Is Naive Realism the Cure for Postmodernism?
Quee Nelson, The Slightest Philosophy. Indianapolis: Dog Ear Publishing, 2007.
Quee Nelson is a classical liberal who has written a book on metaphysics and epistemology, well received by many libertarians. Several libertarians told me it was a book I ought to read. It has received numerous compliments but, as far as I know, no serious discussion. I very much disagree with the main thrust of Quee’s argument and in this review I will try to explain why I disagree.
Quee’s book is entertaining, humorous, unpretentious, readable, and displays evidence of a lot of reading and thought. It’s a book about philosophy but one that clearly aims to include among its readership people who have not read much philosophy. This present review is the same kind of thing; it’s not the sort of review you would expect to find in a philosophy journal, but rather one on about the same level as Nelson’s book, that is, one which takes nothing for granted, which is inclined to ramble, and which does not hesitate to ‘naively’ address some elementary points.
I see this review as a kind of cognitive therapy for libertarians who think like Quee, but unlike most therapy, I am offering it for free. So, you have a real bargain here, and libertarians are supposed to love a bargain. I also see this review as an encouragement to readers to get involved in the marvelous intellectual adventure of philosophy, which entails coming to grips with real arguments at their strongest, and understanding why these arguments do convince intelligent people, rather than pulling faces at conclusions you don’t like.
Nelson claims to adhere to an unpopular, minority view in philosophy—naive realism. I adhere to a different unpopular, minority view in philosophy—critical rationalism. Critical rationalism may be better known as Popperianism, though there is a well-known law of nature that all Popperians have at least one major disagreement with Popper, so we should prefer the more impersonal term, ‘critical rationalism’.
However, on most of the issues covered by Nelson’s book, I adhere to the conventional, consensus, or default position among present-day English-speaking philosophers—representative realism. So, most of the time, I will be defending a fairly mainstream philosophical position against Quee Nelson, though occasionally I will come up with a critical rationalist twist.
Here are the main points I will be making in what follows:
1. Nelson calls herself a naive realist but never lets us know what ‘naive realism’ is.
2. Nelson misrepresents mainstream academic philosophy by claiming it is completely dominated by anti-realism.
3. Nelson mistakenly claims that postmodernism is rooted in skepticism about perception (and that it is derived from Hume and Kant).
4. Nelson doesn’t understand the force of the arguments of Hume and Kant.
5. Nelson mistakenly claims that idealism is a powerful movement in present-day English-language philosophy.
6. Nelson relies upon an argument against representative realism which is purely semantic and therefore inconclusive.
7. Nelson advances a theory about the historical causation of political outcomes by philosophical ideas, which is full of holes.
Naive and Representative Realism
Nelson tells us that she is arguing for “naive” or “vulgar” realism (pp. 2–3). She says she prefers the term ‘naive realism’ to ‘direct realism’, because she thinks that the latter could be taken to deny the complex causal chain involved in perception (pp. 10–12). But other philosophers who advocate what they call ‘direct realism’ don’t deny this at all.
David M. Armstrong argues in favor of direct realism in his 1961 book. As far as I know, this is the best case ever made for direct or naive realism, but although Nelson mentions this work (p. 9), she strangely does not say how much of it she agrees with, or whether Armstrong’s direct realism is the same, or roughly the same, as her naive realism. This is part of a general problem, that Nelson’s actual position, the delineation of what she calls naive realism, is elusive. The reader can only be puzzled as to what Nelson’s naive realism is.
All forms of realism agree that physical entities, such as tables, chairs, rocks, trees, stars, and clouds, exist independently of our minds. Disputes between representative realism (representationalism) and naive (direct) realism have focused on the question of whether, when we see an object such as a tree, we do so by means of seeing a mental representation of that object, or whether we don’t see any representation but only see the object itself. (I don’t approve of that way of framing it, but naive realists usually do frame it in some such way.)
A different distinction is that between common-sense realism and scientific realism. Some people think there’s a troubling conflict between these two. Common-sense realism is the view that the things we suppose we observe as we look around in everyday life exist, independently of our awareness of them. Scientific realism is the view that the entities described by physics and other natural sciences exist, independently of our awareness of them.
I don’t see common-sense realism and scientific realism as competing alternatives. My view is that where common sense and science clash, science is probably right and common sense even more probably wrong. So here my view is contrary to that of Nelson, who thinks that common sense trumps physics (pp. 7–8).
Common sense is not fixed. Today’s common sense among educated people is partly a product of science, or of the scientific modification of more old-fashioned common sense. It used to be common sense that iron boats could not float, and when gas lighting was first introduced, many people couldn’t believe that the pipes carrying the gas did not heat up. Common sense is an assemblage of theories, a product of culture, it consists of memes, it is inculcated into individuals largely by the use of language, it varies among cultures and among sub-cultures, it has evolved over the centuries, and it is always further revisable. Common sense often contains valuable suggestions and it should not be ignored, but it carries no authority.
It would be nice to be able to state Nelson’s own characterization of naive realism and proceed from there, but unfortunately this is not straightforward. She tells us (pp. 2–3) that naive realism is the view that the things we perceive “comprise” an external universe which doesn’t depend on our perception. This implies that cosmic rays, magnetic fields, and dark matter are not part of this universe (they either don’t exist or belong to a different universe).
We can probably assume that this is not what Nelson intended; what she might have meant is that the things we perceive are parts of a universe which also contains many other entities. But this also is unsatisfactory, because this definition would apply to all forms of realism, representative as well as naive. So, this definition would not identify what’s peculiar to naive realism. We never do learn how Nelson wants to define her own naive realism, so as to distinguish it from common or garden representative realism.
Again and again, she seems as though she’s just about to define naive realism, or her version of it, but then she simply defines realism, in a way which would include representative realism. To take just one example, she says that naive realists like herself have an “unwavering faith in the actual existence and intractable mind-independence of locomotives” (p. 10). Yet, allowing for some uneasiness about the word “faith”—but let’s not quibble—this is just as true of representative realists as of naive realists.
The closest Nelson comes to criticizing representative realism is with the brief section headed “The Irrelevance of Representationalism” (pp. 12–15). Here she complains that many different philosophers have advocated many different conceptions of whatever it is in the mind of the perceiver that links the perceiver with the perceived object. She complains about the profusion of terminology as well as definitions. And she says this doesn’t really matter, it’s a “technical side show,” because all that “really matters” is realism versus anti-realism, the question of whether perceived objects exist independently of the perceiver’s mind. But if you’re claiming to advocate naive realism, and you disparage its major alternative, representative realism, and many of your opponents are representative realists, it’s incongruous to say that the issue of representative realism doesn’t matter.
In another brief discussion of representative realism (pp. 5–7), Nelson addresses only the question of color realism. In fact, someone reading this passage with no previous knowledge of these issues might easily conclude that the distinction between naive and representative realism lies entirely in the color question. Galileo, Locke, and many others, have held that whereas some aspects of objects like shape and size are really ‘in’ the perceived objects, aspects like color, smell, and sound are generated in the mind of the human observer.
Today almost the only philosophers who discuss color realism at any length are those specializing in this issue, and most of them take the view that color is a real property of objects (see for instance the article by Johnson and Wright). However, this, if correct, would not entirely dispose of the color question, for there are certain apparent facts about colors (such as ‘There can’t be a reddish shade of green’) which, taken naively, seem to be undeniable facts about objective colors, but are in fact (so science tells us, and here I believe science) entirely due to the ‘design’, or specific structure, of our body’s perceptual apparatus. As Günter Wächtershäuser said, there’s more to vision than meets the eye.
The historically most influential form of non-realism (unless you count Plato) was idealism, classically exemplified in George Berkeley, and fashionable at the end of the nineteenth century in the version known as ‘absolute idealism’, as taught by F.H. Bradley, J.M.E. McTaggart, and Josiah Royce. Idealism claims that all our knowledge is essentially mental and subjective, and therefore the only things we can know about are ultimately products of the mind.
Idealism has little following today. Opposition to realism mostly comes from cultural relativism or social constructivism, sometimes lumped together as ‘postmodernism’. Postmodernism has very little following among philosophers, but it has a huge following—it is almost the reigning orthodoxy—among academics in literary and ‘cultural’ disciplines.
Nelson conveys the impression, once or twice by direct assertion but much more often by insinuation, that non-realism is the dominant position among anglophone philosophers. But this is mistaken; probably most philosophers (meaning faculty and grad students in university philosophy departments) are realists. I will mention some indications of this fact, and of how Nelson misrepresents the current state of academic philosophy, as this review proceeds.
To avoid possible misunderstanding, I should add that philosophy as an academic discipline has become so specialized that many philosophers never have occasion to address metaphysical issues like realism, and also that some people I would classify as representative realists may not call themselves by that term. Representative realism is such a wide and general category—the obvious default category—that some adherents may not see the need for any identifying label.
For the sake of brevity and simplicity, I’m going to cover just two forms of realism—representative realism (representationalism) and naive (or direct) realism. I’m not going to offer my own distinction between naive realism and direct realism, as some writers do, but will treat these as equivalent. Nor will I give separate attention to different aspects of realism, such as truth, objectivity, and mind-independence. Since I am reacting to what Quee Nelson says, I completely neglect a number of important arguments and distinctions which don’t arise in her discussion.
In this review I’m concerned only with realism about perceived physical entities. Realism about moral or aesthetic matters would introduce a lot of additional considerations. Many people are realists about physical objects and non-realists about morality or aesthetics. When I use the term ‘skepticism about perception’, this is short for ‘skepticism about perception as informing us of a world of things independent of our minds’.
Misrepresenting Today’s Academic Philosophy
‘Realism’ may be crudely stated as the view that the universe is largely composed of entities which exist independently of any human awareness of them. For instance, if all conscious minds were to be wiped out, the stars and planets would continue to exist. This is a view I hold, in common with most philosophers.
The mainstream view in English-language philosophy is that perceived objects do independently exist, and this has been the mainstream view since about 1910–1920, by which point the formerly dominant ‘absolute idealism’ had begun to be abandoned, mainly due to the work of Bertrand Russell and G.E. Moore.
To a very large extent, current controversies about realism versus non-realism take the form of a battle between philosophers and non-philosophers. Realism has its stronghold in university philosophy departments while anti-realism has its stronghold among non-philosophers in humanities disciplines such as literary theory, sociology, and culture studies.
Nelson conveys the impression that academic philosophy is a bastion of non-realism. This is not true. She states that “a skeptical anti-realism” is “still more or less in the driver’s seat” (p. xii). In the world of English-speaking academic philosophy, no form of skeptical anti-realism has been in the driver’s seat since the 1890s.
