Reply to Kriegel, Smithies, and Spener


for Philosophical Studies


Eric Schwitzgebel

Department of Philosophy

University of California at Riverside

Riverside, CA  92521-0201


eschwitz at domain:



May 17, 2013




I. Uriah Kriegel.

Uriah Kriegel and I almost agree.  This is a somewhat remarkable fact, given that he is a friend of introspection and I am – one might think – a foe of it.  But I am not as much a foe, in truth, as a despairing and conflicted friend; and Uriah’s species of friendship is far from simple trusting boosterism.  The real foes of introspection are those who think they have some far superior epistemic path to what they care about.  They’re the ones who can afford to say, “Bah, introspection, poppycock!  Forget about it.  Here’s some serious science instead.”  I have no such confidence in purely non-introspective approaches to the stream of conscious experience (if such really exist; and here I fully endorse Uriah’s comments in Section 2 of his commentary).  So I can’t afford to ignore the evidence of introspection.  I just fear it might be too frail to deliver many of the conclusions we want from it.

Uriah suggests that introspection is about as powerful an instrument for learning about our stream of experience as is the human sense of smell for learning about the external environment.  If that were so, our knowledge of the stream of experience would be sketchy indeed.  Uriah writes:

Imagine a race of creatures whose perceptual contact with the external world is limited to a sense of smell (as powerful as ours).  Its perceptual grasp on the external world would be doubly weaker than ours: it would lack information from more powerful senses (such as vision), and it would lack the ability to calibrate information from multiple sources.  Its perceptual grasp on its environment would consequently be much shakier than ours.  It would parallel, I contend, our introspective grasp on our own internal world (**** ref).

If this is the right assessment, we’ve fallen a long way from Cartesian infallibility and indubitability!  I think here of Condillac’s notoriously ignorant olfactory statue (Condillac 1754/1982; cf. also Strawson’s 1959 auditory world).  There’s a bit there, but not much, and not very secure except about the crudest matters.  This is not, I think, how most friends of introspection would have spoken in the 17th-20th centuries.

Uriah defends a position he calls Refined Introspective Minimalism:

For any (normally circumstanced) normal subject S and any non-elusive phenomenology P: If S introspects having P, then S is more likely to have P (than if S does not so introspect) & If S has P, then S is more likely to introspect having P (than if S does not have P) (**** ref).

Note the qualifications: normally circumstanced normal subjects introspecting non-elusive phenomenology.  These qualifications might do a lot of work.  They permit circumstances, even pervasive circumstances, and phenomenology types, even pervasive phenomenology types, where we go persistently and irredeemably wrong.

Uriah suggests “fringe” phenomenology, the phenomenology of “flow”, and the phenomenology of fury as types of phenomenology that might be elusive in the relevant sense.  I’m inclined to think that fringe phenomenology is rather a big deal.  As I suggest in Perplexities Chapter 6, we might need to rely upon introspection to know whether human phenomenal experience is sparse, limited to one or a few topics or modalities or regions at a time, or instead abundant, containing also, in every moment, what Uriah calls a “fringe” across all or nearly all sensory and other modalities, or instead somewhere in between (see also James 1890/1981; Mangan 2001).  Maybe it will turn out that we cannot settle this question introspectively, because attending to the fringe renders it non-fringe and because structural limitations of memory doom attempts retrospectively to recall whether there was fringe experience.  If non-introspective approaches then also fail, we might find ourselves epistemically stuck.  If so, one consequence might be incurable pessimism about general theories of consciousness, since general theories of consciousness seem to build in, from the beginning, assumptions about the relative sparseness or abundance of experience – assumptions that could not possibly be justified in this pessimistic scenario.  I don’t know if such flat pessimism is warranted.  I flirt with it near the end of Chapter 6.  My point here is that dark pessimism about some of the absolutely most basic and pervasive features of consciousness, and about the future of any general theory of consciousness, seems to be entirely consistent with Uriah’s hesitant defense of introspection.

