How are we to account for the epistemic contribution of our perceptual experiences to the reasonableness of our perceptual beliefs? It is well known that a conception heavily influenced by Cartesian thinking has it that experiences do not enable the experiencing subject to have direct epistemic contact with the external world; rather, they are regarded as openness to a kind of private inner realm that is interposed between the subject and the world. It turns out that if one wants to insist that perceptual experiences provide epistemic reasons for perceptual beliefs about the external world as we pre-reflectively take it to be, then one should find a way of avoiding Cartesianism. Here are the two main aims of this paper: firstly, identify the premise that is doing the heavy-lifting work in the Cartesian thinking; and, secondly, formulate an adequate way of denying that premise. The adequacy I claim for my formulation of a way of denying the premise will roughly amount to this: the way I offer is not as susceptible to Cartesian traps as other apparently available ways of denying the premise are.
This is a preview of subscription content, log in to check access.
Buy single article
Instant access to the full article PDF.
Price includes VAT for USA
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
This is the net price. Taxes to be calculated in checkout.
Of course, the idea that perceptual experiences have rationalizing powers contributing to the reasonableness of perceptual beliefs is not a common ground for all philosophers of empirical rationality. In his (1986), Davidson famously defends the claim that “nothing can count as a reason for holding a belief except another belief” (p. 310), the point of which is to exclude experiences from the domain of evidential relations. For the purposes of this paper, Davidsonian views can be safely set aside.
The disjunction here should not be taken in the exclusive sense given that, as Stroud says, “there is the challenge of saying in what ways idealism is superior to, or even different from, the skeptical doctrines it was meant to avoid” (2000, p. 159).
I follow Gupta in taking “Cartesianism” as a shorthand for the thesis of the subjectivity of the given.
As I use the term ‘perceptual judgment’ or ‘perceptual belief’ in this paper, it always means ordinary perceptual judgment or belief defined along these lines.
There are actually four constraints in Gupta’s list. The other two are “Existence” and “Manifestation” (p. 19, 30). These two constraints have no direct bearing on my discussion and can thus be safely left aside for convenience’s sake.
For Gupta, experiences with the same subjective qualitative character are subjectively identical (p. 6, fn. 3). Thus understood, subjective identity amounts to what is more commonly called subjective indistinguishability or subjective indiscernibility.
There is another argument for the same conclusion: the infamous argument from illusion. I agree with Gupta that the sense-datum inference that it invokes—the rule that permits one to derive “there is something \(F\) that is directly perceived” from “there looks to be something \(F\)”—“seems patently invalid” (p. 38).
This move provides Gupta with some dialectical advantage against those attacks that question the viability of the argument from the propositional given by casting doubt on the Reliability constraint. For such attacks, see, for instance, Peacocke (2009) and Schafer (2011). For Gupta’s replies, see Gupta (2009b, 2011).
As Gupta himself notes, “the given aims to capture the contribution of experience from the viewpoint of the experiencing subject” (p. 22, emphasis original).
Anil Gupta raised the objection, in correspondence, that the fact that the Equivalence constraint is formulated in highly general terms, in terms of the idea of the given, renders it unclear what it means for the given to be objective or subjective. I think that this objection can be answered by emphasizing that the idea of the given already incorporates the idea of what is subjectively available to the experiencing subject and thus that the contrast between objectivity and subjectivity applies even at this level of abstractness.
In fact, the second stage of this argument could easily be like this: the given in an experience is either subjective or objective; since as the first stage shows, it is not objective, it is subjective. Moreover, the third stage is, strictly speaking, redundant because the desired generality is already implicit in the first two stages. I formulated the argument in the way above to render it parallel to Gupta’s formulation of the argument from the propositional given.
This argument is very different from Gupta’s version of the Cartesian argument that does not rely on the Reliability constraint (p. 32, fn. 27). First, that version also makes a heavy use of the idea of the propositional given. Second, that version rests on some controversial assumptions about whether two subjects, one having the concept \(F\) and the other lacking it, can have subjectively identical experiences when both have a visual experience of something that falls under \(F\). So, even if Gupta’s version could somehow be relieved of its reliance on the idea of the propositional given, the argument we would thereby have would be vulnerable to some ready-made objections. This is why I have formulated the argument from equivalence in the way above, i.e., without trying to get it through filtering out the propositional given from Gupta’s version.
Furthermore, since the argument from the propositional given is weaker (simply because it rests on an extra substantial assumption, namely, the idea of the propositional given), it seems to me that the argument from equivalence holds more promise for being a charitable interpretation of the classical empiricist line of reasoning from phenomenological constraints on the given to Cartesian subjectivism.
