Melser's reply to Frederick Adams



Derek Melser's
Reply to Frederick Adams

The pre-publication draft of the review that I have posted on my site was kindly, and somewhat ruefully, supplied by Professor Adams himself. It differs from the published version only in the opening sentence — the published one reads, “Life is short and there are too many good books in philosophy for me to recommend that you read this one” — and in containing slightly more literal and grammatical errors. The passages from Adams’ review that I quote below (and leave uncorrected) are from the text published in Mind.

Professor Adams does not like The Act of Thinking. He has no good things and lots of bad things to say about it. However, I don’t think Adams has read my book carefully and I think that much of what he says about it is untrue and unjust. The Mind website states that “Mind has always enjoyed a strong reputation for the high standards established by its editors…” It is intriguing that a distinguished professor of philosophy should write, and Mind should publish, a review containing as many mistakes and misrepresentations as this one does, and employing such unfair tactics. The degree of scorn is intriguing also, given that other senior philosophers have judged The Act of Thinking to be a very credible and interesting contribution to the philosophy of mind. In this reply I go through Adams’ review in detail, justifying my allegation of mistakes, misrepresentations and underhand tactics and setting the record straight as to what is actually in my book. In an appended Discussion section I venture an explanation for the scorn. There are good reasons for expecting that The Act of Thinking would give any committed cognitive scientist, and perhaps a senior spokesperson especially, a rush of blood to the head.

The gist of Adams’ critique is that my book is behind the times, that it returns the reader to the dark days of ‘ordinary language philosophy’ and that the ideas in it are cribbed from Ryle, Wittgenstein, and Lakoff and Johnson. The ‘out of date’ message is hammered home:

Whether we go back fifty years (to Ryle and Wittgenstein) or a mere twenty (to Lakoff & Johnson), this book attempts to move us backward. One cannot overcome the impression of being dragged back into the heyday of ordinary language philosophy, where pronouncements are made upon the metaphysical nature of the mind and action all derived from features of the surface grammar of language. It is déjà vu all over again, and high time philosophers get over it and its vestiges. So to start, the best ideas in the book were stated better by other people first, and cognitive science has taken us past all this (or should have) in the last fifty years. Were people not paying attention? [Adams, p.448.]

This is warmed over Ryle at best, and it ignores most of what happened in the last half the last century, namely, we now understand how to make even guided missiles that are teleological and we are working on naturalizing meaning and mind. Where has Melzer been? Did the news not reach his part of New Zealand? [Adams, p.449.]

One ghost Adams raises can be laid at the outset. Adams tells Mind’s readers that I consider my task in The Act of Thinking to be the exposing of ‘category mistakes’. He uses the expression category mistake five times, beginning his review by stating my book’s thesis — approximately — and continuing, as if it is still my position he is expounding:

Any attempt to say otherwise is a kind of mistake… dare I say ‘category mistake’? Yes, that is right. This book is a re-make of ideas found in Gilbert Ryle’s Concept of Mind (1949) and in the latter Wittgenstein (1963). Melser even explicitly appeals to ‘category mistakes’, (pp.221, 233). [Adams, p. 447, no omission.]

Further on, he tells us that “Here is where Melser seems to see a category mistake… mistaking thinking for something that happens inside of one instead of something one does” [Adams, p.448, no omission]. And category mistakes come up twice in connection with my ‘action metaphors in science’ argument in Chapter 11:

How about the ‘action metaphors in science’ argument? It is the claim that ‘scientific explanation… of physiological events requires the use of action metaphors, and this makes physiological descriptions incapable of explaining things that literally are actions’ (p.228). Again, the best I can make of this is our old friend the ‘category mistake’. The idea is that much of science tries to anthropomorphize nature, make it teleological, or goal-directed. But if events are treated as from agents, then actions cannot literally be explained by science because science is not anthropomorphic, teleological or rational. It would be a category mistake. What is actional is metaphorical, and what is metaphorical cannot be mechanistic or scientific (pp.232-3). [Adams, p.449.]

