disjunctive view of experience. For a helpful overview of the vast recent literature on disjunctivism, see Soteriou (2014).
15 15 See, in particular, Pritchard (2012b).
16 16 For a similar, albeit critical, articulation of this kind of view, see Sosa (1980, 8).