Science, non-science, and nonsense: on the varieties of knowledge[1]
Timothy Chappell
In characterising an episode or a state as that of knowing, we are not giving an empirical description of that episode or state; we are placing it in the logical space of reasons, of justifying and being able to justify what one says.[2]
As usual, “knowledge” is understood as propositional knowledge.[3]
Ce besoin de l’immatériel est le plus vivace de tous. Il faut du pain, mais avant le pain, il faut l’idéal.[4]
Evidently modern science is focused on third-personal propositional knowledge. Indeed, wherever it can, it apparently goes beyond third-personal to impersonal propositional knowledge. As we were taught to write in school chemistry lessons, “A Bunsen was lit”, not “Mr Bloggs the chemistry teacher, or my classmate Bill, lit a Bunsen”.
Analytical philosophy is rather keen on science, and tends to follow it in this focus on third-personal or impersonal propositional knowledge. So for example Duncan Pritchard’s excellent introductory text What is this thing called knowledge? is explicitly focused on propositional knowledge from p.4 on.
An ant might plausibly be said to know how to navigate its terrain, but would we want to say that the ant has propositional knowledge; that there are facts which the ant knows? ...Intuitively not, and this marks out the importance of propositional knowledge over other types of knowledge like ability knowledge, which is that such knowledge presupposes the sort of relatively sophisticated intellectual abilities possessed by humans.[5]
Later on, in his Chapter 11, Pritchard wonders how there can be moral knowledge, taking it to be a sufficiently problematic case to deserve a special case-study. I suggest these two facts about Pritchard’s book are connected. (Perhaps DP thinks so too: I’m not suggesting he hasn’t thought of this.[6]) Moral K does look specially problematic if we focus more or less exclusively on propositional knowledge. My question here is whether we can make it look less problematic by thinking about other kinds of knowledge: varieties of non-propositional moral knowledge.
My answer to that question is a cautious Yes. I think there may be—I am going to try and show that there are—kinds of knowledge which are just as much knowledgeas propositional knowledge is, but quite unlike it, and which are also equally distinctive of human life (even if we share them with creatures which, like Pritchard’s ant, are quite unlike us). If so then we should not judge these other kinds by the standards of propositional knowledge, any more than we should judge propositional knowledge by their standards. Here as elsewhere my aim is not to analyse things down to the simplest possible theory, but rather to display and explore what Louis Macneice called “the drunkenness of things being various”. My motto will be Chairman Mao’s “Let a hundred flowers bloom”, though my methods may differ from Mao’s in one or two respects.
If the argument comes out right, then there is MK, which is a fairly notable consequence. Another consequence is that metaethicists and others might usefully spend more time than they do examining the non-propositional forms of MK. I don’t suggest that no one has ever had the idea that some MK might be non-propositional. On the contrary, Pritchard’s own colleague Andy Clark is one author among plenty who has written about exactly that idea[7]; as Clark observes, and as I shall show by exposition, the idea is at least as old as Aristotle’s Ethics. I do suggest that it would be a good idea for contemporary metaethicists to spend a bit more of their abundant energy and resources on this possibility.
And a third consequence: if there is non-propositional moral knowledge, that implies that not all knowledge is scientific knowledge—which unless we are raving reductionist ideologues, we probably knew anyway—but also that we can be fairly precise both about why some K is not scientific K, and also about how very different this non-scientific K is from scientific K. And this, maybe, we didn’t know anyway.
I won’t say that no MK is propositional. I won’t even say that no important MK is propositional. I will say that much important MK is not propositional, and that no one can have the MK that anyone needs to have without having a lot of non-propositional MK. Like Clark 2000, I want to talk both about these non-propositional varieties of MK, and also about the ways that they can combine (with each other, and with propositional MK) to create the sort of rich, flexible, and multi-purpose web of capacities and sensitivities that might be worth calling practical wisdom.
Looking beyond the putative category of moralknowledge to a broader putative category that we might call humane knowledge, I want to explore the possibility that there might be other kinds of non-scientific knowledge in two cases in particular, aesthetics and religion.
First, then, we need to know (a) what varieties of knowledge there are, and (b) how they count both as a variety (as genuinely different from each other) and also (c) as knowledge (as genuinely worthy of the name). Once we have these matters as straight as we can get them, both for what I shall call the Probable Candidates and (d) for the Possible Candidates, we can ask (e) what these varieties of knowledge look like in the cases of moral, aesthetic, and (f) religious knowledge.
(a)
Various things might be distinct both from propositional K and also from each other. Call them the Candidates.[8]I will put the Candidatesin two groups. I do this because there are two cases where I am quite sure that we have distinct varieties of K, and two cases where I am not sure at all that that is what we have. The first two Candidates—call them the Probables—are ability-knowledge and experiential knowledge; the other two Candidates—call them the Possibles—are what I shall call second-personal knowledge and self-knowledge.