Nelson not only falsely identifies anti-realism with mainstream philosophy, but also falsely roots present-day anti-realism in skepticism about perception. Skepticism about perception was originally at the root of idealism, the philosophical movement which had its heyday in the nineteenth century. Today’s anti-realism is normally rooted in cultural relativism and social constructivism, tendencies extremely popular among people in non-philosophy humanities disciplines and decidedly unpopular with philosophers. Cultural relativists and social constructivists rarely (if ever) make arguments which appeal to skepticism about perception.
The Professor and the Student
After the first two chapters, Nelson develops her argument by means of a dialogue between a “Student” and a “Professor.” She identifies with the Student, while the views she opposes are identified with the Professor. Her Professor is testy and dogmatic, shifty and evasive, making feeble arguments with a display of arrogance, and frequently saying things that are blatantly ignorant or silly, while her Student embodies sweet reasonableness, judicious fair play, encyclopedic erudition, and wisdom beyond his tender years.
The views preached by Nelson’s Professor are, taken in their totality, views which no one holds. They are views made up by amalgamating different philosophical doctrines (or selected portions of these doctrines) which Nelson doesn’t like, and which are unlikely to be simultaneously held by the same person. You will never find anyone who is simultaneously a Berkeleyan idealist, a Kantian, a post-Kantian, a Hegelian, a phenomenalist, a postmodernist, a Kuhnian, a pragmaticist, and a pragmatist, but Nelson’s imaginary “Professor” is such a chimera. In fact you would be extremely unlikely to find anyone who combines even two of these, though I admit that conceivably could happen. On almost every page, the Professor says something that no professional philosopher would ever say.
The net effect of this portrayal of the Professor is to further emphasize Nelson’s misleading claim that anti-realism is the dominant, orthodox, or mainstream view. Since Nelson purports to be arguing for naive realism, it would be more appropriate to have the Professor as a representative realist, or perhaps to have five interlocutors, representative realist, naive realist, idealist, phenomenalist, and postmodernist—with the understanding that idealism was included as a historical curiosity.
The Specter of Postmodernism
Nelson begins her book by talking for some pages about postmodernism. Why does she do this?
There’s no agreement about the definition of ‘postmodernism and I won’t try to come up with an authoritative definition. I will say that postmodernism in philosophy, postmodernism in the arts, and postmodernism in discussions of popular culture, are often very different, and attempts (including attempts by some of their followers) to represent them as being aspects of a single movement don’t work. The word ‘postmodernism’ has different usages, and in some of the more popular areas, it may have little, if anything, to do with non-realism or non-objectivity.
In philosophy, postmodernism is usually taken to refer to several French philosophers (including Michel Foucault, Jacques Derrida, Jean-François Lyotard, and Jean Baudrillard), but these writers don’t follow a single party line, and it’s hard to arrive at a single doctrine which they all advocate. As an example of variation in the use of the term, the ‘postmodernist’ writings of the cultural-Marxist (or perhaps former cultural-Marxist) literary theorist Fredric Jameson, immensely influential among students of literature and popular culture, advance a concept of postmodernism which has little in common with that of the French writers labeled postmodernist, or with the concept of postmodernism as equivalent to social constructivism and cultural relativism, or with any kind of non-realism.
Philosophical postmodernism is often identified with the claims that ‘reality is socially constructed’, ‘truth is culturally relative’, and ‘there is no truth, only various interpretations’. Here I’m going to accept these positions as a rough working definition of philosophical postmodernism, which seems to broadly agree with the way Nelson uses the word. In this sense, postmodernism can be seen as a form of non-realism, since it denies that there is a single objectively true account of facts. But postmodernism is very different from old-fashioned idealism.
Use of the word ‘postmodernism’ to denote a style or movement of philosophy dates from the 1970s, but postmodernism’s influence on American literary and culture studies began earlier, when it, or something like it, was called post-structuralism, deconstruction, social constructionism, or constructivism.
There’s no dispute that some writers commonly labeled postmodernist have said things which cause our eyebrows to elevate. To take the most famous example, when archeological researchers, after examining the mummy of the Egyptian pharaoh Rameses II, concluded that he had probably died of tuberculosis, the French philosopher Bruno Latour, sometimes described as a postmodernist, objected that this finding must be wrong, because tuberculosis hadn’t yet been ‘constructed’ by medical science in ancient Egypt, and therefore couldn’t have existed at that time!
You misunderstand the situation in philosophy today if you don’t appreciate that the majority of philosophers, including French philosophers, view Latour’s assertion as totally hilarious, just as I do or just as Nelson (presumably) does. Notice that the way of thinking that leads Latour to suppose that tuberculosis didn’t exist until the medical concept of tuberculosis was ‘constructed’ is not like old-fashioned idealism and has nothing to do with skepticism about perception.
Scholars in literary and ‘cultural’ disciplines have lower IQs and less exacting standards than philosophers, and these disciplines are happy hunting grounds for Marxism, Freudianism, postmodernism, and other fanciful belief systems currently rejected by philosophers. (I wish I could add feminism and critical race theory, but I have to acknowledge that even professional philosophers are often susceptible to these unsightly conditions.) These disciplines taken together have a much higher head count than philosophy departments. (Last time I looked, some years ago but I doubt it has changed much, faculty and grad students in philosophy departments in the US amounted to about 7,000, whereas disciplines such as history, sociology, psychology, religion, and ‘culture studies’ each amounted to several times that number—I’m including women’s studies, African American studies, and so forth, among “culture studies.”)
Postmodernism is one of a succession of French philosophical tendencies, beginning with existentialism in the 1940s, which came into anglophone, mainly American, literary theory and from there into more popular discussion, largely bypassing anglophone philosophy. Generally speaking, these tendencies had proportionately far more support from American non-philosophers than from American philosophers or even from French philosophers. Most French philosophers were never existentialists and never post-modernists.
So, the recurring pattern is that a trendy but distinctly minority ‘coterie’ movement within French philosophy is transmitted into American literary and ‘culture’ disciplines, gets media attention, is swallowed by pundits, educational bureaucrats, and other ignoramuses, and is resoundingly rejected by American and British philosophers, who then occasionally offer criticisms of that tendency. (Perhaps even before World War II, Bergsonianism might have followed much the same pattern, but I don’t know enough to be sure of that.)
Nelson says: “Unfortunately postmodernists didn’t get that way on account of ignoring the teachings of the philosophy department, but on account of sincerely imbibing them. The terrible truth is that postmodernism is what happens when honest, intelligent people read the canonical philosophers and believe them” (p. x). This “terrible truth” is, at best, a wild and unsupported surmise, and Nelson offers no corroboration for it. But it does explain why Nelson begins her book with postmodernism even though the great majority of her book is not about postmodernism and the great majority of philosophers are not (in any sense) postmodernists.
Her claim is that postmodernism (though she acknowledges that most philosophers reject it) follows from what most philosophers believe, and that is, in her account, Hume’s and Kant’s views on perception. But it is not remotely accurate that postmodernists became postmodernists because of the Humean-Kantian views on perception they learned from “the teachings of the philosophy department.”
If it really were the case that postmodernism were due to Hume and Kant, we might wonder why the great majority of philosophers, familiar with Hume and Kant, have no time for postmodernism, while people in literary and ‘culture’ fields, knowing nothing about Hume or Kant, subscribe to postmodernism. We might also wonder why postmodernism waited till the 1960s to put in an appearance, instead of beginning in the eighteenth century.
The Downfall of Idealism
Indeed, we might take our curiosity about history further and wonder why non-realism in the form of idealism dominated English-language philosophy until the 1890s, and was then rapidly dislodged from this dominant position by Russell and Moore, since when representative realism (in various forms) has had considerably more influence. (Probably the major rival of representative realism would be phenomenalism, which I will not pursue here, except to say that it owes something to Hume and nothing to Kant.)
The story of how the dominance of idealism in anglophone philosophy was (rather dramatically and suddenly) overturned is told in Peter Hylton, Russell, Idealism, and the Emergence of Analytic Philosophy. It’s true that Russell later moved to a position known as ‘neutral monism’, an attempt to avoid both idealism and realism but which Karl Popper claims is fundamentally similar to idealism (Realism and the Aim of Science, pp. 90–91). However most anglophone philosophers didn’t necessarily accept neutral monism and probably continued to embrace some form of representative realism. Many of them became materialists. We can define materialism (or ‘physicalism’) as realism plus the view that reality consists entirely of what used to be described as ‘matter in motion’ but is now more fashionably rendered as ‘particles in fields of force’.
A good insight into what happened to philosophy in the English-speaking world is provided by A.C. Ewing’s fine book of 1934, Idealism: A Critical Survey. Ewing was a realist, at a time when avowed idealism had become a rapidly dwindling minority among philosophers. He wanted to explain just what the disappearing idealism had been and why it was demonstrably incorrect, while preserving certain valuable insights he believed some idealists had contributed. Ewing’s book is a respectful autopsy on idealism; it gives a meticulous account of the arguments which had led earlier philosophers to embrace idealism, and the more recent counter-arguments which had led them to abandon idealism.
Many of these counter-arguments were not so much demonstrations that idealism was wrong as demonstrations that arguments in favor of idealism were flawed. Remember, Aristotelian or syllogistic logic had recently been replaced by modern logic, and this was a big deal.
One thing we should be clear about is that, historically, idealists were no less respectful of the objectivity of truth than realists. They didn’t suppose that they could make up the truth about reality according to their taste, or that any theory was as good as any other. They thought that the only reality we could know was constructed by our minds; they did not think that we had any discretionary control over the way this happened. Confronted with the contention that the Copernican account of the solar system is no more or less objectively accurate than the traditional account of some hunter-gatherer tribe (a view now commonly held in literary and culture-studies circles), Bradley, McTaggart, or Royce would have had pretty much the same response as Russell, Quine, or Searle, including astonishment that any functioning biped could countenance anything so ridiculous.
Idealism held that we’re not free to choose for ourselves the way in which the mind shapes reality: this is something involuntary, determined independently of our will. Idealists and realists would agree completely on the facts of astronomy, mechanics, or medicine; it’s just that idealists considered these facts to be inescapably and irreducibly molded or ordered by our minds, while realists maintained they were descriptions of a reality which was independent of our minds—though of course the descriptions themselves were products of our minds.
In one sense, idealism is the diametric opposite of postmodernism, because idealism holds that the necessary structure of the mind—the same for all minds and indeed for all possible minds—determines how we must inescapably conceive of the world, while postmodernism holds that different and contradictory ways of conceiving the world can be equally valid (a view that would have been quite baffling to idealists). This diametric opposition was understood by at least some of the originators of postmodernism, who deliberately included Kant and Hegel among the ‘moderns’ they were repudiating.