Consider another debate: the debate about whether there is some sense in which a tilted coin looks elliptical (which I discuss in Chapter 2 of Perplexities).  I suspect that people who think the coin does in some sense look elliptical are mistakenly over-analogizing visual experience to flat, projective two-dimensional media like paintings and film.  But opponents might insist that the phenomenology of ellipticality is real and say that I am blinded to this aspect of my phenomenology because I’m too invested in rejecting majority opinion whatever it is or because I’m too theoretically attracted to a view of our phenomenal space as fully three-dimensional.  These accusation have merit; I am partly unseated by them.  Suppose, then, that this dispute continues to boil, and that we also fail to settle the issue by non-introspective means.  Such persistent irresolvability would be consistent with Refined Introspective Minimalism, as I understand it: Each party need only think of the other party as not “normally circumstanced” because overly biased.

Maybe even there are some matters of wide consensus that we get radically introspectively wrong because very few of us are normal introspectors normally circumstanced?  “Normal”, here, I assume to be a normative term, rather than a majoritarian description.  So then maybe, as some mystics might say, almost all of us are clueless about the phenomenology of the Inner Self because we are too fascinated by External Things?  Or maybe most of us are normal perceivers normally circumstanced, but the introspective benefit is tiny, so that we are only ever so slightly more apt to get it right about something when we introspect?  Maybe if someone introspectively reports having P, she is 4% likely to have P, but she is only 2% likely to have P if she does not introspectively report having P.  Uriah’s conditions for introspective minimalism could thus be satisfied despite our being wrong almost all of the time.

So Refined Introspective Minimalism is consistent with very substantial pessimism about the accuracy and reliability of introspection, even if it avoids absolutely 100% flat pessimism.  It is consistent with massive, pervasive, unconquerable error, with the failure of any hope for scientific consensus, and with the collapse of the attempt to study consciousness.  It is consistent with widespread failure to meet “Schwitzgebel’s Challenge” – consistent, that is, with the widespread failure of introspective methods, or any other methods, to provide us a scientifically respectable understanding of the stream of experience, across a wide range of centrally important cases.  If Refined Introspective Minimalism is the position to which friends of introspective reliability are forced, then I have shown all I intended to show.  Presumably Uriah would want a bit more than just minimal Refined Introspective Minimalism – but it remains unclear how much more can really be justified.  The olfaction analogy is suggestive, but still also compatible with quite substantial pessimism, as I think Uriah intends.


II. Declan Smithies.

Declan Smithies’ commentary turns upon the distinction between brute error and basing error.  Brute errors are justified false beliefs properly based on justifying evidence, and basing errors are unjustified false beliefs that are not properly based on justifying evidence.  If unbeknownst to me I have been put in a situation in which I have excellent perceptual grounds for thinking there’s a dog outside – maybe someone is playing a realistic sound recording of a dog – then I have a justified false belief and my error is “brute”.  In contrast, if I have no good reason for thinking there’s a dog outside and instead leap to that conclusion based only on wishful thinking as I hear a cart roll by, then my belief is unjustified and I’ve made a “basing” error.  Declan uses this distinction to defend a certain kind of limited introspective infallibilism: Introspection, he says, unlike perception, is immune to brute ignorance and brute error.

The distinction between brute and basing error seems potentially valuable.  And furthermore, Declan’s limited claim about the infallibility of introspection is true! – or rather, I should say, it’s true given his definition of “introspective justification”.  Given his definitions, in fact, the infallibility of introspective justification is a priori true no matter how cruddy people in fact are in reaching judgments about their ongoing stream of experience.  This apriority makes Declan’s thesis, I think, too unbeatably true for its own good.