The two Probable Candidates, EK and KH, are easy to characterise, at least in broad terms, since both are just as familiar as propositional knowledge.
KH, ability-knowledge, is knowledge how to do things, the kind of K that I have in virtue of knowing, as in fact I do, how to juggle, read Italian, or make an apple pie.
Experiential K is K by experience, phenomenal awareness: as it is sometimes put, it is K “what it’s like”, e.g. K what silk feels like when you rub it, or what woodsmoke smells like, or the taste of apricots. To take a more controversial example, EK is the new relation to red that Mary the colour scientist enters when she finally leaves her monochrome bubble where she’s been doing the physical theory of red and other colours, and actually sees red. We can use the word “acquaintance” for EK if we like, provided we don’t fall into the trap of taking that word to commit us to the specific details of Russell’s famous account of K by acquaintance. (Similarly, we shouldn’t make the mistake of thinking that believing in KH must mean believing specifically in Ryle’s famous account of KH.)
So much for the Probables. The two Possible Candidates, 2PK and SK, are much harder to characterise, and, for that reason and others, much harder to discuss philosophically. But very briefly indeed, what I have in mind when I speak of 2PK is the apparently distinctive kind of cognitive achievement that is involved in responding to another person, say Jane, as another person. And what I mean by SK is the apparently sui generis cognitive achievement of awareness and understanding of oneself.
Maybe contemporary science uses or depends on some or all of these other candidate kinds of knowledge.[9] But even if it does, it does not seem focusedon any of them; apparently its focus is on explicit, impersonal, testable propositional knowledge. It looks like science works something as follows: There can be (there is) a scientific study of what EK involves, but the object of such a study is to produce propositional knowledge about EK. Being a competent scientist involves knowledge-how, but the skill you need to earn a living in (theoretical) science is to know how to produce bodies of propositional knowledge. Again, good scientists typically work in teams, and productive team-work with other scientists is impossible, or at any rate difficult, without reasonably high degrees both of awareness of others and of self-awareness. (Witness the widespread mayhem that has been caused down the years by some very famous scientists who were famously lacking in both.[10]) Nonetheless, it is obvious enough that no scientific knowledge is well categorised in either of the putative kinds second-personal knowledge or self-knowledge. Even when a scientist studies the phenomena I am tentatively grouping into those kinds, her aim, as before, is not to produce either second-personal knowledge or self-knowledge, but structures of propositional knowledge about those phenomena. (Or so it seems, but this appearance will be questioned before we’re finished.)
When I speak, under the heading of 2PK, about “responding to Jane as a person”, I don’t just mean “recognising that Jane exists”. You don’t get 2PK merely by learning the proposition that something in your vicinity is a someone, isanother person. What you need for 2PK, to put it in a metaphor that I find almost unavoidable[11], is to come face to face with Jane.
As Stephen Darwall has recently argued, second-personal address seems quite different from third-personal or impersonal interaction with what we might, with a nod in the direction of Martin Buber, call “the world of things”. As Sir Peter Strawson famously put it in “Freedom and Resentment”, there seems a pretty fundamental distinction between “interpersonal reactive attitudes” on the one side, and on the other, what he rather ominously called “objective attitudes”. Or to quote a more recent exposition of the idea, consider Shaun Gallagher’s words[12], “in ordinary instances of interaction with others, I am not in the observer position; I am not off to the side thinking or trying to figure out what they are doing. Rather, I am responding to them.”
I share the sense that is evidently felt by Darwall, Buber, Strawson, Gallagher, and many other philosophers—notably including Immanuel Kant and Emmanuel Levinas—that there is something distinctive about second-personal knowledge. One reason why 2PK is only a Possible Candidate is that it’s rather hard to say exactly what is distinctive about it—or indeed whether there is one phenomenon here at all, rather than a whole range of different phenomena that might all be worth calling 2PK for different reasons. Without these clarifications, it’s not easy to be sure that 2PK really is a distinctive kind of K, not reducible to PK, in the way that EK and KH (as I shall argue) pretty obviously are. The interesting possibilities remain open that 2PK might be an amalgam of other types of K, or perhaps even identical with one other type of K. (After all, one very familiar way we use the word acquaintance is to talk about knowing other persons.)
To see how 2PK might be an amalgam of other types of K, consider a further form of K that I have not yet considered: objectual K.[13] OK is K of objects, K of particular things. Apparently this form of K was the basic form for the classical Greeks: Aristotle, for example, takes substance to be the primary existent, and so naturally takes K of substance to be the primary K. The three first forms of K that I have distinguished here, PK, KH, EK, are all for the Greeks derivative from OK. It is only by knowing the object Socrates-as-snub-nosed that we come to know the proposition “Socrates is snub-nosed”; it is only by knowing the craft of building that we come to know how to build houses; it is only from experiential knowledge of Achilles (and in particular the colour of his hair) that we come to know what redness is.
