Turning Descartes on His Introspective Head
August 23, 2011
Green Guy, Pete Mandik, 2003. Photograph by Rachelle Mandik
by Eric Schwitzgebel
Many philosophers consider the era of “modern” philosophy to begin with René Descartes’s Discourse on Method (1637) and Meditations on First Philosophy (1641). In these works, Descartes aims to ground human knowledge of the external, material world – the world of the newly arising modern science – upon a secure, indubitable foundation. That secure, indubitable foundation is knowledge of one’s own mind, one’s immaterial soul, the thinking self of Descartes’ famous dictum “I think therefore I am”. We cannot err about our own minds, Descartes asserts, if we limit our judgments to such matters as our current ideas and imaginings, our current intentions, emotions, and desires, and our current sense experiences – that is, our current stream of conscious experience or “phenomenology”. Knowledge of the external world, and all of science, is, according to Descartes, to be grounded upon our prior perfect knowledge of the soul.
Until the middle of the 20th century, philosophers tended to agree with Descartes about our perfect knowledge of the current contents of the stream of experience and thus our secondary or derivative knowledge of the world beyond that private stream. So, for example, in his Essay Concerning Human Understanding (1690), John Locke defends a broadly Cartesian epistemology: People have perfect knowledge of their current ideas, and this serves as the ground of knowledge in general. Much later, Bertrand Russell writes about “sense data” – the colors, sounds, smells, etc., that we experience – that “My knowledge of the table as a physical object… is not direct knowledge. Such as it is, it is obtained through acquaintance with the sense-data that make up the appearance of the table. We have seen that it is possible, without absurdity, to doubt whether there is a table at all, whereas it is not possible to doubt the sense-data” (Problems of Philosophy, 1912/1959, p. 47). Even David Hume, the great Scottish skeptic who doubted that 2 + 2 = 4, nonetheless found something it was impossible to doubt: “For since all actions and sensations of the mind are known to us by consciousness, they must necessarily appear in every particular what they are, and be what they appear” (Treatise of Human Nature, 1740/1978, p. 190).
Although in the latter half of the 20th century strict infallibilism about the stream of experience fell into disfavor, challenges to it tended to be indirect or tepid. The psychologists Sigmund Freud and Richard Nisbett, amongst others, convinced the academic community that people could be wrong about their motives, about the causes of their actions, and about the existence of hidden beliefs and desires, but such challenges do not directly undercut infallibilism about the stream of experience. It’s one thing to be wrong about why you chose a particular pair of socks, why you quit your job, or whether you secretly want to have sex with your mother, quite another to be wrong about whether or not you are experiencing a headache or seeming to see red. In his famous 1977 paper with Timothy Wilson, Nisbett even says, about current sensations, emotions, evaluations, and plans, that such “private events… can be known with near certainty” (p. 255). The contents of the Cartesian stream of experience are not the subject of Freudian or Nisbettian doubt. Some philosophers (e.g., David Armstrong, Paul Churchland, and Hilary Kornblith) did explicitly challenge infallibilism about such apparent Cartesian certainties as subjective pain and redness, but their challenges tended to be limited to far-fetched science fiction scenarios or wild delusion, or they tended to posit only minor or fleeting errors.
In a series of papers beginning in the year 2000, culminating in my current book Perplexities of Consciousness, I argue that this Cartesian picture is not just wrong at the margins but almost exactly backwards. Although it may be difficult to doubt the existence of severe, canonical pains and visual experiences of centrally presented uniform colors in standard conditions, such cases are not typical cases of introspective knowledge but rather unusual, best cases. To start epistemological inquiry by thinking about cases of that sort, and to generalize swiftly from there, is to rig the game – something like taking Babe Ruth and Hank Aaron as typical representatives of human skill at baseball.
Perplexities of Consciousness is composed mostly of detailed case studies of aspects of the stream of experience about which people seem to err, grossly and persistently, even in circumstances that might seem favorable to reflection. We aren’t just fallible in hard cases, I argue; we err systematically and pervasively about even the most basic facts of the stream of experience, and even when we set our minds to it carefully and conscientiously. Our knowledge of our immediate physical environment is much better than our knowledge of our stream of experience, and in fact to a large extent our knowledge of our physical environment is the ground of whatever knowledge we do manage to cobble together about our stream of experience. Not: I know what my visual experience is as I look at this tissue box and what my gustatory experience is as I eat this burrito and on the basis of such self-knowledge of the contents of my experiences I know the outward physical properties of this tissue box (that it’s blue and rectangular) and this burrito (that it’s cheesy and spicy). Rather, instead: I know that there’s a blue, rectangular tissue box out there and that the burrito in my hand is cheesy and spicy, and largely on that basis I conclude that I am experiencing blueness and rectangularity and cheesiness and spice. Even the words we use to describe the stream of experience – words like “blue”, “rectangular”, “cheesy”, and “spicy” – refer in the first instance to the properties of outward objects and only secondarily and derivatively to the properties of the sensory experiences that those objects typically produce.