According to Declan, “the source of one’s introspective justification to believe that one has a certain kind of experience is constituted by that experience itself….  [I]ntrospective justification is identical with its subject matter” (**** ref).  In other words, the epistemology of introspection is distinctive in a certain way: What introspectively justifies beliefs about one’s experiences just are the believed-in experiences themselves.  We are, I think, to understand this as a stipulative definition of introspective justification on Declan’s account (see also Smithies 2012a).  If so, infallibility to brute error follows straightaway a priori: If one does not have the experience one believes one has, one cannot, by definition, be introspectively justified in believing one has that experience.  But brute error, by definition, occurs only when one is justified and nonetheless mistaken.  So introspective brute error is, by definition, impossible.  My metaphilosophy teaches me that the question should then become a pragmatic one: Is this a useful way to carve up the definitional territory, given our values and the empirical facts?

A thought experiment can set up my response.  Declan seems willing to grant the skeptic quite a bit of empirical territory: We might, in fact, be highly unreliable in our introspective judgments, and some debates might persistently resist solution given our non-ideal capacities.  Declan’s position can, he says, “be reconciled with Schwitzgebel’s pessimism about [introspection’s] impotence in settling hard questions about the nature of conscious experience” (**** ref).  Let me, then, exploit Declan’s concessions on this front.  Imagine two parties, Party A and Party B.  Members of Party A think we have constant tactile experience of our feet in our shoes (see Perplexities, chapter 6) and that phenomenology involves much more than just sensory and imagistic content, including distinctive “cognitive phenomenology” and other such phenomena (see Perplexities, chapter 7).  Members of Party B think experience is always limited to only one or a few things at a time in the field of attention and thus that we usually have no experience whatsoever, not even peripheral “fringe” experience, of our feet in our shoes, and also that sensory and imagistic phenomenology is all the phenomenology there is.  Suppose, further, that the disputes between Party A and Party B prove stubborn over time and involve no particular intellectual failings by either Party, such as unusual bias or lack of sincere introspective effort or commitment to dubious auxiliary hypotheses.  There is no deficiency in justification by “ordinary standards”, as Declan puts it, but only by “ideal standards” (**** ref).  To the extent conceivable, the only difference between the two parties is in the truth of their opinions: One party happens to have it right and the other happens to have it wrong.  An ideal introspector would get it right, let’s suppose, but a mere human, applying her best merely human capacities, is as a matter of empirical fact just as likely to err as to succeed.  Finally, let’s suppose that Party B is actually right.  On Declan’s account, Party B will be introspectively justified in their opinions, but Party A will be introspectively unjustified. 

Objection 1: It seems odd to say that Party A and Party B differ in their degree of justification.  There’s an asymmetry in the truth of their opinions, of course.  But asymmetries of justification are widely held to differ from asymmetries of truth.  The notion of “justification” as used by epistemologists normally serves to mark conditions in which we exhibit adequate epistemic virtue, or right operation, or something like that, and yet might unfortunately nonetheless fail to arrive at the truth.  To make truth a necessary condition of justification deprives us of the potentially useful distinction between making the right kind of effort and succeeding in that effort.  Since Parties A and B, if I have imagined them coherently, are making similar effort, doing their best with the resources they have, and neither party is guilty of any vice the other lacks, other than simply the vice of error itself, it seems more natural, less potentially misleading, to say either that both parties have justification or that neither does.

I have pressed this concern upon Declan in discussion, and he has two responses.  First, he says, Party B, which I am assuming to be correct, might have “propositional justification” but not “doxastic justification”.  The proposition “experience is entirely sensory and imagistic” is one that they have introspective justification to believe, but their “doxastic dispositions may be insufficiently sensitive to the facts about conscious experience that determine which propositions we have introspective justification to believe” (**** ref).  Declan expands more fully on this distinction elsewhere: One is doxastically justified in believing a proposition only if one believes the proposition “on the basis of one’s justification” (Smithies 2012b, p. 722-723).  Furthermore, one’s belief will be justified only if one has both epistemic and doxastic justification (2012b, p. 722).  Compare: If a reliable source tells me that the Dodgers won, but if I believe that the Dodgers won not on those grounds but rather because the tea leaves said so, then I have propositional justification for believing that the Dodgers won but not doxastic justification.  Parties A and B then, would presumably both lack doxastic justification despite Party B’s having propositional justification; and then neither party justifiably believes.  The implications thus seem to be skeptical rather than anti-skeptical: It makes tea-leaf readers of both parties.