A Misleading Depiction
One of the misleading things about Nelson’s account is that she supposes that because Hume and Kant are ranked by knowledgeable people as outstanding philosophers, therefore philosophy students are encouraged to read them uncritically. This is ludicrously far from the case. No one is taught Hume or Kant in a philosophy course without being given a barrage of standard objections to their arguments. The student will be told about naive realism, representative realism, and perhaps two or three forms of non-realism, the points in favor of each of these positions and the points against. There will be explanations of Descartes, Locke, Berkeley, Hume, and Kant, including criticisms which explain why much of these authors’ work is not accepted by philosophers today.
Kant’s Critique of Pure Reason is acknowledged to be difficult (though no more difficult than, say, first-year college physics minus the math; Nelson’s contention, p. 143, that no one can make any sense of it is mistaken), and the student will probably use a short published ‘commentary’ or introductory guide. A short commentary will draw attention to things in Kant which can’t be accepted, things which are problematic because of specified objections, things where there is dispute about exactly what Kant meant (with attention to the vulnerability of what he meant on each interpretation), and so forth.
Anyone who regularly talks to a lot of faculty and grad students in philosophy departments knows that postmodernists are very thin on the ground, whereas they’re very thick in literary and ‘culture’ disciplines. The year before Quee’s book appeared, a little book by Paul Boghossian came out, Fear of Knowledge, straightforwardly debunking “relativism and constructivism” from a very conventional philosophical standpoint.
Here Boghossian points out that “anti-objectivist conceptions of truth and rationality” are generally rejected within academic philosophy, and as a result, there has been “a growing alienation of academic philosophy from the rest of the humanities and social sciences, leading to levels of acrimony and tension on American campuses that have prompted the label ‘Science Wars’” (Fear of Knowledge, p. 8).
Despite its simplicity and brevity, Boghossian’s book was favorably reviewed in prestigious philosophy journals. The review by Harvey Siegel concludes: “Boghossian has wise things to say concerning the contemporary split between ‘academic philosophy’, which by and large rejects the target views [relativism and constructivism], and the rest of the humanities and social sciences, which, unfortunately in Boghossian’s view as in my own, are far more welcoming of them.”
The truth is that contemporary philosophy tends to be realist and philosophers suffer because of their opposition to the fashionable anti-realism prevalent in other humanities disciplines. Meanwhile, Nelson spreads the story that mainstream academic philosophers are responsible for the non-realism of these non-philosophers, a story which is some considerable distance from the truth.
Nelson maintains that postmodernism derives its anti-realism from Hume and Kant and in general from skepticism about perception. But if you look at the arguments proponents of postmodernism offer for their anti-realism, you find that they appeal to cultural relativism and social constructivism, not to skepticism about perception, and if you look at current philosophical critiques of postmodernism, such as Boghossian’s Fear of Knowledge, you find that they barely mention skepticism about perception.
To take another example, in his excellent little introductory book on metaphysics, Peter van Inwagen, a realist philosopher well acquainted with non-realist thinking gives an account of idealism, the arguments for it and against it (pp. 58–67), and here discusses skepticism about perception, and separately he gives an account of modern non-realism, what I have been calling postmodernism, the arguments for it and against it (pp. 93–108), and here he doesn’t mention skepticism about perception.
I have said that Nelson gives the false impression that anglophone philosophers are predominantly non-realist. Mostly she does this by innuendo and rhetorical spin, but on page xi, she offers two pieces of direct evidence for her claim.
The first is a reference to John Heil as cited in Michael Devitt’s book Realism and Truth. Heil reported in 1989 that the number of current books advocating anti-realism exceeded the number of pro-realist books. This doesn’t tell you how many philosophers belong in one camp or the other, and the realist Devitt seems to acknowledge (p. xii) that his earlier impression that anti-realism was “rampant” was mistaken.
The second piece of evidence Nelson reports as follows: “One of the latest books from Oxford University Press still assures us of “our epistemological enlightenment, where we have corrected our ordinary, naive view, and accepted that external items are not accessible to sense-perception.” Here Nelson conveys the impression that the stated view is alive and kicking among philosophers, and perhaps that being published by Oxford University Press is a seal of approval for a work’s conclusions.
The book referred to here is by John Foster (1941–2009), well known as that peculiar and possibly unique anomaly, a contemporary advocate of Berkeleyan idealism. Oxford University Press publishes dozens of philosophy books every year, few of them advocate anti-realism, and almost none of them advocates idealism. John Searle’s 2015 book advocating direct realism (naive realism) was also published by Oxford University Press.
If you do a quick online search for reviews by philosophers of Foster’s books, you’ll easily find half a dozen, and if you read them you’ll find that almost all the reviews mention, in passing, the odd, bizarre, or unfashionable nature of Foster’s idealist position.
So, Nelson’s reference to Foster is misleading, and it is part of a seriously misleading pattern.
Where Hume and Kant Were Coming From
David Hume (1711–1776) thought that “philosophy” showed that belief in a ‘real’ world, existing independently of our awareness of it, was unfounded and indefensible. Since Hume was by temperament a hard-headed Scot, he found himself unable to accept this conclusion. He never did accept it, and he spent much of his life discussing the world based on the assumption that realism is true.
As Hume himself puts it, though “profound and intense reflection” leads to skepticism anent the world of independently-existing physical entities, “carelessness and inattention” come to the rescue, and anyone briefly convinced by skeptical arguments will find himself returning to realism within an hour (A Treatise of Human Nature, p. 218).
This is often called “the Humean shrug.” Though he believes that “philosophy” demonstrates that realism is indefensible, he thinks we have no alternative to accepting realism, even though we can only do so on completely slapdash and illogical grounds, because our rational faculties are overpowered by habit and short attention span, which automatically cause us to accept realism as a practical matter. Hume found this conclusion unwelcome, but he couldn’t see any way out, and he went on to take realism for granted in all his writings on religion, history, society, and economics.
Hume’s starting point is empiricism in the strict sense. Empiricism in the strict sense is the view that all human knowledge comes from experience or observation, which involves relying on the evidence of our senses. We accumulate knowledge by observing the world around us and by performing logical operations on our observations. We have sensory experiences, and from these we deduce the existence of tables, chairs, mountains, stars, and the rest of it. We start with an empty mind, a ‘tabula rasa’ (blank slate), and anything that gets into our mind gets there from observation, and therefore comes through our sense organs, such as our eyes and ears.
‘Empiricism’ in the looser, everyday sense means that we should take empirical evidence very seriously. All schools of philosophy are empiricist in this platitudinous sense, and from now on I will use the word ‘empiricism’ to mean only strict or empty-mind empiricism.
The impulse behind empiricism is the conviction that our view of the world is, or ought to be, derived from evidence about the world, and should not be prejudiced by gratuitous preconceptions. Since our evidence about the world can (according to empiricism) only be the information we get through our senses, our view of the world has to be derived from what our senses tell us, and from that alone.
Empiricism thus gives rise to the empiricist project or challenge: show that our common-sense or scientific ideas about the world are or can be derived from our observations of the world, and from nothing else, by a process of pure deduction.
Hume concludes that this cannot be done, that we can’t get, by any rationally defensible method, from accumulated observations (or sensory experiences) alone to a common-sense theory of the world (involving material objects, space and time, arithmetic, cause and effect, and so forth). For example, no observation of the world can ever, by itself, give us good evidence for causation. Hume acknowledges that some truths, such as 2 + 2 = 4, can be established by logical analysis, without any appeal to experience. But these are what Hume calls “relations of ideas,” not specific claims about material reality.
Since empiricists normally start out by wanting to be able to accept realism, empiricists down the centuries have labored long and hard to come up with a defensible way to reason from the evidence of the senses to the existence of physical entities (from now on let’s stipulate physical objects, except where otherwise stated, as the easiest type of physical entities to talk about).
Unfortunately for the empiricist project, once we accept that everything has to be deduced from the evidence of our senses, what ultimately must follow is that all our knowledge about the world is inferred from or constructed from our sensory experience. But since all our experiences are necessarily subjective and mental, this seems to imply that our view of the world is composed of elements that are subjective and mental. Thus, empiricism has sometimes led to idealism, the view that the world (or of any aspect of the world we can think about and talk about) is itself made up of subjective mental elements.
Some philosophers, to this day, agree with Hume throughout (only in broad outline, of course; all Hume scholars accept numerous detailed criticisms of Hume). They are empiricists who agree that empiricism shows realism to be rationally indefensible and they agree that we are in practice bound to accept realism, and thus they defend the “Humean shrug.” Hence the realist Willard Quine’s remark that “The Humean predicament is the human predicament” (quoted by Nelson, p. 38). Others are still trying to find a way to rehabilitate empiricism by reasoning from observations alone to an objective world of physical objects, or in other words, by demonstrating that induction can be valid. Some still hope to refute Hume’s demonstration of the impossibility of valid induction by deriving induction from Bayes’s theorem (a well-known theorem in probability theory). Good luck with that.
Quee’s Student and Professor make a total hash of Hume’s problem of induction. The Student comes out with inane remarks, and the Professor, since he is just Quee’s other glove-puppet, has no idea what to say and burbles irrelevantly. Nelson’s Student triumphantly asserts that something can be logically possible but physically impossible (p. 210), as though this were something Hume hadn’t thought of! Of course, Hume’s point here is precisely that the conclusion that something is physically impossible can never be deduced from observations alone.
Where do we get the notion that anything is physically impossible? It’s not a truth of logic, so according to empiricism, it must be derived from observation. What Hume has seen is that the claim that something is physically impossible is a conclusion supposedly derived from a finite number of instances, applied to an infinite number of instances. As we might say today, it is a conclusion derived from an infinitesimal bit of spacetime, applied to all of spacetime. Any such supposed inference is deductively invalid. (It is invalid according to modern logic and it is also invalid according to the more primitive and incomplete Aristotelian or syllogistic logic known to Hume.) So, Hume’s question is: What’s the basis for this conclusion, since it is not a truth of logic nor a logical inference from observations? A universal generalization, such as a scientific law or a piece of folk wisdom, can never be deduced from observations alone.
Hume demonstrates the incompatibility of empiricism and realism, but Hume doesn’t address the fact that this conclusion leaves it open which of these is to be discarded. He seemed to take for granted that empiricism is equivalent to “philosophy,” or at least, to high-quality philosophy. However, instead of rejecting realism, we can consider rejecting empiricism.
If empiricism be discarded, then we don’t have to start with an empty mind which is then filled with information from observations. We can start with a mind which, before it experiences anything, is already furnished with preconceptions or expectations. According to this approach, the human mind is not a blank slate at birth; it has plenty already written on it (by billions of years of natural selection, but neither Hume nor Kant knew that), and without that stuff that’s already written on it, the mind would be unable to form a picture of the world.