Perchance to Dream, Episode 9, Season 1 of The Twilight Zone, CBS, 1959
Often we turn to better-known outward media and technology to try to make sense of our barely-graspable stream of experience – and in doing so, we are prone to err by over-analogizing, that is, by attributing to our stream of experience properties that rightly only belong to those outward media. One case of this constitutes a chapter of Perplexities. I trace the evolution of the view that people dream primarily in black and white. Before the 20th century, it almost impossible to find instances of purported black-and-white dreaming. In the early 20th century, there start to be some reports. By the 1940s the overwhelming majority of Americans say they dream almost exclusively in black and white, with at most the occasional splash of color here and there. Now, in the 21st century, most Americans say they dream primarily or exclusively in color. In early 21st-century China, Changbing Huang, Yifeng Zhou and I found a similar diversity across socio-economic status: low SES rural Chinese, with a history of access primarily to black and white film media (at the time of study) reported dreaming mostly in black and white; high-SES urban Chinese, with access primarily to color film media, reported dreaming mostly in color; and intermediate-status urban Chinese, with intermediate media access, reported intermediate rates of color dreaming.
One possibility is that the media profoundly influence our dreams, so that the typical American in 1950 and rural Chinese in 2005, exposed primarily to black-and-white film media, really does dream primarily in black and white, while the typical American and wealthy Chinese of the 21st century dreams primarily in color. There are several reasons to be doubtful about this possibility, but here is one: The rates of color term use in dream reports collected upon awakening did not change in the United States between 1950 and 2000. Americans were just as likely to use terms like “orange”, “green”, and “purple” to describe features of their dreams in 1950 as in 2000 – though in the 1950s they insisted that such colors were only occasional splashes in a sea of black and white.
More likely, I think, than the possibility that media exposure radically alters the coloration of one’s dreams is the possibility that dreams remained largely the same over time, and it’s only the reports that changed. If so, at least one group is radically mistaken about the coloration or not of their dreams. I don’t attempt to adjudicate which group is mistaken, but I suggest that one interesting possibility is that practically everyone is mistaken, and that dreams are neither predominately colored nor predominately black and white – that the attribution of pervasive coloration or black-and-whiteness to dreams is a mistake, much as it would be a mistake to attribute pervasive coloration or black-and-whiteness to a novel that left the coloration of most of its objects unspecified. In that case, it would not be only Americans in 1950 and the rural Chinese in 2005 who were over-analogizing dreams to movies, but all of us. Movies are all, or nearly all, either black and white or colored, but it doesn’t follow that dreams must be.
Dreaming may not be the best case for the Cartesian inversion argument since judgments about dream coloration are (mostly) retrospective and Cartesian infallibility pertains only to currently ongoing experience, but it’s a nice place to start since it brings out the sources of error so clearly. In Perplexities I also discuss other cases: One is the claim, common both among philosophers and among ordinary Americans, that tilted pennies look elliptical and distant objects look small, a view that I suggest derives in part from over-analogizing visual experience to flat, projective film media like paintings and photographs. Another case I put forward (in a chapter with Mike Gordon) is that people’s auditory experience is massively echolocatory – that we have pervasive auditory experience of silent objects, much as bats and dolphins do – though few people know this about ourselves. In another chapter I consider whether people have constant tactile experience of their feet in their shoes and of the hum of the refrigerator; I suggest that the question may be methodologically irresolvable, at least in the medium term, and that unless it can be resolved there is no hope for arriving at a well justified general theory of consciousness.
The modern and contemporary philosophical tradition, which has emphasized the specialness and security of self-knowledge, especially self-knowledge of the stream of conscious experience, and in comparison the relative insecurity or derivativeness of our knowledge of the physical world around us, has the epistemic situation upside-down.
About the Author:
Eric Schwitzgebel is a Professor of Philosophy at the University of California at Riverside. He is author of two books: Perplexities of Consciousness (MIT Press, 2011) and, with Russell T. Hurlburt, Describing Inner Experience? Proponent Meets Skeptic (MIT Press, 2007). He also studies the moral behavior of ethics professors.