Declan’s second response is to suggest that what the best members of Party B have is epistemic justification by ideal standards even if they lack it by ordinary standards.  This move, it seems to me, considerably narrows the interest of his infallibilism.  Lacking ideal minds, ordinary standards are what we real human beings have to work with.  Compare the case of mathematical justification.  Before the proof of Fermat’s Last Theorem, presumably an ideal mind would have been justified in believing Fermat’s Last Theorem.  But an ordinary mind might well not have been justified.  If Jane leaps to belief in Fermat’s Last Theorem based on an unjustifiably inflated confidence in Fermat’s genius, she might still be “ideally” justified in believing the theorem in the sense that it compels belief by ideal standards; but the epistemic facts are more helpfully expressed by instead denying that Jane has adequate justification for her belief.  So also, it seems natural to say, the members of Party B lack justification, if the empirical situation is sufficiently dire.

Objection 2: Declan’s view overplays the importance of the distinction between brute and basing errors.  Although I see merit in Declan’s distinction, the matter of greatest epistemic import is whether or not I am, or am likely to be, in fact mistaken, and only secondarily the path by which I got there.  Descartes errs, I think, when he sets aside the possibility that he is a madman, and Declan (and most of the rest of the tradition) errs in following Descartes in this, too swiftly limiting the range of skepticisms they are willing to consider.  Why, exactly, does Descartes reject the thought that he might be like madmen who “say they are dressed in purple when they are naked, or that their heads are made of earthenware, or that they are pumpkins, or made of glass” (1641/1984, p. 13)?  He offers only one sentence in answer: “But such people are insane, and I would be thought equally mad if I took anything from them as a model for myself” (ibid.).  Maybe it was wise of Descartes not to risk the social stigma of being thought insane, but this doesn’t seem compelling epistemic grounds to dismiss the case if one’s aim is absolute certainty.  Or is Descartes instead saying that it would in fact be insane to consider the possibility that one is insane?  That doesn’t seem right.  It’s not clear that we should rule out the possibility of our own insanity without argument, if it’s possible that the evidence will lead us to conclude that we are mad.  I have occasionally met otherwise intelligent people some of whose opinions were mere clinical insanity (e.g., paranoid schizophrenics) who I wished were more open to considering the evidence of their madness; and if something similar might be true of me, I think it reflects rather better than worse on my overall sanity were I willing to consider the evidence for and against it.  (See al Christensen 2010.)  Furthermore, if our opinions about our stream of experience might be as wildly off the ground as a glass-headed madman’s, it seems more likely to worsen than to cure our madness to develop a set of definitions according to which those of us who happen by lucky chance nonetheless to be correct have a certain kind of infallible justification.

Objection 3: By casting his thesis as a priori and true by definition and ceding the empirical territory to the skeptic, Declan has surrendered the game that matters.  Suppose we imagine a skeptic more radical than I – one who denies even the weakest possible version of Uriah’s Refined Introspective Minimalism, one who sees nobody’s belief about his experience ever as even the tiniest bit responsive to the facts about his experience, who thinks that all claims even about experiencing canonical pain and canonical foveal red in seemingly maximally favorable introspective situations are the merest stabs in the dark, as likely to be accompanied by, say, the experience only of tasting jellybeans as by pain or red.  Even this skepticism is compatible with Declan’s infallibilism about introspective justification.  But then it seems misleading to highlight this infallibility, to craft a set of definitions around it, and to present one’s view as a partial defense of the Cartesian tradition according to which we know our stream of experience better than we know the world outside.  The most important issues here are the empirical ones, and the definitions should fall out secondarily, in whatever way best helps us negotiate the empirical territory.