If this is right, then to defend realism requires abandoning empiricism. One way to abandon empiricism is to say that the mind is not tabula rasa but has built-in preconceptions. We can’t form a mental picture of nature without putting something of ourselves into our picture of nature, right from the get-go. This is what Kant thinks, and to this extent, Kant is right.
Kant maintains that the mind comes equipped with general faculties which impose ‘categories’ on our experience. These categories include time, space, causation, and number. Kant holds that, since these concepts cannot be logically derived from the data of experience, they must be already innate in the mind. He considers them indispensable preconditions to having meaningful experiences of the world, and not themselves logically derivable from experience of the world. This leads him to make a distinction between noumena (things in themselves) and phenomena (things as they appear to us in experience). We can’t experience things without ordering them according to the categories, and so we can never get at pure ‘things in themselves’ (things as they are before they are ordered by the categories).
Kant maintains that all observations are the combined product of external reality and mind-imposed ‘categories’, and that we cannot get anywhere by questioning these categories. For instance, we can’t conceive of objects existing without their existing in space. We cannot but conceive of objects as positioned in space, and we cannot question the fundamental nature of spatial relationships, because our minds are so constructed that we can only make sense of the world by thinking of it in terms of spatial relationships. Later neo-Kantians got rid of ‘things in themselves’, and thus became more unambiguous idealists, but this was a departure from Kant.
Everyone now accepts that no one can defend original Kantianism, not only because of the anti-idealist arguments of Russell and Moore in the 1890s and 1900s, but also because of Einstein’s revolution in physics. One of Kant’s assumptions (a very widely held assumption in the eighteenth and nineteenth centuries) is that Newton’s theory of celestial mechanics simply has to be true. Newton’s theory includes the Euclidean conception of space. We now accept that Newton’s theory is false and that space is not Euclidean.
Euclid is correct in the sense that the theorems follow from the axioms, but the axioms do not correctly map reality. Euclid’s world is fictional and therefore Newton’s world is fictional. The shortest distance between two points is not a straight line and the square of the hypotenuse is not equal to the sum of the squares of the other two sides, because spacetime is curved, something which Kant assumes that we can never seriously entertain as a possibility. Because of this, present-day Kantian metaphysicians like Michael Friedman have to accept that the ‘bounds of sense’ are much less constraining than Kant allowed, though they persist in maintaining that there are some bounds there.
If empty-mind empiricism can’t give us realism, whereas we can hope to get to realism by accepting that the human mind has built-in preconceptions, one possible conclusion, at first glance a very natural one, would be that some kind of guarantee of the truth of some propositions, perhaps very general propositions, is innate in the mind. That’s the way Kant reasoned.
But this step of Kant’s is mistaken, according to critical rationalism. Critical rationalism combines the ‘Kantian’ view that the mind has to contribute something logically prior to all experience with the ‘empiricist’ view that there are no guarantees of the truth of any claims about the way the world is, even the most general claims. (For a clear exposition of the critical rationalist position, as it emerges from the criticism and refutation of Hume and Kant, see Realism and the Aim of Science, pp. 31–88.)
(There is much more to Kant than I am mentioning here. For example, I am saying nothing about his argument from the “antinomies” or his use of the transcendental type of argument, both of which give him additional reasons to reject pure empiricism and supplement empiricism with self-evident truths known independently of experience.)
Critical rationalists say that Kant is right to conclude that the mind cannot make sense of the world unless it has built-in preconceptions which it tries to impose upon the world, but Kant is wrong in thinking that these preconceptions have to take the form of unquestionable, immovable truths. From a critical rationalist point of view, humans have an innate drive to jump to conclusions about the world, a thirst to believe in theories, along with the capacity to abandon any individual theory and replace it with a new and different theory. These theories can’t be inferred (by any valid logical process) from experience but can be suggested by experience, and once formed, can sometimes conflict with experience, and therefore can be tested against experience, possibly leading to their abandonment and replacement with new theories.
The mind is not a passive recipient of observations, but is an active explorer. Any active explorer has to begin with expectations or preconceptions. Observation must always be preceded by theory (including the unconscious theories which we only become aware that we held when they are surprisingly contradicted by experience).
The Ascent from Naivity to Physics
A classic argument against naive realism is the illusion of the bent stick. A straight stick half immersed in water looks (to someone with no prior experience of half-immersed sticks and not having been told about optics) as if it were bent. Usually, most of us learn at some point in childhood that the stick is not really bent, even though it looks bent.
Nelson’s Student has a fine old time ridiculing this point by appealing to the fact that most of us have experience of half-immersed sticks or have been told about optics (pp. 56–64). But this entirely misses the point. (Indeed, it implicitly denies the point, which is absurd. A straight stick half-immersed in water does indeed look bent.) It is question-begging to appeal to the common-sense conclusions we have accepted, as data, when what we are evaluating is precisely the claim that those conclusions were arrived at invalidly—that there is no logically sound way to get from experience to those conclusions.
Learning that the bent-looking stick is straight is one of many corrections we make to the infant’s ‘naive realism’ as we go through life. Another well-known case is the understanding that objects in the distance look smaller because they are further away. This understanding is not automatic but has to be discovered.
Traditionally, the BaMbuti Pygmies lived in the forest and never left it, their visibility always restricted to about a hundred yards. Colin Turnbull went to live among the BaMbuti to study them, and he took one of them outside the forest to the plains:
“As we turned to get back in the car, Kenge looked over the plains and down to where a herd of about a hundred buffalo were grazing some miles away. He asked me what kind of insects they were, and I told him they were buffalo, twice as big as the forest buffalo known to him. He laughed loudly and told me not to tell such stupid stories, and asked me again what kind of insects they were. He then talked to himself, for want of more intelligent company, and tried to liken the buffalo to the various beetles and ants with which he was familiar.
“He was still doing this when we got into the car and drove down to where the animals were grazing. He watched them getting larger and larger, and though he was as courageous as any Pygmy, he moved over and sat close to me and muttered that it was witchcraft. . . . Finally when he realized that they were real buffalo he was no longer afraid, but what puzzled him still was why they had been so small, and whether they really had been small and had suddenly grown larger, or whether it had been some kind of trickery.”
I grew up in England, where ambulances have a clearly marked two-note siren. As a child, I was vaguely aware that when an ambulance went by me, the pitch of its siren would drop. (According to my recollection, the fall in pitch was well over a whole-tone, in fact nearly a third, but this seems incredible now, and perhaps my memory has exaggerated it. Of course, the exact drop would depend upon how fast the ambulance was going.) I am sorry to say that I never figured out by myself that the drop in pitch was an illusion, caused by the fact that things emitting a sound and moving away from the hearer are perceived as having a lower pitch than things emitting the same sound and moving towards the hearer. It was only when I read about the Doppler Effect that it suddenly dawned on me that this was the explanation of something I had heard many times without paying attention (and, by the way, that I was an idiot). I surmise that there might even be some adults who never learn this, and continue to think that the pitch of a sound has dropped when really it has remained the same.
When I was four or five, I spent the summer with my grandparents in Rothesay, Isle of Bute. Walking with my grandfather I pointed to the sky and said “What are those white things?” He looked at me intently and said “Clouds,” or since he was a Scot, “Cloods.” I surmise he had a puzzle. Was I seriously retarded, or could it be that children from English cities had never seen clouds?
I had a puzzle, which, being an introvert and half-Scots, I would never mention. Did he mistake my question as referring to those big fluffy white things which were, obviously, duh, clouds? Or was he correctly answering my question, and was it the case that those objects I was asking about really were a special, rare type of cloud? Soon after this, I came to understand that these small semi-translucent circular white objects were not in the sky at all, but in my visual apparatus, that they were ‘floaters’ (and, by the way, that I was an idiot). My sharp awareness of the floaters was temporarily enhanced by the unusual experience of a ‘big sky’, undistracted by buildings, tall trees, or other objects.
We repeatedly make corrections to our picture or theory of the world, as we learn by trial and error to interpret the evidence of our senses more accurately. The naive realism of the toddler gives way to the less naive realism of the adult, and then to the even less naive realism of the scientifically informed adult.
You might wonder what happens in the first few weeks, months, and years following birth. We have recently come to know a lot about this, thanks to Alison Gopnik and her colleagues. The answer is that even more elementary ways of interpreting the world have to be learned by conjecture and refutation, or trial and error, the only way they could be arrived at. The fact that objects can continue to exist when they disappear from view has to be discovered by trial and error. Some of these things we learn so fast that it seems likely we have an inborn proclivity to learn them. For example, contrary to what used to be supposed (on the basis of armchair speculation), the baby understands he is an individual separate from the rest of the world by no later than the first few weeks after birth.
We learn from science that the Sun does not move across the sky, instead, the appearance of the Sun’s movement arises because the Earth is spinning. The stars do not twinkle; the appearance of twinkling is due to the interference of our planet’s atmosphere. We sometimes see a rainbow in the sky but there is no object there, corresponding to this rainbow; it’s a trick of the light. There is no pot of gold at the foot of the rainbow, not just because there are no leprechauns, but because there is no foot of the rainbow. There is no water in the desert in the experience of witnessing a mirage; another trick of the light. A pain in my toe is not in my toe at all; my brain makes a map of my body and the pain is in the ‘toe’ part of that map—hence, I can experience a completely genuine pain in my toe even if my legs have been amputated.
As we learn more and more, our realism becomes less and less naive. Educated people take for granted that the vast majority of the volume of a solid object such as a granite rock is empty space, and that there is more heat in an iceberg than in a red-hot poker, though these assertions, now common sense among the scientifically literate, would have sounded like mystical riddles as recently as two hundred years ago.
Bertrand Russell famously made the remark, quoted by Nelson, that “Naive realism leads to physics, and physics, if true, shows that naive realism is false.” (The Slightest Philosophy, p. 68).
Russell was fond of paradoxical-sounding bons mots, but the truth stated here, slightly misleadingly for paradoxical effect, is not a genuine paradox. When science corrects naive realism, the scientist replaces naive realism with a less naive realism, and this is in principle no different from the corrections a normal adult non-scientist has been making all her life. When a correction is made, all the previous observational history of the person making the correction is re-interpreted. Some parts or aspects of the earlier naive realism are preserved in the not-so naive realism that supplants it. The person, so to speak, checks to make sure that nothing she did when getting to the new theories relies upon the now discarded parts of the old theories.