Whether I am (rationally) entitled to hold those judgments depends, for Gupta, on whether I am (rationally) entitled to hold the view that yields them. This complication is inessential for my purposes here. Moreover, this short presentation cannot even begin to do justice to the sophisticated theory provided by Gupta, which includes inter alia a viable account of how hypothetical entitlements may give rise to categorical ones and an interesting turning—the—tables sort of response to radical skepticism. However, these aspects of Gupta’s overall theory do not directly bear on my discussion here and can be safely set aside.
Strictly speaking, not all views that yield subjective judgments are rationally acceptable presumably because there are other criteria such as simplicity and coherence that may be relevant in assessments of rational acceptability. However, since these further criteria are not directly relevant to our discussion, it is safe to formulate the argument on the convenient assumption that what rational acceptability requires is only that assessing the truth of a view is within the subject’s epistemic capacities.
There is indeed a stronger Cartesian argument, which I unimaginatively call the new and strong argument, for the same conclusion, stronger in the sense that unlike the new argument, it does not take as a premise that assessing the truth of the judgments yielded by a view is a must for assessing the truth of that view. The new and strong argument simply holds as a premise that a certain view is rationally acceptable by a given subject only if that view yields judgments whose truth is assessable by the subject. Since even the weaker argument as stated above is strong enough for my purposes and captures the essentials of a sort of typical Cartesian reasoning (see below). I will stick with it.
The locus classicus of the classical Evil Demon is, of course, Descartes’ Meditations published in 1641. For the new Evil Genius, see Lehrer and Cohen (1982).
This is, of course, where Descartes’ non-deceiving God is supposed to come to the rescue and cut the Gordian knot.
Coherentism about justification in its generic form is, as I see it, the thesis that whether a judgment is justified is a matter of how it “hangs together with” (whatever this may mean) other judgments; and as such, it is not committed to a denial of (V). However, a proponent of a sort of coherentist theory of justification may want to object to the priority of assessing the truth of perceptual judgments over assessing the truth of views that seems to be implicated in this statement. However, this sort of coherentism is built on the rejection of the assumption as to the appropriateness of a question around which this paper is structured: how do experiences contribute to the reasonableness of perceptual judgments? And rejecting the appropriateness of this question is the very reason why the coherentist of that sort may want to object to the priority: since, on the sort of coherentism in question, the priority of perceptual judgments may only derive from something non-existent, that is, the rational force of experiences conceived as non-belief-like states, there is no good reason to maintain it. In any case, please note that the argument does not really need (V), see (fn. 17).
Whilst the new argument focuses on the notion of truth, it does not demand truth as a necessary condition for reasonableness. In particular, the new argument is not built on the assumption that the truth of a particular view is a necessary condition for the reasonableness of that view. Clearly, whether it is known by the subject that the truth of a particular view is assessable for her is a different matter from whether that view is true. A particular view might be false and known by the subject to be assessable for her. Thanks to an anonymous referee for pressing on this issue.
One might counter the new argument by holding that it rests on premises that are false and hence that it is not sound. (Thanks to an anonymous reviewer for raising this worry). A Cartesian would reply that if there are reasons for thinking that the premises in question are false that outweigh the reasons for thinking that they are true, then the new argument can be rationally thought to be unsound. So, a Cartesian would request some reasons for thinking that they are false and would be unmoved by the mere assertion that the premises of the argument are false. As I have stated above, a Cartesian has reasons to think that the premises in question are true. For instance, take (O): objective judgments are such that their truth is not assessable by the subject. A Cartesian argues that (O) is supported by the idea that assessing whether an objective judgment is true requires that one tell in any given case a veridical experience from a subjectively identical hallucinatory counterpart. If this idea is true, then given the subjective identity of veridical and hallucinatory experiences, (O) simply follows. A Cartesian is armed with a strong argument for (O) and is within her rights when she makes a request for an equally strong counter-argument from an anti-Cartesian.
In this section, I will assume, for convenience, that the given in experience is propositional in form. However, it is worth noting that what is called “the theory of appearing” holds both that what is given in experience is objects appearing to percipients and that the Equivalence constraint is false. See, for instance, Langsam (1997).