In conversation, Declan has acknowledged that his approach can only have value under certain non-skeptical assumptions about our knowledge of our experience.  The question then arises: How much skepticism his view can tolerate before his structures start to lose their value?  Without a fuller sense of where Declan is going, this is a tough question to answer.

Let me end with a concession to Declan’s view.  Despite my complaints, I acknowledge that there is some truth in the thought that there is a kind of error that is possible in perception for which there is no straightforward parallel in introspection.  If a malicious demon horribly enough hijacks my perceptual inputs, I can (presumably) be radically wrong about the external world despite being the ideal model of rationality and reason within.  Introspective judgments don’t seem straightforwardly demonizable by quite the same method.  This difference between perception and introspection is presumably in some way important.  Different modes of knowing are, by their nature, subject to different sources of error – an interesting fact about them!  But I see no real comfort here for the (neo-)Cartesian without a system of additional commitments in the background.  In his body of work as a whole, Declan is working to construct just such a system, and maybe in the end it will prove sufficiently attractive to justify accepting his definitions despite the seemingly odd consequences I’ve discussed above.


III. Maja Spener.

Maja Spener finds Perplexities to have an inconsistent voice.  Sometimes, I express pessimism about our ability to have any but the smallest sliver of introspective knowledge.  Sometimes, I hint at optimism for the prospects of introspection more carefully done.  At still other times, I seem to stand perplexed about how much optimism or pessimism is appropriate.  Maja is entirely correct to notice this inconsistency.  The origin of this inconsistency is, I think, my perplexity plus my sometimes confessional approach.  I am genuinely unsure how much optimism or pessimism is warranted.  Because I employ confession to help induce the reader personally to share my doubts, I allow this perplexity to manifest as wavering opinion that shifts with the flow of argument.  So, for example, throughout Chapter Four, optimism reigns (about introspection properly done) as Mike Gordon and I build the case that ordinary people have pervasive auditory experiences of the shapes, positions, and textures of the silent objects around them.  At the end, though, having completed the case, I step back and allow myself again to fall into doubt.  My discussion of the flight of colors in Chapter Five concludes with considerations that seem to favor radical doubt about my knowledge of the color of my afterimages, alongside confession that those considerations leave me cold, though all I have against them is my probably untrustworthy epistemic intuitions – some hangover Cartesian feeling of indubitability.  I cannot resolve the question; I don’t feel the doubt that my argument seems to imply that I should feel.  But I don’t want only to pretend to doubt.  I would disrespect the reader if in my doubt-induction exercises I asked her authentically to assess her feelings of confidence or doubt but I masked my own.  So I am left with meta-level doubt: uncertainty about whether I should doubt, what Maja calls my perplexed voice.  To the extent this book is a treatise defending a single point of view, this wavering is an inconsistency of thesis.  But I prefer to think of the book as also partly an essay, confined within the bounds of its broad thesis – an essay in something like Montaigne’s (1580/1948) sense of an exploration that needn’t have a fixed conclusion but instead follows the rabbit trails of thought.

And I’m chicken.  Maja is right about that, too.  (To be clear: Maja doesn’t use exactly that term.)  My reasoning leads me to the abyss of pure dark skepticism and I cannot bear to leap.  No explicit reason holds me back – only “epistemic intuition” or “common sense” of dubious provenance.  I offer no positive reason to trust my introspective judgment that I am visually experiencing red as I look at this red object in good light or that I am experiencing pain as a pinch myself sharply on the back of my hand.  If so many other introspective judgments are false that I had originally thought secure, why not these, too?  I haven’t yet happened upon evidence that casts them positively into doubt, but perhaps that’s mere accident.  I do have a metaphilosophical excuse for this cowardice, though.  Although I mistrust intuition and common sense (see also Gopnik and Schwitzgebel 1998; Schwitzgebel 2013a), I also mistrust science and philosophy, in matters such as this.  So I cannot entirely abandon the one for the other – not that the three are entirely distinct, anyway.  I feel left only with very imperfect tools, which almost, but not quite, crumble in my hands.  My meta-skepticism partly tempers my skepticism: I am unsure how skeptical to be.