Science refutes many notions held by naive realists, but it does so in such a way that we can (and do) adapt our common-sense notions to certain corrections by science. In doing so, science explains the partial or approximate truths contained in our early version of naive realism. It is a key part of critical rationalism that false theories, including theories once held to be true but now acknowledged to be refuted, can continue to be very useful.
The air of paradox in the Russell quote arises from our tendency to read it as saying that science is somehow indebted to uninformed naive realism. Although this is true in terms of chronological progression, it is not true in terms of the logic of statements. By the time someone is undertaking scientific enquiry, they have corrected and replaced their earlier naive realism. Thus the statement “Naive realism leads to physics” is parallel to “The belief that the Earth is stationery leads to the belief that the Earth is moving,” or “Newton’s theory of gravity leads to Einstein’s.”
Common sense is a set of theories of how things are, and as we become more scientifically educated, we understand that common sense uninformed by science includes false theories which require correction, though often false theories that have some workable approximation to the truth within a limited range. But a new common sense emerges among scientifically educated people. The new common sense is less naive than the common sense of the scientifically illiterate. The old picture is abandoned and the new picture is closer to the truth.
The new-born baby’s naive realism is corrected and revised, eventually leading to the very different naive realism of the toddler, which is further corrected and revised leading to the very different naive realism of the adult. If the adult paid attention in high-school science (and if the teacher still teaches any science instead of leftist ideology), his realism will be even more different and his conception of the world even more accurate. But the picture always remains incomplete; it may be indefinitely modified and improved.
If we ever get an opportunity, it would be fruitful to analyze the ‘common sense’ of a feral human, such as one raised by wolves. (Thanks to the techniques developed by Gopnik and others, we can now analyze what babies, and presumably feral humans, think about the world, even though we can’t literally ask them.) Past incidents of this phenomenon seem to show that the feral human is, once past a certain age, permanently unable to learn some aspects of even the most naive form of normal adult ‘common sense’. I conjecture that such a research project would find that exposure to language is essential for arriving at the adult’s naive common-sense view of the world. If so, this would indicate that mere observations of physical objects would never be enough for the child to acquire the rudiments of common-sense understanding. Even the most naive common sense of which we are normally aware is a highly elaborate theoretical system that might require an input from culture, especially language. If this were true, it would predict that a child raised from birth by a single adult in an isolated cottage, where the adult didn’t talk much, would be mentally retarded.
The Semantic Argument for Naive Realism
Nelson has an argument, which she apparently thinks is an argument for naive realism, which I can summarize as follows:
Skeptics have claimed that when we think we see external physical objects, what we really see are impressions, or appearances, or sensations, or sense-data from which we infer the existence of external physical objects. But this is wrong because we really do see external physical objects.
Her Chapter 3 is particularly concerned with this issue, though it makes numerous appearances elsewhere in her book.
The thing you have to be clear about when approaching any argument like this is that it is semantic. It’s not talking about the way things are in the world, independently of our discussion, but only about the rules governing the way we talk about things in the world. Here the assertion is that we’re not allowed to use the verbs ‘see’ or ‘perceive’ in a particular way, even though that way is part of actual English usage. If you fail to grasp this point, you will be bewitched by mere words and unable to talk any sense about actual perception.
The simple fact is that in English, we can use ‘see’ or ‘perceive’ either to refer to the experience of seeing or perceiving, or to refer to the experience of seeing or perceiving when that experience is appropriately linked to the existence of an independently-existing object seen or perceived.
Nelson supposes that the word ‘see’ must be used either for seeing a physical object (when it’s really there, and the appropriate causal connections between the object and our visual apparatus are in place), or for having the experience of seeing a physical object (when this might be a hallucination, and the object might not be really there, or the appropriate causal connections might not be in place).
In correct English, the word ‘see’ can be used for both, and it has several other meanings too (such as ‘understand’, ‘ensure’, or ‘match a bet’). Many words in natural languages have several meanings, sometimes related, though distinct, other times not obviously related. Take the word ‘table’ for instance. And I recently wrote an article in which I discussed two quite different though related meanings of the word ‘fact’.
Nelson’s “Student” and “Professor” go on for page after page arguing to no effect because they don’t acknowledge the simple truth that the word ‘see’ has both meanings in English. This is allowing yourself to be bewitched by words.
Nelson says that “if the relationship between your brain and your retina is called ‘seeing’, then we’re going to need a new word to refer to the relationship between your eyeball and a boat on the horizon” (p. 53). No, that is simply a false statement about English usage. We have to accept that, in the English language, the word ‘see’ has more than one meaning, just as many other words do. The word ‘see’ can be a phenomenological report, describing a subjective experience, or it can be a claim to have had that subjective experience in an appropriate causal relation to a real external object.
Both uses or senses of ‘seeing’ occur in English. It’s somewhat ironic that Nelson gives a chapter the title “Seeing Things.” ‘Seeing things’ is a standard and very common term in English for hallucinating—seeing things that aren’t there and don’t exist at all.
It’s a fact about the English language that ‘see’ has more than one meaning, and can be used to apply to subjective visual experiences with no external correlate. If you were instructing a foreigner learning English that the word ‘see’ could not be used in this way, you would be telling him an untruth and impairing his grasp of English.
You might wish that ‘see’ did not have this meaning in English (and the corresponding term in, at least, all closely related languages I’m aware of, such as French, German, and Spanish). You might propose that we adopt a different convention for epistemological discussions, and get rid of this meaning of ‘see’. This is what some philosophers, including some quoted by Nelson, have proposed. You might even make this proposal as a linguistic change in the English language, the way some folks advocate that we should load the language up with 272 pronouns to stand for 272 ‘genders’. But as things are, the use of ‘see’ to denote the experience of awareness of a mental image is normal, correct English.
The fact that there is a logical gap between our experience of seeming to perceive an external object and the actual existence of the external physical object is not a fact that can be made to go away by any mere analysis of terminology.
Since I was reviewing Nelson’s book, I skimmed through John Searle’s recent book in which he advocates naive realism, though he prefers to call it “direct realism.” Searle is an outstanding philosopher and I was much impressed by his work on intentionality and on consciousness (though I was already prepared for the possibility that he might screw up badly, because of his really terrible attempt to derive ‘ought’ from ‘is’.)
I found that he has an argument essentially the same as Nelson’s. Searle says that we “directly” perceive external physical objects, and that when we mistakenly think we’re perceiving an object (mistaken because there’s really no object there) we’re not actually perceiving anything.
This argument, in its Nelsonian or Searlian form, is a linguistic or semantic argument. It is pure talk about talk, and even as such it is false. It claims that what we mean by ‘seeing’ is seeing an external physical object, and that therefore it must be wrong to say that we ‘really see’ some intermediate mental entity between ourselves and the external physical object. This is simply a false allegation about correct English usage. Other writers, some of them quoted by Searle, who say that we only ‘see’ in the sense of having a visual experience, and that we don’t really see external objects, are of course equally wrong.
Thus Nelson’s and Searle’s arguments (really the same argument) fail. We can be said to see something intermediate between the object and ourselves. But also, it’s not essential to talk that way. In other words, it’s a fact about English that ‘seeing’ (in one sense) can be used as synonymous with ‘seeming to see’ (where ‘see’ has another sense). We can put the same point differently by saying, for example, by saying that we can have the same visual experience whether or not there is really some external object (and precisely the appropriate external object) causing that visual experience. Whether or not we choose to say that the experience without the object is a case of ‘seeing’ is simply a question of freedom of choice within linguistic convention and gives us no information whatsoever about perception or epistemology.
The fact that the Nelson-Searle argument is purely semantic can be confirmed by translating the traditional discussion of perception, including the arguments for representative realism, into Nelsonian or Searlian terminology. Instead of saying that we ‘see’ an immediate ‘object’ of sensory experience, we can say that we ‘seem to see’ or that we have ‘a visual experience of’. The arguments are unchanged by this translation, and still make the same sense, confirming that the Nelson-Searle argument is exclusively about the meanings of words.
Other Arguments for Direct Realism
Searle, of course, gives other arguments for what he calls direct realism, and I won’t try to cover most of them here. I will mention one argument, because it’s very simple, and he finds it completely convincing while I find it totally unpersuasive. The same argument is given by Armstrong—both of them acknowledge getting it from Berkeley (who used it to argue against representative realism and therefore in favor of non-realism).
Searle’s argument is that representative realism claims that the sense data (or whatever we want to call the subjective experience of perceiving) resemble the object perceived. He says this can’t be right because the object perceived is invisible and undetectable except via the sense data, therefore the sense data and the object cannot be said to resemble each other. He claims this is like saying I have two cars in the garage which resemble each other even though one is completely invisible (Seeing Things as They Are, pp. 225–26).
But how can Searle avoid saying that the experience we have of seeing something has some resemblance to the object seen? If it doesn’t, it can’t allow us to conclude that we are ‘seeing things as they are’, can it?
Anyway, this argument proves too much, because it rules out all cases of becoming aware of B through awareness of A, where there is no other way to become aware of B. Searle’s argument would imply that it must always be wrong to say that a model we have mentally constructed of some phenomenon we have detected resembles the phenomenon. And this is trivially false.
For example, consider the technique of observing the structure of deep layers of the Earth by means of tomography. (Think of the guy looking at the computer screen in Dante’s Peak.) There is no other way to observe this structure except through tomography, but we don’t therefore conclude it’s meaningless to say that the graphic which appears on the screen resembles the structure of the deeper levels of the Earth.
You might object to this that it is not a pure case of perception, since we are using normal vision to look at the graph on the computer screen. This is actually irrelevant, but if it bothers you, imagine that, some time in the future, when for some reason it becomes vitally important to be aware of changes in the structure of deep levels of the Earth, people’s brains are wired to the tomography equipment, so that they don’t look at a screen but just see the graphic as a visual experience. People would be perceiving the structure of deep layers of the Earth, by dint of the fact that they were having a subjective experience of seeing something which resembled the deep layers of the Earth. Searle would have to say they are directly perceiving the deep layers of the Earth and Nelson would have to say they are naively seeing them.
After all, what does ‘resemblance’ mean, in the context of perception? It means that the mental entity gives us information about the external entity perceived. For example, a histogram showing the weight by age of the US population resembles the distribution of weight by age in the US population, and we can say this because the former gives us information about the latter. Again, if it bothers you, imagine that people’s brains are wired to the output of the research organization which collects and processes this statistical data, and they then have the experience of seeing a histogram. We would then be perceiving the distribution of weight by age in the US population, and we could say that our experience of perceiving the histogram (which would exist only inside our skulls) resembles the actual distribution of weight by age in the US population, something that we would have no other means of observing.
Someone might feebly object that providing information is not the same as resemblance, but in that case we can say that representative realists don’t need to employ the word ‘resemblance’. The representative realist can simply say that the experience of seeing, when it is veridical, gives us information about the object seen.
