There’s a familiar issue that underlies many of the most interesting discussions on TPF. Many forum members are committed to a broadly naturalistic view of existence. This view (which I don’t share, but that’s not important right now) holds that something like physicalism, or at any rate “what science studies,” is fundamentally what there is. We may of course find other discourses, other ways of describing the “life-size” phenomena that we encounter – Sellars’ “manifest image” – but the naturalistic view would insist that these other ways must be derived in some sense from a scientific picture, or at the very least be reducible to such a picture.
The problem is that many of the things that philosophers (and a lot of other people) care most about don’t seem to be an easy fit for this view. Huw Price, whose paper “Naturalism Without Representationalism” I’m about to cite in some depth, offers this list: “meaning, value, mathematical truth, causation and physical modality, and various aspects of mentality.” Price notes, “The difficulties stem from the fact that in many interesting cases it is hard to see what natural facts we could be talking about.” And so, uneasily, we wonder if perhaps these seemingly significant terms don’t refer at all, or not in the ways we care about.
Price calls this a “placement problem” – how do we “locate topics of these kinds within a naturalistic framework? . . . We seem to be faced with a choice between forcing the topic concerned into a category which for one reason or another seems ill-shaped to contain it, or regarding it as at best second-rate – not a genuine area of fact or knowledge.”
Richard J. Bernstein, in Pragmatic Naturalism, connects this problem with the work of both McDowell and Nagel. He quotes McDowell:
[O]n a familiar modern understanding of nature, a contrast opens between saying how something is placed in the space of reasons – a logical space that is organized by justificatory relations between its inhabitants – and saying how something is placed in nature. . . The structure of the space of reasons stubbornly resists being appropriated within a naturalism that conceives nature as the realm of law. (Naturalism in Question, 91)
An example may help illustrate this, and one of the most familiar is that of rationality, which I think Price would include in “aspects of mentality.” On the one hand, we want to regard rationality as a natural phenomenon like any other, since after all we humans are natural creatures. So surely science ought to be able to explain rationality, just as it explains heat, or planetary motion. But on the other hand, the claims of rationality push against this very sort of explanation. If rationality can be given, for instance, an evolutionary explanation, what does that say about the key terms within rationality – terms like truth, consistency, warrant, logic, objectivity, et al.? Can these terms really characterize what rationality is and does – its authority to do it, so to speak – if at the same time they are purportedly now explainable in naturalistic terms? In Nagel’s formulation, reason seems to demand “the last word” here, rejecting deflation into a mere form of life or species-specific practice. To oversimplify, Nagel argues that we need the standard conception of rationality in order to make the scientific judgment that anything has been adequately explained.
But how could we defend that, while remaining naturalistic? Where do we place rationality in our naturalistic picture?
Price’s paper begins from this uncomfortable tension – not about rationality, which is my own example, but about the placement problem in general – and goes on to offer what he considers to be a third way, an alternative that does justice to both naturalism and our wish not to push a topic into a naturalistic form so that it’s no longer recognizable for what it is – “distorting the very phenomena that a naturalistic study is supposed to explain.” His paper is extremely well analyzed and argued, and I recommend it highly. But here I want to talk about just one aspect of Price’s case, a distinction he makes between a material view and a linguistic view of the “problem” topics in question.
The most intuitive way to look at the naturalism question would be to see it as a problem about items in the world – “object naturalism,” as Price calls it, making the connection with Sellars’ focus on the primary objects of the manifest and scientific images. He says, “The issue is thus how some thing, X, can be a natural thing – the sort of thing revealed by science (at least in principle).” So, for example, we’re familiar with values like courage, compassion, truthfulness, etc., and we ask ourselves, “How can such items be given a naturalistic explanation?” Or we might ask the same question about mathematical entities, or about my intention to open a door. But the question is formed as a question about X.
Brief interruption about neutral terms: It’s always hard to find a neutral term to describe something that’s allegedly “out in the world” or “mind-independent”. Any one we pick becomes immediately tendentious. If we talk about “things” or “entities”, we start to picture objects exclusively. If we say “events” or “processes,” we lose a sense of stable identity. “Properties” or “qualities” come with their own implied biases. My favorite term – because it’s the most neutral one I can think of – is “item”, so that’s the one I’ll use here. “Item” should be understood as potentially including absolutely anything that the speaker wishes to assign to an object-naturalist world. It carries no further commitments.
So, back to the placement problem. The other way to view this problem is that it’s linguistic in origin. It’s not about items in the world, it’s about what we say. Here is Price:
Roughly, we note that humans (ourselves or others) employ the term “X” in language, or the concept X in thought. . . . [So] we come to wonder how what these speakers are thereby talking or thinking about could be the kind of thing studied by science. (6)
It’s crucial to see that a linguistic interpretation of the problem does not begin by assuming that there must be a referent for the piece of linguistic behavior. It does not start with a question about X, but rather about the term “X”. So a possible answer to the question, “What are we talking about?” could be “Nothing.”