And that is why I find Maja’s positive project very attractive.  It promises to help reveal the bounds of introspective reliability in a principled way.  Her thought is that we have certain introspection-reliant abilities – real-world skills, like adjusting binoculars and ordering the right amount of food at a restaurant, that depend upon our making accurate introspective judgments.  We should expect introspective judgments to be accurate at least when they underwrite the successful deployment of such skills.  I needn’t simply fall back upon confessing an ineradicable intuitive confidence that I’m visually experiencing red right now; I can support that intuition empirically and systematically if I can find a measurably-successful behavioral skill that requires my introspective accuracy for its operation.  (For a rather different, but also promising, principled effort to limn the conditions of reliability and unreliability, see Bayne and Spener 2010.)

Despite the promise of Maja’s approach, though, I see a serious problem at its center: determining what qualifies as an “introspection-reliant ability”.  To paint the problem starkly, perhaps too starkly, consider epiphenomenalism.  If conscious experiences are causally irrelevant to behavior, then no behavioral skill, it seems, need require accurately detecting them.  All that would really be necessary would be to detect the “neural correlates” of the conscious experience, or some other regular accompaniment of it.  (This is presumably how Chalmersian (1996) zombies would avoid overeating.)

In order to display the full range of ordinary human skills, we need to have some sensitivity to our internal states.  But sensitivity to internal states is not yet introspection in the relevant sense.  Introspection must involve some sensitivity to one’s stream of conscious experience, if introspection-reliant abilities are going to do the work Maja wants them to do, that is, the work of establishing that we have good knowledge, at least in a certain range of cases, of our stream of conscious experience.  And it’s not clear that we need a specifically that type of sensitivity to order the right amount of food at a restaurant.  Maybe what I do when I accurately judge that I’m only a little hungry and so order only a small dish is introspect my phenomenology of hunger, detect that I am having an experience of only mild hungriness, and base my order on that.  But alternatively, maybe judgment is driven instead by proprioceptive sensory or quasi-sensory knowledge of the state of my stomach, plus knowledge of the time of day and what I have recently eaten, plus other sorts of physiological and contextual cues.  Indeed, the judgment that I am hungry seems somewhat separable from the judgment that I am experiencing the phenomenology of hungriness; I might judge the former while judging that I lack the latter and vice versa.  And to the extent they are separable, the former might be the more reliable and relied-upon guide to proper eating.

It’s 1:21 p.m. right now.  At noon I ate a burrito.  I’ve been drinking small amounts of coffee all day, but perhaps a bit less than usual.  As I sit here thinking about the phenomenology of hunger, I find myself growing confused.  I think about what I feel in the area of my stomach.  It feels not quite normal – maybe hungry?  I have some impulse to eat, but I’m unsure about its connection to the phenomenology in question; maybe it would be equally satisfied by drinking coffee instead?  My head feels a little foggy – undercaffeination, undereating (despite the burrito), poor sleep?  When I’m riotously hungry or seriously stuffed, I cannot mistake the gross contours of my phenomenal hunger or satiety – same when I am in intense pain or blasted by noise.  But in moderate cases, intermediate cases, and focusing carefully upon my hungriness phenomenology as distinguished from my desire to eat and my quasi-sensory knowledge of my body, and thinking about anything other than the grossest contours, I begin to feel my confidence slipping.  I don’t know quite what I am feeling.