There are apparently other ways of denying Cartesianism, which do not rely on a denial of the Equivalence constraint. One of them is Pryor-style “dogmatism,” according to which “whenever you have an experience as of \(p\)’s being the case, you thereby have immediate (prima facie) justification for believing that \(p\)” (Pryor 2000, p. 532). Propositions that can be replaced by the place-holder \(p\) are, on this view, ordinary perceptual propositions such as “There are hands.” Note also that on Pryor’s dogmatism, “whether or not the experience counts as a case of veridical seeing is irrelevant to one’s justification” (ibid, p. 545, fn. 30). So, on this view, the (prima facie) evidential strengths of the reasons subjectively identical veridical and non-veridical experiences as of seeing that there are hands provide for the belief that there are hands are identical. My only concern in the rest of this paper is to formulate an adequate denial of the Equivalence constraint and I will thus not be concerned with those accounts that keep it. However, it is worth mentioning that Pryor-style dogmatism is vulnerable to powerful Cartesian attacks. One way to motivate the Cartesian attack is through a problem Pryor himself recognizes with dogmatism: “It is not plausible that every proposition we believe on the basis of perception is immediately justified. I may believe on the basis of perception that a certain car is a Honda Accord; but this belief does not seem to be justified just by the fact that I have the experiences that I have. So we have a challenge: we have to determine which of our perceptual beliefs are immediately justified, and which depend for their justification on more than just our having the experiences we have” (ibid., p. 536). If this problem having to do with what “goes beyond the strict deliverances of my experiences” (ibid, p. 538) is pressed adequately by the Cartesian, then it is doubtful whether dogmatism can adequately insist that propositions that can be replaced by the place-holder \(p\) in “experiences as of \(p\)’s being the case” are ordinary perceptual propositions rather than propositions that describe, in some way, “the inner world” of the subject.
To be sure, ruthless disjunctivism is not an entirely novel idea. Pritchard’s epistemological disjunctivism (see for instance (2012)) is, at least in one of its better considered statements, ruthless. The point of the caveat “at least” will become clear in the following sub-section.
William Child gives an especially clear statement of the epistemological motivation: “A central motivation concerns the consequences for epistemology and content of the way we conceive of experience. According to the disjunctivist, to think of conscious experience as a highest common factor of vision and hallucination is to think of experiences as states of a type whose intrinsic features are world-independent; an intrinsic, or basic, characterization of a state of awareness will make no reference to anything external to the subject. But if that is what experience is like, the disjunctivist objects, how can it yield knowledge of anything beyond experience? ...Given that motivation, we must see the disjunctive conception as a thesis about experience as it features in epistemology” (1992, pp. 200–201). See also Pritchard (2012), a considerable part of which is devoted to arguing that epistemological disjunctivism can function as a cure for skepticism.
In his (2012), the work in which I believe one can find the official statement of epistemological disjunctivism, Pritchard concurs: without substantial assumptions that connect mere seemings to facts, that I seem to see hands gives no better reason to believe that I have hands than it gives to believe that I am a brain in a vat.
Notice that in order for the skeptical problem to arise, it is not required that all experiences are non-veridical but only that there are some non-veridical experiences. (In fact, it does not even require this much but only that it is possible that there are some non-veridical experiences. For convenience, I will work with the former existential requirement). If there are some non-veridical experiences, then the mild position claims that they support perceptual judgments by providing reasons of the seeming-to-see form. Now, on the assumption that reasons of the seeming-to-see form can support perceptual judgments only when they are conjoined with substantial background assumptions that connect seemings to facts, the skeptical problem about those substantial assumptions does arise. Simply put, the skeptical problem is about the justification of the substantial assumptions in question, and it arises once it is acknowledged that they are needed for mere seemings to rationally support perceptual judgments. Thanks to an anonymous referee for pressing on this issue.
If, for instance, the best explanation of the fact that the subject seems to see that things are thus and so is that things are thus and so, then the belief that things are thus and so that is held by the subject on the basis of her seeming to see that things are thus and so is justified (see Vogel 1990). I take it that the best explanation in question is not part of the reason the subject has for her relevant belief: the explanation gives an account of why seeming to see by itself gives rational support to the perceptual belief. Otherwise, we would have a problem similar to the one about validity, which is raised by the Tortoise in Lewis Carroll’s famous piece (1985).
It may be thought that the fact that one has an epistemic reason of the form that one seems to see that \(p\) for one’s perceptual belief that \(p\) can perhaps silence the skeptical worries about the reasonableness of perceptual beliefs but not the skeptical worries about perceptual knowledge: that one is epistemically justified for believing that \(p\) does not entitle one to claim that one knows that \(p\). There are two points I want to make in response to this query. First, it is not clear that the subject whose perceptual belief that \(p\) is epistemically (not pragmatically or aesthetically) justified is not in a position to claim that she has a good reason to think that she knows that \(p\). Second, what I am principally concerned to do in this section is to formulate an adequate rejection of the Equivalence constraint, which is a constraint on the rational contributions of experiences to the reasonableness of perceptual beliefs.