Price goes on:
In favour of the material conception, it might be argued that the placement problem for X is a problem about the thing X, not a problem about the term “X”. In other words, it is the problem as to how to locate X itself in the natural world, not the problem about how to locate the term “X”. (6-7)
But Price points out that certain philosophical approaches which begin with language instead would be automatically rendered “wrong” if the placement problem were, by stipulation, about the material conception. He does not find that plausible. Non-cognitivism is the example he gives – and which I alluded to just now – and it’s a good one. If talk of “Xs” has no objective referent, then there can’t be a material problem at all; we’re simply speaking vacuously, or perhaps emotively or poetically. But those who begin with the material conception have to rule that out; they have to say that linguistic analysis gets the cart before the horse, and misunderstands the problem by focusing on the term “X”. Now non-cognitivism may be false, but Price’s point is that we can’t just ban it from the discussion on the grounds that it misunderstands the focus of the placement problem. That focus is the very thing under debate.
Suppose one were truly neutral on the topic of where the placement problem begins. Such a person would hear humans talking about many things and experiences that may or may not have references in the world, and, if they do, may or may not be treated satisfactorily by science. True neutrality, remember, demands that science not be given ontological priority, a primacy of perspective in terms of objects. A perfectly good scientific account might analyze “meaning,” say, in terms that reduce it to something that doesn’t resemble our everyday “meaning” at all. But this need not be eliminative, any more than a scientific account of heat eliminates being scorched. Heat is not “really” about feeling a hot temperature; perhaps meaning is not “really” about something that events or words contain or display.
This is Price’s “subject naturalism,” which doesn’t deny the manifest image, but merely begins with a scientific picture of human nature rather than a scientific picture of the image or the object. (Bernstein traces this line of thought back to Dewey: “Dewey advocates a subject-naturalism in which philosophy’s starting point must lie in science’s disclosure about ourselves.”)
Price makes this important observation:
Object naturalism rests on substantial theoretical assumptions about what we humans do with language – roughly, the assumption that substantial ‛word - world’ semantic relations are a part of the best scientific account of our use of the relevant terms. (10)
In other words, to make the placement problem be about items, you have to assume or accept that semantic relations are real – that this is what language does – and these relations can be based in scientific understanding. You need a bridge of some sort between our use of language and a language-independent world, such that we get “a genuine shift of theoretical focus . . . to an issue about the nature of non-linguistic objects.”
In contrast, if the placement problem is linguistic, and if we don’t appeal to representationalist semantic relations (and that’s a big “if”), then we have what Price calls “a plurality of ways of talking.” He says, “The challenge is now simply to explain in naturalistic terms how creatures like us come to talk in these various ways.” This is subject naturalism, clearly, and for Price it is good enough. We can remain naturalists about human beings without committing ourselves to any dubious natural objects that require placement in the world.
The question is, is this a satisfactory response to the original formulation of the placement problem? At the start, Price stated it this way: “We seem to be faced with a choice between forcing the topic concerned into a category which for one reason or another seems ill-shaped to contain it, or regarding it as at best second-rate – not a genuine area of fact or knowledge.”
If the placement problem is really about how humans use language, aren’t we “forcing the topic” to go there? Yes, knowledge about linguistic behavior is “a genuine area of fact or knowledge,” but is it the right area in which to address the placement problem? Price doesn’t claim that this must be so. He’s defending naturalism as such; the farthest he will go is to say:
The availability of the subject naturalist alternative makes clear that the problems of object naturalism are not problems for naturalism per se – not a challenge to the view that, in some areas, philosophy properly defers to science. (21)
And I think this is true. The naturalist can certainly try to explain “what role the different language games play in our lives,” such that talk about values is explained differently than talk about electrons. But will this solution satisfy someone – naturalist or not – who wants a more comprehensive way to discriminate about putative objects/items in the world, one in which philosophy doesn’t necessarily “properly defer to science”? Isn’t it possible that some “problem” words refer after all? Is it science that will tell us which ones? I think a naturalist has to reply, Yes, we must defer to science. And that has an uncomfortable aura of circularity.
Another way of putting this is to ask what the difference is between Price’s subject naturalism and non-cognitivism about all the problem items. Price says that non-cognitivism doesn’t have to be true, but is this disingenuous? This comes back to the question about whether the subject-naturalist account is or is not eliminative concerning the items involved. Is the analogy I drew with the scientific account of “what heat is” a good one for a subject-naturalist account of “what meaning is”? Heat as molecular motion may not eliminate my perspectival experience of being scorched, but it does eliminate, for instance, the existence of heat understood as “a natural element that constitutes one of the four bases of the world.” Science is clear that such an item does not exist. Is that closer to what a scientific account of meaning might declare? I don’t want to take more space here with discussions about reductionism and eliminativism, but I think we can see where the problems lie.
Clearly I have reservations about Price’s suggested solution to the placement problem. But I hope you’ll read his essay for its perspective on why the problem needs attention. And I don’t mind going into more of the details of his case against representationalism, which I’ve hardly touched on here, if there’s interest in doing that.