We come up with the hypothesis that physical objects exist, in order to make sense of our subjective experiences of perception. Generally, we come up with this hypothesis in our first couple of weeks outside the womb. How can that seriously be disputed?
If I have two cars in the garage, one of which is invisible, and we’re thinking about the hypothesis that properties of the visible car are caused by properties of the invisible car, then it makes perfect sense to say that the visible car resembles the invisible car. This assertion would be the statement of a hypothesis, a guess, a surmise, a conjecture, of course. But all our statements or beliefs about the world are hypotheses, guesses, surmises, or conjectures. They can never be anything else, or so I surmise.
The title of Searle’s book is Seeing Things as They Are. But this might be taken as hyperbole. Aside from optical illusions, which are everywhere, our perceptions are highly selective: we don’t see an object’s mass, chemical composition, electrical charge, radio-activity, ultra-violet ‘color’, or temperature. There could be an organism which saw all these, but did not see size, shape, or color in the human-visible range. Presumably such an organism would, according to Searle, also be seeing things directly (or, according to Nelson, naively) and ‘as they are’, even though its visual experience or sense-data would be very different from ours. Presumably a bat which perceives objects by echolocation also perceives things, naively or directly, as they are. And the same goes for a dragonfly or a paramecium. So, there could be many quite distinct ways of ‘seeing things as they are’ which were vastly different. I won’t say this is an incorrect use of the phrase ‘seeing things as they are’, but it’s a bit different from the most natural and immediate understanding of that phrase, according to which there would be just one form of visual appearance which could be called ‘seeing things as they are’. In that sense, there is no such thing as seeing things as they are, and never conceivably could be, because there are many different ways of seeing things, not one of them privileged over the others. In that sense, we can only see things as they are represented, though we can speculate about the accuracy of the representation, and even subject it to tests, perhaps improving it.
Is There Some Representational Entity between Perceiver and Perceived Object?
People who defend naive or direct realism sometimes frame it like this: when we perceive a physical object, there is nothing, such as a mental state or distinctive subjective experience, intermediate between us and the object.
This strikes me as so absurd that it is not even intelligible. Nonetheless, naive or direct realists do tend to use turns of phrase that evoke it. They seem uncomfortable with any admission that there is anything which might be called sensa or sense-data. The vague notion that it’s possible to deny the existence of any ‘intermediate entity’ may be what unconsciously lies behind the appeal to the purely semantic argument I refuted earlier.
Nelson agrees that “it’s hard to object to the claim that we can perceive an oncoming freight train only by means of data we have gathered by means of our senses” (p. 14). Indeed, very hard! This looks like a grudging admission that sense-data do exist, but one page later, Nelson refers to “representationalist reifications,” insinuating that the data we have gathered by means of our senses don’t really exist. (To reify is to culpably treat an abstraction as though it had concrete existence.) As we read on, later in her book, we continue to get the feeling that Nelson has a hard time letting go of the ‘absolutely unmediated’ theory.
Some naive realists apparently feel that if they grant the existence of something intermediate, such as a person’s subjective experience identical to that involved in perceiving an external object, they will have given the representative realist a foot in the door. That’s right!
Historically, some sense-data theorists got themselves into a pickle because, being mostly materialists, they felt they had to try and explain sense-data in terms of the physics of perceptual processes. If my experience of seeing a tree is not the tree (obviously correct), and not in the tree (also obviously correct), then perhaps it’s somewhere in my optical apparatus, such as in my retinas or in my brain. Hence the many different views and coined jargons in this area. But I would say that we might not yet know enough about subjective mental processes to explain them in worked-out physical specifics. (There can be little dispute that the subjective experience of perception occurs somewhere inside the perceiver’s skull.)
We can explain the essential point here and elsewhere in purely phenomenological terms (we can stick to the subjective experience without trying to translate it into physics or physiology). Whether we see a tree or have a hallucination of seeing a tree, we have a certain type of visual experience. The visual experience is common to seeing a tree and hallucinating seeing a tree. (It’s also common to a third type of possibility, for instance that we’re seeing a hologram of a tree.) That visual experience of seeing something is the kind of thing that used to be called a sense-datum. The term ‘sense-datum’ is currently still in some bad odor (which arose because of many philosophers’ involvement, in the 1950s and 1960s, with the vogue for ‘linguistic philosophy’ or ‘ordinary-language philosophy’, derived from J.L. Austin and the later Wittgenstein). I don’t care whether we rehabilitate it or drop it. We can call it a perceptual-seeming experience, or whatever. It is something that objectively exists, as any subjective experience objectively exists, and in the case where the perception is veridical, it is intermediate between the perceiver and the external object perceived.
I will briefly mention one elementary blunder often made by naive or direct realists I have talked with. They think that seeing something on television, or in a mirror, or through a telescope is indirect whereas seeing something with the naked eye is ‘direct’. This distinction is bogus. If seeing something with the naked eye is direct, then seeing something on television, or in a mirror, or through a telescope must be direct.
I see some things with the aid of spectacles. I could use contact lenses. Or I could have laser surgery on the lenses of my eyes. It should be obvious that there is no distinction in principle between these three. My body has equipment, and it can make no difference in principle if I artificially modify or augment my body’s equipment. When Armstrong or Searle call themselves ‘direct realists’, the directness does not lie in the causal process of perception, but in the alleged non-existence of the sensory experience as an object of perception.
Some deaf people can be cured of their deafness by installing a cochlear implant in their skull. This equipment works on completely different principles than the natural human apparatus of hearing. But, after a bit of practice by the patient, the result (when the operation is successful) is very similar (as described in Michael Chorost’s book Rebuilt). It is clear that we can’t reasonably say that the cochlear implant is any more or less direct than the natural system. Artificiality in itself does not make perception any less direct (and epistemology fails unless it easily encompasses cyborgs).
If any perception is direct, then all perception is direct. However, as a matter of fact, all conscious perception is indirect, and can only be indirect, in the sense that the experience of perception is not the external object perceived, and persons do conjecture the existence of the external object perceived to account for their experiences of perception.
Is This a Hallucination which I See Before Me?
In an attempt to head off the implications of the fact that people sometimes hallucinate—seeing things that aren’t there—Nelson (through her glove-puppet Student) maintains that hallucinations are always misinterpretations of something that’s really there (pp. 74–75). So there are no genuine hallucinations, only misinterpretations of things perceived. This bold claim heroically contradicts everything that psychologists know about hallucinations.
For example, there are about one million sufferers from Parkinson’s Disease in the US, and over a third of them experience hallucinations, most commonly seeing someone they know who isn’t really there, often someone who has died (Oliver Sacks thinks that these hallucinations are not due to the disease, but to the medications). These Parkinson patients see a real person, large as life, in complete detail, every hair in place, a few feet away from themselves.
Are these sick people misinterpreting a speck of dust or a ray of light as a human being? There’s no evidence for this, and if it were true, the gap between the objective stimulus and the hallucinated object would be enormous; for most purposes the situation would be the same as an apparent perception with no external stimulus at all.
In any case, arguing for skepticism about perception by appealing to hallucinations or illusions is ultimately merely illustrative and rhetorical. If, as far as we could tell, perception were always one hundred percent veridical, there would still be a logical gap between the subjective experience of perceiving an object and the independent existence of that object, though I admit it might then be tougher, as a practical matter, to get anyone interested in that fact.
Searle says that he prefers the term ‘direct realism’ to ‘naive realism’, because ‘naive realism’ has become associated with a group of philosophers known as disjunctivists (Seeing Things as They Are, p. 15). The fact that Nelson seems to deny that one can have the same subjective experience when hallucinating as when seeing a real object makes me surmise that possibly Nelson is a disjunctivist. But since I haven’t read much of the literature on disjunctivism and since Nelson’s definition of naive realism is so very obscure, I’m not sure of that.
Armstrong’s Three Arguments against Representationalism
I have said that Armstrong’s 1961 book is the best statement I have seen of a case for direct or naive realism. Armstrong starts by assuming that in the theory of perception there are three live alternatives, representative realism, direct realism, and phenomenalism. Armstrong has a chapter on the refutation of phenomenalism (much of which I agree with) and a chapter on the refutation of representative realism, in which he presents three arguments.
First, he says that according to the claim that sense-impressions are the only immediate objects of perception, we can have no reason to believe that there are physical objects. But, as I have pointed out, talk about what is or is not an immediate object of perception (or immediately perceived) is nothing more than talk about talk. If we recast the same point in different language, we remain with a logical gap between experience and external objects, and so the same objection applies to direct realism. The direct realist does not deny that a subjective experience of perception is essential to perception, nor that the subjective experience of perception is not the external object perceived, nor that the properties of the subjective experience of perception are not the properties of the external object perceived.
Armstrong states that if the representative theory is correct, “we have no evidence at all for passing from the immediate perception of sense-impressions to the mediate perception of physical objects” (p. 29). This is pure Hume, and as far as it goes, properly interpreted, it is perfectly correct.
Armstrong then considers the point that although we have no inductive evidence of the existence of the physical world, “we might form the hypothesis of the existence of the physical world; and, having formed it, we might find that it served to explain the regularities and irregularities in the flow of our sense-impressions” (p. 30). This is excellent, but Armstrong avoids the natural conclusion with the following interesting passage:
“The objection seems just, and blunts the edge of the argument. But it does not turn it altogether. For surely we are not prepared to degrade bodies into hypotheses? We want to say that our assurance of the existence of the physical world is far stronger than any assurance we could obtain by indirectly confirming a theory. If the Representative theory were true, it would be proper to have a lurking doubt about the existence of the physical world. Yet such a doubt does not seem to be proper” (p. 30).
This passage begins with a careless slip, since it is not bodies themselves which are being “degraded” into hypotheses, but our assertion or belief or theory that there are bodies. We may want to say that our assurance of this or that is stronger than any assurance we could obtain by indirectly confirming a theory, but if so, that want must be forever unrequited. The assurance we get from indirectly confirming a theory is the strongest possible assurance for any statement or belief. That’s as strong as it gets. Some might say that logical truths are stronger, and I don’t dismiss that out of hand, however no one claims that the existence of physical objects is a logical truth. The whole passage tends to confound our subjective feelings of conviction with what we can logically demonstrate.
Armstrong’s second argument is the one about resemblance, which I have already refuted in reference to Searle. And again, the essentially semantic nature of the question as posed by Armstrong and Searle means that the objection simply reappears with direct realism, for, setting aside the language of direct and indirect perception, the fact remains that the subjective experience of perception has qualities quite different from the objective properties of physical objects. This logical gap cannot be made to go away, which is of course why we ought to acknowledge that the subjective experience represents the physical object.