Recall that I believe that the official statement of Pritchard’s epistemological disjunctivism is ruthless in the sense specified above. Thus I believe that it is a mistake to think that Pritchard is caught on the horns of the dilemma and the digression that leads to the dilemma may humbly serve as an invitation for clarification. There is a world of difference between saying that merely seeming to see justifies a perceptual belief to a degree less than a state of seeing does and saying that it does not justify it at all, even though it may be dialectically effective in certain contexts to state (a part of) the thesis of epistemological disjunctivism as a disjunction of the two.
The contrast between veridical and non-veridical experiences is not the one that is directly relevant to McDowell’s disjunctivism. This is because there are some veridical experiences, e.g., “veridical hallucinations,” that are not knowledge-yielding. (Thanks to McDowell for letting me appreciate the relevance of this possibility). However, nothing in my discussion will turn on this, and I will set my discussion in terms of the distinction between veridical and non-veridical experiences.
It would be very unreasonable to understand McDowell’s point in writing “Givenness should be givenness for knowing” in the quotation above as imposing strict restrictions on one’s choice of terminology. Rather, it is preparatory for McDowell’s disjunctive conception of the given for knowledge in an experience. Hence I do not think McDowell would raise his hand against my interpretation of what he says about the relation between experience and rational belief in terms of the terminology of the given. Moreover, it is instructive here to take note of the fact that McDowell himself uses the terminology of the given to articulate his own view about the rational contribution of mere appearances to the reasonableness of perceptual beliefs: “We can acknowledge that an experience that merely seems to be one of perceiving can give its subject a reason for belief” (2009, p. 470, emphasis added).
It is worth noting that the given in an experience is not responsive to anything. Thanks to McDowell for pressing on this issue.
For a closely related controversy on the appropriate way of understanding McDowell’s take on the relation between states of having it looking to a subject as if \(p\) and states of her seeing that \(p\), see Haddock and Macpherson (2008, p. 8).
Pritchard’s parenthetical remark reports a view that is the same as the one Comesana attributes to McDowell (according to which (3) is false), and there are, as I have pointed out above, good reasons for being suspicious of the claim that McDowell holds that view.
For McDowell, the disjunctive conception better allow what is given in experience in veridical and non-veridical cases to be the same to accommodate “the occurrence of deceptive cases experientially indistinguishable from non-deceptive cases.” (1998a, p. 389). A further issue is what McDowell means when he adds the qualification “in so far as it is an appearance that things are thus and so” to his claim that what is given in a veridical experience is the same as what is given in its hallucinatory counterpart. I think Putnam’s metaphor of portraying the environment is helpful here. The experiences in question present the subject with subjectively identical portrayals of the environment; and that one of them is a mere appearance and the other a worldly fact does not have anything to do with what is given in the experience for rational belief.
Consider also this: “What I urge in my paper [“Knowledge and the Internal”] is precisely that justification adequate to reveal a state as one of knowing must be incompatible with falsehood and can be had” (McDowell 2002, p. 98, emphasis added). I argue that McDowell’s claims about knowledge-adequate rational basis should not be taken as having a direct import on what he has to say about knowledge-inadequate rational basis. As McDowell himself puts it: “It might be rational (doxastically blameless) for that subject—who only seems to see a candle in front of her—to claim that there is a candle in front of her. But this is not the notion of entitlement or justification that should figure in a gloss on the Sellarsian thought that knowledge is a standing in the space of reasons. The right notion for Sellars’s point is precisely...a notion for which entitlement and truth do not come apart” (2002, p. 99). I take it that that this quotation makes clear that there are two notions of what is given in an experience, one adequate to knowledge and the other not, in McDowell’s thinking.
McDowell says: “I agree with [Elizabeth Fricker] that we lose the point of invoking the space of reasons if we allow someone to possess a justification even if it is outside his reflective reach” (1998c, p. 418, fn. 7).
See McDowell (2006).
There are other ways of describing the controversy between internalism and externalism. One is via what is called “mentalism,” a sort of epistemic internalism defended by Feldman and Conee (2001), and the other via the new Evil Genius case, presented by Lehrer and Cohen (1982). For a nice classification of different conceptions of the controversy between internalism and externalism, see Neta and Pritchard (2007).
Pritchard (2012, p. 92) writes that it is an “undeniable fact that there are pairs of deceived and non-deceived cases...which are introspectively indistinguishable.”
According to Pritchard (2012), merely granting the reflective accessibility of the factive reason that justifies a perceptual belief is sufficient for his epistemological disjunctivism to count as epistemically internalist. What i argue above is that this is mistaken.