I have fallen back again on epistemic intuition!  Maja might recommend that I consider science instead.  I would like to if I could.  It’s a scientific question at root, maybe: To the extent people are skilled at proper eating, is that skill driven by a psychological mechanism that centrally involves good introspective attunement to the phenomenology of hunger?  Or is it driven mainly in some other way?  Assuming we can set aside epiphenomenalist worries, maybe we could start angling in on this by looking at the mechanisms underlying our food choices and, especially, attempting to vary the hungry phenomenology while holding steady the non-phenomenal markers of hunger and the phenomenal associates of hunger that are distinct from the phenomenology of hunger proper, and vice versa.  But such knowledge might still be a long way down the road, and a variety of methodological worries might arise, including how we are to judge when we have successfully varied or held constant the hungry phenomenology, if we allow – as we must, if we are not to beg the question against the skeptic – that the phenomenal reports might not be accurate.  Richard D. Mattes has led the way in recent empirical research on the phenomenology of hunger, finding an array of problems in measurement and report, as well as tenuous connections between self-reports of subjective hunger and actual eating patterns (Friedman, Ulrich, and Mattes 1999; Mattes, Hollis, Hayes, and Stunkard 2004; Mattes 2010); and yet I’m inclined to think that even he is not fully sensitive to the nuanced differences that would be crucial to evaluating Maja’s view.

Similar issues arise for Maja’s case of adjusting binoculars to reduce their blurriness.  One might worry, at least in principle, about the difference between relying upon introspective knowledge of one’s visual phenomenology vs. relying upon a non-introspective sensitivity to states of one’s visual system correlated with that phenomenology.  A camera can autofocus with (presumably) no introspective attunement to its phenomenology whatsoever.  Maja’s argument for introspective reliability only works if introspective attunement to visual experiences of blurriness is carrying the causal load rather than guidance driven by something merely correlated with visual experience of blurriness; and fully to assess that question might require knowledge not currently within reach about relationships among visual processing, phenomenology, and visually-guided behavior.

But here at least – as not in the case of the phenomenology of hunger – I do find myself intuitively compelled to say that I know darn well how clear or blurry my vision is.  (Well, I do in most cases.  I have recently changed from monofocal spectacles to progressive corrective lenses, and I find myself sometimes confused about the sharpness of my vision in various combinations of eye-, head-, and spectacle-position.)  But now I wonder to what extent this judgment of blurriness is a phenomenal judgment about my visual experience vs. an outward-directed judgment about my optical environment.  The two seem rather hard to pull apart.

Consider other cases of optical reflection and refraction.  I see an oar half-submerged in water.  In some sense, maybe, it “looks bent”.  Is this a judgment about my visual experience of the oar or instead a judgment about the optical structure of my environment? – or a bit of both?  How about when I view myself in the bathroom mirror?  In the normal case during my morning routine, I seem to be reaching judgments primarily about my bodily condition and secondarily about the mirror, and hardly at all about my visual experience – not unless I am always reaching judgments about my visual experience to ground my judgments about the outside world (an introspective foundationalism I reject in Chapter 7 and in Schwitzgebel 2012).  Or suppose I’m just out of the shower and the mirror is foggy; in this case my primary judgment seems to concern neither my visual experience nor my body but rather the state of the mirror.  Similarly, perhaps, if the mirror is warped.  If I can reach such judgments directly without antedecent judgments about my visual experience, perhaps analogously in the binoculars case?  Or is there maybe instead some interaction among phenomenal judgment and optical or objectual judgment, or some indeterminacy about what sort of judgment it really is?  What if I am looking through a warped window or a fisheye lens, or using a maladjusted review mirror at night, or looking in a carnival mirror?  We can create, I’m inclined to think, a garden-path argument from arrayed intermediate cases, from the normal bathroom mirror case (maybe non-introspective) to the blurry binoculars case (maybe introspective), going in either direction, and thus into self-contradiction.  (For related cases feeding related worries, see Schwitzgebel 2013b.)