Armstrong’s third argument I find difficult to understand. He claims that it makes no sense to think of a physical object that can in no way be “immediately perceived” (p. 33). Again, this is leaning heavily on the semantics. He says that we can’t say that all perception of external objects is mediate, because this implies that they could be immediate, and if they could be immediate they must sometimes be immediate (that doesn’t sound like a fair summary, as it is so obviously wrong, but it is the best I can make of what he says on p. 33).
This kind of reasoning is fallacious. Just as we can entertain the possibility that there are no disembodied minds, no living things that don’t respire, or no particles lacking relativistic mass, so we can entertain the possibility that there are no examples of conscious perception without a subjective representation which is distinct from the external object perceived (but which conveys information about the object perceived and in that sense resembles it).
Political Implications of Epistemology
Views like Nelson’s have often been prevalent among libertarians, usually stemming from Ayn Rand, though as far as I noticed (Nelson has no index) Rand is not mentioned by Nelson. And I don’t know whether Rand, though she was certainly extremely naive in some ways, would have admitted to being a naive realist.
Libertarians who embrace certain metaphysical or epistemological views often believe that these views are somehow congenial to libertarianism, whereas any views they find unconvincing are the first steps to the Gulag. I criticized some of these theories about the link between epistemology and politics in my 2002 article ‘Ayn Rand and the Curse of Kant’ and here I will just add a few observations.
As Nelson seems to uneasily acknowledge, historical claims about the causal relation between philosophy and politics have nothing to do with the merits of the philosophical theory in question. For instance (an example she mentions, p. 140), supposing it to be true that Darwinism caused the Holocaust, this would have exactly no bearing on the truth or value of Darwinism as a scientific theory.
So, even if it were true that skepticism about perception had to lead to the Gulag and the Final Solution, this would not affect the question of whether skepticism about perception is sound or reasonable. But it would be saddening, and to some discouraging, if it were the case that the dissemination of a sound (or even minimally defensible) philosophical analysis led by necessity to a horrendous political outcome. I don’t think we have to worry about that, because the kind of historical causation from philosophy to politics advanced by Rand, Peikoff, Kelley, and Nelson clashes with the historical evidence at so many points.
Nelson gives us no extended discussion of politics in this book, but she sprinkles in brief remarks about politics here and there. Here are some examples.
She repeatedly associates David Hume with Nazism (pp. 222, 231, 239). The only tangible link mentioned is that Johann Georg Hamann and Johann Gottlieb Fichte each admired Hume and each influenced German nationalism. So this, as Huck Finn might say, is a stretch.
She attributes Jonathan Edwards’s role in the Great Awakening, and the fact that some of Edwards’s followers burned books, to the fact that Edwards held to a quasi-Berkeleyan idealism (pp. 228–230). But many evangelical preachers with more impact than Edwards, such as the Wesleys and Whitefield, don’t seem to have been influenced by skepticism about perception. Evangelical ‘revivals’ may be explicable by common factors such as residual Christian beliefs among the population at large, plus people’s innate desire for an all-embracing theory that will help to make sense of their lives, plus the new theological idea (preached by the Moravians and from thence transmitted to John Wesley) that a kind of emotional born-again experience could provide the believer with assurance of salvation.
Book burning has been an occasional feature of Christianity for two thousand years. Edwards was a Calvinist, and the Calvinists sometimes burned books, two centuries before Edwards (or Berkeley). The Calvinists in Geneva not only burned Servetus’s writings, but, just to be on the safe side, burned Servetus too. It’s excessively creative to scan the history of Christian book-burning, find this one preacher who was a philosophical idealist, and attribute the book-burning by some of his followers (not by him) to his idealism (of which those followers were probably unaware). Nelson says Edwards was “inspired” by his idealism, which goes beyond the evidence. Robert Lowell wrote a poem perhaps implying that Edwards was inspired by his observations of spiders, equally a stretch, though more forgivable in a poem.
Out of all the dozens of evangelical preachers, who are realists, just one of them is (in his philosophical writings unknown to the wider public) an idealist. Therefore idealism causes evangelical revivals. The logic is certainly . . . impressive.
Since I have already warned that I might ramble, I will also draw attention to the fact, often briefly mentioned in the literature on Berkeley, that there is a tension between idealism and Christian doctrine. Berkeley, a bishop in the Church of Ireland, had to watch his step. The Bible is realist. The teaching that God became flesh in Christ does not easily harmonize with the notion that flesh is nothing more than an idea in the minds of various persons. Genesis 2:7 tells us that God made man out of the dust of the ground. So, the dust of the ground predated man. And so on.
The fact that proponents of traditional religion would sometimes point to the skeptical literature to support their resort to ‘faith’ doesn’t have the ominous ramifications that Nelson imputes to it (pp. 40–41). There are many such historical facts; people use whatever arguments they find to hand. For instance, defenders of traditional religion will sometimes say that science keeps changing its mind (which is true), while some sacred scripture stays the same (also true). This shouldn’t make us look askance at the fact that science is always open to revision.
Sixteenth- and seventeenth-century Christian teachers pointed to skeptical arguments as showing how we couldn’t rely on our own abilities to get at the truth, and should therefore accept what Christianity taught. Nelson apparently concludes that philosophical skepticism strengthened belief in Christianity. But does Nelson really suppose that if these skeptical arguments hadn’t been published, those Christian teachers would have volunteered that Christian doctrine was open to doubt? In any case, this was just one response. The more orthodox line was that basic elements of the religion, such as the existence of God, could be demonstrated by natural reason.
One of Nelson’s cute throwaway lines is: “A Cambodian guerrilla deep in a steaming jungle carries a paperback copy of Rousseau, and the next thing you know, a million people are dead” (p. 17). Did I somehow miss the memo that the Khmer Rouge renounced Marxism-Leninism and went back to Rousseau? This would imply, for example, that the Khmer Rouge must have repudiated collectivism in favor of private ownership.
Although historians disagree about the extent of Rousseau’s influence on the American Founding Fathers, no one would dispute that there was some appreciable influence. A thousand times more likely than a Khmer Rouge soldier carrying a copy of Rousseau is one of George Washington’s soldiers carrying a copy of Rousseau or Hume, and this would account for the fact that the United States immediately became a totalitarian dictatorship. Oh, wait . . .
How might metaphysical anti-realism lead to hideously repressive forms of government? Two stories are detectable in Nelson. The first is that being skeptical about perception, or doubting objective reality, directly has the effect of making you more prone to totalitarian views. The second is that skepticism about perception historically caused romanticism (which includes disbelief in the efficacy of reason, or valuing emotion above reason), and romanticism historically caused totalitarianism. That’s where Rousseau comes in, since he has been seen as the father of romanticism. But Rousseau influenced Kant, not Kant Rousseau, which looks to be the wrong way around.
A problem with historical cause-and-effect stories like this is that they depend on numerous thinking individuals reacting deterministically in a specific manner to a specific situation. So, people who are skeptics about perception must be bound to respond by valuing emotion above reason (and anyone who values emotion above reason must not be doing it for any other reason than skepticism about perception). But if they’re bound to, why does it take generations for them to do it? Then, someone who values emotion above reason (and we must assume that there’s nothing else to romanticism) must respond by becoming politically totalitarian in outlook (and there must be no other reason why anyone would become totalitarian in outlook). If any of these postulates doesn’t hold, the theory is in trouble.
Other questions follow thick and fast. For instance, German metaphysics when the middle-aged Kant started his revolution, was dominated by the rationalism of Christian Wolff. Would something less or more anti-realist have come along if Kant had died at the age of fifty-six, or would Wolffian rationalism have continued? I suggest we just can’t say. Were Rousseau’s proto-romantic writings somehow connected with skepticism about perception? I doubt it. And is it self-evident that Rousseau’s influence, in its net effect, favored totalitarianism? Nelson alludes to something called “romantic totalitarianism” (p. 231). Really? Is that a thing? If the works of Byron, Berlioz, or Poe somehow advanced the cause of totalitarianism, must we accept the corollary that Balzac, Stravinsky, or Joyce fought back on liberty’s behalf?
What tends to happen with believers in such a wondrously far-fetched story is that they cite a few cases which comply with the story, or cases which their own limited knowledge enables them to falsely suppose comply with the story. However, a few compliant cases do not really corroborate such an ambitious theory of historical causation. All cases, or at least a big majority, must conform, or we ought to discard the story.
For example, what made a lot of people support totalitarianism in the twentieth century? What we find, if we look at the evidence (and I have looked), is that totalitarianism emerged out of economic collectivism. And people became economic collectivists for specific, identifiable reasons: popular theories about economics pointed to collectivism. These economic theories owed nothing to skepticism about perception or to romanticism.
So, to take one strand out of many (but the most prominent strand), if we examine the historical record of those who became socialists, and more particularly Marxists, we observe a number of things. They were epistemological realists, and they denounced skepticism about perception as the worst philosophical crime. They were not romantics by ideological background, but if anything, anti-romantics, adherents of what has sometimes been called scientism or ‘scientific ideology’. They had quite specific reasons for thinking that socialism was both preferable to capitalism and the inevitable successor to capitalism, and these reasons (the most important of which I enumerate in ‘Ayn Rand and the Curse of Kant’) had nothing to do with romanticism and everything to do with a self-consciously ‘rational’, scientific approach.
Nelson’s epistemological views are similar to those of dialectical materialism, the ruling ideology of the USSR. Lenin’s Materialism and Empirio-Criticism has essentially the same anti-skeptical metaphysics as The Slightest Philosophy, and The Slightest Philosophy (leaving out the few sentences directly about politics) would have been heartily applauded by the ideological commissars of Soviet Communism in its heyday. The Slightest Philosophy could, making allowance for references to more recent developments, have been written by a very devout Communist Party member circa 1930. (I know, you’re wondering whether this means that The Slightest Philosophy will cause a million or more deaths. I hope not, but we can’t be too careful.)
How does it come about that the first and greatest totalitarian regime of the twentieth century, and a major causal influence on all the others, had a strictly imposed official philosophy which made a huge point of insisting that skepticism about perception is philosophically the root of all evil? Why did Communist philosophers always talk about perception exactly like Quee Nelson?
Nelson alludes to Frederick Engels’s 1843 claim that German philosophy ought to culminate in communism (pp. 30–31). But if we look at this remark by Engels, it should be understood in exactly the opposite way to Nelson’s construal.