Alston, W. P. (1988). An internalist externalism. Synthese, 74(3), 265–283.
Carroll, L. (1985). What the tortoise said to Achilles. Mind, 4(14), 278–280.
Comesana, J. (2005). Justified vs. warranted perceptual belief: Resisting disjunctivism. Philosophy and Phenomenological Research, 61(1), 367–383.
Child, W. (1992). Vision and experience: The causal theory and the disjunctive conception. The Philosophical Quarterly, 42(168), 297–316.
Davidson, D. (1986). A coherence theory of truth and knowledge. In E. LePore (Ed.), Truth and interpretation: Perspectives on the philosophy of Donald Davidson (pp. 307–319). Oxford: Blackwell.
Feldman, R., & Conee, E. (2001). Internalism defended. In H. Kornblith (Ed.), Epistemology: Internalism and externalism (pp. 231–260). Oxford: Blackwell Publishing.
Gupta, A. (2006a). Empiricism and experience. Oxford: Oxford University Press.
Gupta, A. (2006b). Experience and knowledge. In T. S. Gendler & J. Hawthorne (Eds.), Perceptual experience (pp. 181–205). Oxford: Oxford University Press.
Gupta, A. (2009a). Précis of empiricism and experience. Philosophy and Phenomenological Research, 79(2), 461–466.
Gupta, A. (2009b). Equivalence, reliability, and convergence: Replies to McDowell, Peacocke, and Neta. Philosophy and Phenomenological Research, 79(2), 490–508.
Gupta, A. (2011). Replies to Selim Berker and Karl Schafer. Philosophical Studies, 152(1), 41–53.
Haddock, A., & Macpherson, F. (2008). Introduction: Varieties of disjunctivism. In A. Haddock & F. Macpherson (Eds.), Disjunctivism (pp. 1–24). Oxford: Oxford University Press.
Langsam, H. (1997). The theory of appearing defended. Philosophical Studies, 87(1), 33–59.
Lehrer, K., & Cohen, S. (1982). Justification, truth, and coherence. Synthese, 55(2), 191–207.
McDowell, J. (1998a). Criteria, defeasibility, and knowledge. Originally published 1982. In J. McDowell (Ed.), Meaning, knowledge, and reality (pp. 369–394). Cambridge: Harvard University Press.
McDowell, J. (1998b). Knowledge and the internal. Originally published 1995. In J. McDowell (Ed.), Meaning, knowledge, and reality (pp. 395–413). Cambridge: Harvard University Press.
McDowell, J. (1998c). Knowledge by hearsay. Originally published 1994. In J. McDowell (Ed.), Meaning, knowledge, and reality (pp. 414–443). Cambridge: Harvard University Press.
McDowell, J. (2002). Knowledge and the internal revisited. Philosophy and Phenomenological Research, 64(1), 97–105.
McDowell, J. (2006). The disjunctive conception of experience as material for a transcendental argument. Teorema, 25(1), 19–33.
McDowell, J. (2009). The given in experience: Comment on Gupta. Philosophy and Phenomenological Research, 79(2), 468–474.
Neta, R., & Pritchard, D. (2007). McDowell and the new Evil Genius. Philosophy and Phenomenological Research, 74(2), 381–396.
Peacocke, C. (2009). Perception, content and rationality. Philosophy and Phenomenological Research, 79(2), 475–489.
Pritchard, D. (2011). Epistemological disjunctivism and the basis problem. Philosophical Issues, 21(1), 424–455.
Pritchard, D. (2012). Epistemological disjunctivism. Oxford: Oxford University Press.
Pryor, J. (2000). The skeptic and the dogmatist. Nous, 34(4), 517–549.
Putnam, H. (1999). The threefold cord: Mind, body, and world. New York: Columbia University Press.
Schafer, K. (2011). The rationalism in Gupta’s empiricism and experience. Philosophical Studies, 152(1), 1–15.
Stroud, B. (2000). Understanding human knowledge. Oxford: Oxford University Press.
Vogel, J. (1990). Cartesian skepticism and inference to the best explanation. The Journal of Philosophy, 87(11), 658–666.
I would like to thank Anil Gupta, John McDowell, Erdinç Sayan, Teo Grunberg, David Grunberg, Steven Voss and three anonymous referees for their comments on the previous versions of this paper.
About this article
Cite this article
Demircioglu, E. The given in perceptual experience. Synthese 192, 2667–2693 (2015). https://doi.org/10.1007/s11229-015-0730-4
- Epistemic given
- Empirical rationality
- John McDowell
- Anil Gupta