Another avenue of resistance to the binoculars case might be this.  Suppose I’m looking at a picture on a rather dodgy computer monitor and it looks too blue, so I blame the monitor and change the settings.  Arguably I could have done this without introspection: I reach a normal, non-introspective visual judgment that the picture of the baby seal is tinted blue.  But I know that baby seals are white.  So I blame my monitor and adjust.  Or maybe I have a real baby seal in a room with dubious, theatrical lighting.  I reach a normal, non-introspective visual assessment of the seal as tinted blue, so I know the lighting must be off and I ask the lighting techs to adjust.  Similarly perhaps, if I gaze at a real baby seal through blurry binoculars: I reach a normal, non-introspective visual assessment of the baby seal as blurry-edged.  But I know that baby seals have sharp edges.  So I blame the binoculars and adjust.  Need this be introspective at all, need it concern visual experience at all?  In the same way maybe, I see spears of light spiking out of streetlamps at night – an illusion, some imperfection of my eyes or eyewear.  When I know I am being betrayed by optics, am I necessarily introspecting, or might I just be correcting my normal non-introspective visual assessments?  This is a nest of issues I am currently struggling with, making less progress than I would like.  Maybe Maja is right, then.  Something in me thinks she is partly right.  But how much so, it will take further work to discover.


IV. Conclusion.

Thanks to Uriah, Declan, and Maja for their troublesome commentaries.[1]










Bayne, Tim, and Maja Spener (2010).  Introspective humility.  Philosophical Issues, 20, 1-22.

Chalmers, David J. (1996).  The conscious mind.  Oxford: Oxford.

Condillac, Etienne Bonnot de (1754/1982).  A treatise on the sensations.  In Philosophical writings of Étienne Bonnot, abbé de Condillac, vol. 1, trans. F. Philip.  Hillsdale NJ: Erlbaum.

Christensen, David (2010).  Higher-order evidence.  Philosophy & Phenomenological Research, 81, 185-215.

Friedman, M.I., P. Ulrich, and R.D. Mattes (1999).  A figurative measure of subjective hunger sensations.  Appetite, 32, 395-404.

Gopnik, Alison, and Eric Schwitzgebel (1998).  Whose concepts are they, anyway?  The role of philosophical intuition in empirical psychology.  In M.R. DePaul and W. Ramsey, eds., Rethinking intuition.  Lanham: Rowman & Littlefield.

James, William (1890/1981).  Principles of psychology.  Cambridge, MA: Harvard.

Mangan, Bruce (2001).  Sensation’s ghost: The non-sensory fringe of consciousness.  Psyche, 7 (no. 18).

Mattes, Richard D. (2010).  Hunger and thirst: Issues in measurement and prediction of eating and drinking.  Physiology & Behavior, 100, 22-32.

Mattes, Richard D., James Hollis, Dayle Hayes, and Albert J. Stunkard (2004).  Appetite: Measurement and manipulation misgivings.  Journal of the American Dietetic Association, 105 (supplement), 87-97.

Montaigne, Michel de (1580/1948). The complete essays of Montaigne.  Stanford: Stanford.

Schwitzgebel, Eric (2011).  Perplexities of consciousness.  Cambridge, MA: MIT.

Schwitzgebel, Eric (2012).  Introspection, what?  In D. Smithies and D. Stoljar, eds., Introspection and Consciousness.  Oxford: Oxford.

Schwitzgebel, Eric (2013a).  The crazyist metaphysics of mind.  Manuscript available at .

Schwitzgebel, Eric (2013b).  The problem of known illusion and the resemblance of experience to reality.  Manuscript available at .

Smithies, Declan (2012a).  A simple theory of introspection.  In D. Smithies and D. Stoljar, eds., Introspection and Consciousness.  Oxford: Oxford.

Smithies, Declan (2012b).  Mentalism and epistemic transparency.  Australasian Journal of Philosophy, 90, 723-741.

Strawson, P.F. (1959).  Individuals.  London: Methuen.


[1] And thanks to Janet Stemwedel for organizing the Pacific APA session on which this symposium was based.