The young Marx and Engels, formerly Young Hegelians, were suddenly converted to the atheism, realism, and materialism of Ludwig Feuerbach and David Friedrich Strauss. This conversion signified a conscious and systematic rejection of idealism. Engels considered that Feuerbach’s materialism brought an end to German philosophy and directly pointed to communism (though Engels thought it was a failure on Feuerbach’s part that he did not perceive that materialism implies communism, just as presumably Nelson might think it a failure on Quine’s part that he didn’t see that realism implies laissez-faire capitalism). It’s no exaggeration to say that in Engels’s thinking as in Marx’s, it is the total repudiation of all idealism and the unconditional acceptance of realism which points to communism.
Engels’s slightly coy statement in his brief article of 1843, quoted by Nelson, was succeeded by The German Ideology (1846) and Ludwig Feuerbach and the End of Classical German Philosophy (1886), where Engels’s and Marx’s hostility to idealism and their view of a tight connection between materialism and communism are spelled out in great detail. Just to be clear, I don’t accept that any metaphysical or epistemological theory implies or causes any social or political theory, much less any political movement. Yet it would be easy to formulate a thesis that the rejection of skepticism about perception and the embrace of metaphysical realism lead to totalitarianism; after all, we observe that totalitarians are usually motivated by certainty, not doubt. I don’t think there’s anything in such a thesis, but at least it would not be as spectacularly contrary to the historical evidence as Quee’s thesis is.
In Nineteen Eighty-Four, Orwell has the Party boss O’Brien propound the theory of “collective solipsism,” according to which reality is whatever the Party says it is. Nelson says that Orwell is “caricaturing collectivist epistemology” (p. 31). But was there, as a matter of fact, any collectivist epistemology to caricature? I discuss this point in my book on Orwell (Orwell Your Orwell, pp. 307–311).
Although O’Brien’s argument is a remarkable anticipation of some aspects of social constructivism (these had been prefigured by Otto Neurath and by Mannheim’s ‘sociology of knowledge’ but it’s doubtful that Orwell knew of these), we have to be clear that no actual totalitarian regime has ever employed the “collective solipsism” type of reasoning, and it’s very unlikely that any actual totalitarian regime ever could, because totalitarian regimes, when they tell untruths (or even truths which they find it useful to draw attention to), always intend to communicate that what they say is true by the traditional absolute and objective standard of truth. Orwell is not caricaturing any existing “collectivist epistemology” but inventing a fictional epistemology in order to caricature totalitarian practice.
Notoriously, the Communist parties would sometimes suddenly switch their position on various issues, saying the opposite today of what they had said yesterday, and sometimes propagating falsehoods to help justify the current party line. Actually, this aspect of Stalinism was barely noticed by most non-leftists, while non-Communist socialists like Orwell were always acutely aware of it. The Wobblies even had a song about it (“Our Line’s Been Changed Again”).
Orwell applies to this phenomenon the type of satire he had learned from Jonathan Swift: he has the totalitarians preaching what he considers them to be practicing. The Communists never did preach anything like this; in fact, as fanatical adherents of materialism, a form of metaphysical realism, they always preached the opposite.
Orwell’s hero Winston Smith expresses the view, against the Ingsoc Party, that “reality is something external, objective, existing in its own right.” It doesn’t surprise anyone knowledgeable about Communist thinking to learn that the leading Communist ideological authority Rajani Palme Dutt, responding in 1955 to Nineteen Eighty-Four, commented that this remark by Winston Smith states what is in fact the Communist view (Meyers, George Orwell, pp. 287–88).
Finally look at this sentence by Nelson: “In the same century that Heidegger, Habermas, and DeMan imbibed totalitarian collectivism as National Socialists, Althusser, Gramsci, Sartre, Camus, Putnam, and Rorty imbibed it as international socialists” (p. 30). You might think that Nelson is here giving us evidence—looks like quite an accumulation of evidence!—for her historical thesis. But exactly what does this all amount to?
What three writers “imbibed” (one of them as a child, one as a young adult, the other in middle age) hardly shows that their distinctive philosophical views resulted from the ideology they imbibed, or vice versa. After the war, none of these three showed any obvious political influence of National Socialism. It’s not playing fair to smear Habermas because he was a schoolboy under the Third Reich. Heidegger seems to have cheerfully embraced the National Socialist regime partly because of career opportunism. To what extent he was a realist or an idealist is sometimes debated (this is the ambiguity of Husserl’s legacy, though Heidegger’s ‘being in the world’ has been seen by some as an attempted answer to Humean skepticism), but at any rate he was no romantic and he didn’t endorse National Socialism before it came to power or after it had been overthrown. De Man (presumably Paul, not his uncle Henri/Hendrik) wrote some antisemitic stuff in Nazi-occupied Belgium, apparently for reasons of self-preservation and self-advancement (for he discreetly helped out individual Jews he knew). After the war he promoted deconstructionism in literary theory. He doesn’t appear to have been influenced by skepticism about perception. He wrote a book on romanticism, which might be considered an unmasking of romanticism’s pretensions.
Gramsci and Althusser were both materialists, therefore hardcore realists (Gramsci’s concept of revolutionary praxis is not, despite what you sometimes hear, a departure from philosophical materialism; there is a question mark over Althusser’s last writings, but they have had no influence). Neither Gramsci nor Althusser were romanticists. Sartre and Camus were philosophical realists and in their literary output decidedly unromantic. Putnam and Rorty vacillated on the question of realism and they were not associated with romanticism. Of course, they had generally leftist worldviews, and that may be what bothers Nelson, but that’s the predominant fashion amongst all twentieth-century intellectuals, including the realist and anti-romantic ones. Some of the harshest attacks by realists on postmodernism have come from leftists (Chomsky, Detmer, Eagleton, Sokal).
So, all in all, if we deconstruct Nelson’s flurry of names, we just don’t find much corroboration for her historical thesis. Nelson’s method, as she scans a lot of historical data, is to pick out a handful of instances which seem to confirm her story, while ignoring the far greater number of instances which starkly contradict it.
Armstrong, D.M. 1963 . Perception and the Physical World. Routledge.
Boghossian, Paul. 2006. Fear of Knowledge: Against Relativism and Constructivism. Oxford University Press.
Chorost, Michael. 2005. Rebuilt: How Becoming Part Computer Made Me More Human. New York: Houghton Mifflin.
Devitt, Michael. 1991 . Realism and Truth. Blackwell.
Ewing, A.C. 1934. Idealism: A Critical Survey. Methuen.
———. 1996 . A Short Commentary on Kant’s Critique of Pure Reason. University of Chicago Press.
Feuerbach, Ludwig. 1986 . Principles of the Philosophy of the Future. Hackett.
———. 1989 . The Essence of Christianity. Prometheus.
Foster, John, 1982 The Case for Idealism. Routledge.
———. 2000. The Nature of Perception. Oxford University Press.
———. 2008. A World for Us: The Case for Phenomenalistic Idealism. Oxford University Press.
Friedman, Michael. 1992. Kant and the Exact Sciences. Harvard University Press.
Gopnik, Alison. 2009. The Philosophical Baby: What Children’s Minds Tell Us about Truth, Love, and the Meaning of Life. Farrar, Straus, and Giroux.
Gopnik, Alison, and Andrew N. Meltzoff. 1998. Words, Thoughts, and Theories. MIT Press.
Gopnik, Alison, Andrew N. Meltzoff, and Patricia K. Kuhl. 1999. The Scientist in the Crib: Minds, Brains, and How Children Learn. Morrow.
Hardin, C.L. 1988. Color for Philosophers: Unweaving the Rainbow. Hackett.
Hayek, F.A. 1963. The Legal and Political Philosophy of David Hume. Il Politico 28.
Hume, David. 1888 [1738–1740]. A Treatise of Human Nature: Being an Attempt to Introduce the Experimental Method of Reasoning into Moral Subjects. Oxford University Press.
Hylton, Peter. 1990. Russell, Idealism, and the Emergence of Analytic Philosophy. Oxford University Press.
Jameson, Fredric. 1991 . Postmodernism: Or the Cultural Logic of Late Capitalism. Duke University Press.
Johnson, Kent, and Wayne Wright. 2006. Colors as Properties of the Special Sciences. Erkenntnis 64.
Latour, Bruno. 1998, Ramses II, Est-il Mort de la Tuberculose? La Recherche 307 (March).
Lenin, V.I. 1927 . Materialism and Empirio-Criticism: Critical Comments on a Reactionary Philosophy. In Lenin, Collected Works, Volume 13. International.
Lowell, Robert. 1946. Lord Weary’s Castle. Harcourt, Brace.
Meyers, Jeffrey, ed. 1975 . George Orwell: The Critical Heritage. Secker and Warburg.
Miller, David. 1994. Critical Rationalism: A Restatement and Defence. Open Court.
———. 2006. Out of Error: Further Essays on Critical Rationalism. Ashgate.
Nelson, Quee. 2007. The Slightest Philosophy. Dog Ear.
Popper, Karl R. 1968 . Conjectures and Refutations: The Growth of Scientific Knowledge. Harper.
———. 1979 . Objective Knowledge: An Evolutionary Approach. Oxford University Press.
———. 1983. Realism and the Aim of Science. Rowman and Littlefield.
———. 2009 . The Two Fundamental Problems of the Theory of Knowledge. Routledge
Radnitzky, Gerard, and William Warren Bartley III, eds. 1987. Evolutionary Epistemology, Rationality, and the Sociology of Knowledge. Open Court.
Sacks, Oliver. 2012. Hallucinations. Knopf.
Searle, John R. 1964. How to Derive ‘Ought’ from ‘Is’. The Philosophical Review 73:1 (January).
———. 1983. Intentionality: An Essay in the Philosophy of Mind. Cambridge University Press.
——— 1992. The Rediscovery of the Mind. MIT Press.
———. 2015. Seeing Things as They Are: A Theory of Perception. Oxford University Press.
Siegel, Harvey. 2007. Review of Paul Boghossian, Fear of Knowledge: Against Relativism and Constructivism. Notre Dame Philosophical Reviews.
Steele, David Ramsay. 2002. Ayn Rand and the Curse of Kant. Liberty (August).
———. 2017. Orwell Your Orwell: A Worldview on the Slab. St. Augustine’s.
———. 2018. Is It a Fact that Facts Don’t Matter? In Daniel Yim, Galen Foresman, and Robert Arp, eds., Scott Adams and Philosophy: A Hole in the Fabric of Reality (Open Court).
Turnbull, Colin. 1961. Some Observations Regarding the Experiences and Behavior of the BaMbuti Pygmies. American Journal of Psychology 74.
Van Inwagen, Peter. 2009. Metaphysics. Third edition. Westview.
Watkins, J.W.N. 1984. Science and Scepticism. Hutchinson.
Wilkerson, T.E. 1976. Kant’s Critique of Pure Reason: A Commentary for Students. Oxford University Press.