It comes naturally to us, then, to say that OK is an amalgam of the three varieties of knowledge that I have distinguished in this paper. As a matter of fact I think Aristotle at least would deny even this; he would say, I suspect, that OK is what Tim Williamson calls “prime”, i.e. that it is notfactorisable into PK, EK, and KH, even though all three forms of K naturally flow from it—just as belief, according to Williamson, flows naturally from PK without being an ingredient of PK. (Perhaps Plato would agree with Aristotle about this, though it is harder to situate him relative to this question.)
Without getting into an argument with Aristotle,[14] I think we can at least see what it would be for OK to be factorisable into PK, EK, and KH. Knowing an object, we could say, means knowing what propositions are true of it; and it means being directly acquainted in experience with the object; and it means knowing how to interact with the object. If these three kinds of knowledge are all there is to OK, and if they can be neatly separated out from each other, in principle at least, then OK is just an amalgam of PK, EK, and KH. If 2PK is similarly structured, then it too is no more than an amalgam of other more basic kinds of K. But then, presumably, 2PK should not appear on our list of these basic kinds—no more than OK should by the parallel argument.
Analogous problems and more apply in the case of the other Possible Candidate, SK. When philosophers today talk about “self-knowledge” they can have a whole variety of different things in mind. One thing they mostly don’t have in mind is the famous Greek saying that Socrates liked to quote—“Know thyself”;[15] which is a pity, perhaps. In any case, with SK as with 2PK, it isn’t entirely obvious which of the various phenomena that we could call SK is the central phenomenon. There does seem to be something suigeneris about my own awareness of what I am thinking, feeling, and doing now; about my own relation to my own memories of the past and my own intentions about the future; and so on. (It’s tempting to ask: how could there not be something sui generis about all this?) But here too we face the possibility—a possibility taken very seriously by contemporary writers on self-knowledge such as Fred Dretske and Alison Gopnik—that what seems, at least experientially, to be another distinctive kind of K is really, on analysis, no more than an amalgam of other kinds of K.
There is also a rather intriguing, but again elusive, thought about the Possible Candidates that, at least since the late nineteenth century, has been swimming in and out of focus in both analytic and continental philosophy. This is the thought that 2PK and SK are constitutively interrelated, that I cannot know myself without knowing others, or others without knowing myself. Or again, you can hear it suggested that it is only within the context of a prior relatedness that either SK or 2PK can emerge, as two poles of the same axis. Or that K of the external world presupposes K of myself and of others. You may even hear people say that any other knowledge at all always depends on a prior 2PK, or a prior relatedness.
The preconditionality of second-personal knowledge seems to be asserted in Edmund Husserl’s Cartesian Meditations: “A priori my ego… can be a world-experiencing ego only by being in communion with others like himself: a member of a community of monads”[16]. Slightly differently, it is also quite often suggested, for example by Heidegger and perhaps by Buber, Wittgenstein, and Levinas as well, that there can be neither objectivity nor subjectivity unless there is intersubjectivity first: to be is to be in relationship, and it is only out of relationship that knowledge of ourselves, of other persons, and of the world of things can ever arise.
Such claims are attractive for a number of reasons. But at least in the phenomenological form in which they are usually presented, they are not compelling.
One reason why not is because such claims seem not to have the anti-sceptical bite that Heidegger in particular evidently thinks they have. That (what we call) awareness of other people is a logical precondition of knowledge of anything else seems too strong a thesis to be plausible. That (what we call) awareness of other people is a causal precondition of knowledge of anything else does not give us a sound argument to the conclusion that there are certainly other people; perhaps (what we call) awareness of other people is an Eden from which we get expelled as, on the basis of that putative awareness, we come to learn more—for example, as we come to learn the sceptical arguments that put in doubt the existence of those other people.
Moreover, these claims about the primordiality of intersubjectivity sound very much—at least in their more-plausible causal versions—like a posteriori not a priori claims. Now phenomenology—self-examination, introspective analysis of what is revealed in consciousness—is certainly a way of making some a posteriori discoveries. Is it a way, or a good way, of making the discovery that intersubjectivity is primordial in the required sense? I doubt it. The thesis that awareness of other people is a causalprecondition of knowledge of anything else sounds to me like a thesis about child development. I therefore recommend that we take the research programme of trying to establish whether or not this thesis is true as a research programme, not in phenomenology, but in experimental child psychology. I look forward to seeing results from that programme. (For all I know, there already are such results.)
I don’t myself intend to argue, or even to imply, any such theses; for instance, that 2PK is a magic bullet to kill off the problem of other minds. We can no more solve the problem of other minds by positing a mental faculty with the specific role of detecting other minds than we can solve the problem of moral scepticism by positing a mental faculty with the specific role of detecting moral properties. (Why couldn’t such faculties exist, yet have nothing to detect?) Of course the thesis that 2PK is a distinctive kind of K is relevant to the problem of other minds. It’s just relevant in subtler and more interesting ways than this.