First, how could the (correctly quoted) page 228 passage put anyone in mind of ‘category mistakes’? Second, the sentence beginning “The idea is…” is arguably similar to the first premise of my ‘action metaphors in science’ argument, but the remainder of Adams’ ostensible exegesis of this argument, from “But if events…” to the end, is entirely Adams’ creation — a kind of semantic stink-bomb with “It would be a category mistake” in the middle. He mails this horrid thing, with what seems to me a quite arbitrary page reference, to The Act of Thinking. But the main point is: the impression is being fostered in the above that I invoke Ryle’s concept of a ‘category mistake’, and that this concept is integral in my book’s argument. Professor Adams’ testimony in the page 447 passage above, that “Melser even explicitly appeals to ‘category mistakes’”, leaves no room for doubt. He is claiming I use the term category mistake (this is what “explicitly appeals to” means) at least twice in my book — namely, on pages 221 and 223.

The truth is that neither the term category mistake nor any implication or even inkling of it occurs at all, anywhere, in my book. What we find on page 221 is a description of people’s actions and natural processes as “two very different categories”. Neither the word category (or any cognate of it) nor the word mistake occurs on page 233. Adams is ‘quoting’ what he imagines between the lines. How sporting is that?

In chapters 8-11 of The Act of Thinking I do speculate about mistakes philosophers are prone to — including (1) the mistaking of figurative talk (especially metaphor, synecdoche, and/or nominalisation of a verb or adjective) for straightforward literal talk and (2) the methodological mistake, or naivety, of assuming that just because we can see actions being performed, they must be scientifically explicable. But none of the philosophical mistakes I talk about bears any resemblance to Ryle’s ‘category mistakes’.

Reading his review, it is often difficult to square what Adams says with the book’s contents. For the record, The Act of Thinking contains:

º an Introduction (of 15 pages) which moots, and argues a prima facie case for, the idea that thinking is a learned activity and not a process in the brain;

º Chapters 1 and 2 (totalling 37 pages) which critically review six twentieth-century theories of thinking that are ‘action-based’ rather than ‘brain-based’;

º Chapters 3-7 (101 pages) proposing a new developmental theory of thinking and showing (with the help of a lot of evidence from developmental psychology) how children are taught and otherwise acquire the ability to think;

º Chapters 8-10 (63 pages) which propose my ‘metaphorical-origin theory’, i.e., that our notion of mind is a byproduct of certain very useful metaphors, each of which highlights a noteworthy feature of thinking, and

º Chapter 11 (29 pages) containing among other things my ‘empathy argument’ arguing that people’s actions (including their acts of thinking) are not objectively observable and hence not credible candidates for scientific explanation.

Biological process or learned action?

In the Introduction of The Act of Thinking I argue that biological processes differ from learned actions, and that thinking looks much more like a learned action than a biological process. My pages 6-12 offer five different arguments both that (1) natural biological processes and people’s actions are fundamentally different and that (2) thinking is an action, something people do, and not a natural biological process. Adams’ verdict on these introductory arguments of mine is that they are bad and that they don’t exist.

Next, the arguments for Melser’s thesis are bad. For instance, early on (pp. 3-6) Melser paints as mutually exclusive the ideas that what goes on in one’s mind are information processing events that go on in one’s brain (on one hand) versus that thinking is a kind of activity that a person does (on the other hand). He (as Ryle and Wittgenstein before him) seems to take it as an unargued presupposition that it cannot be both. [Adams, p. 448.]

It is unclear whether Adams is talking here about the process/action distinction, or the different question of which one thinking belongs with, or both questions. But whichever he is talking about, his “he seems to take it as an unargued proposition” is untrue. As I say, there are five different arguments on pages 6-12, each clearly identified with a heading and each clearly arguing both of the propositions in question. The false impression that I don’t provide argument on these issues is reinforced later in his review, when Adams refers, in relation to my empathy argument, to “his presupposition”.

He mainly just repeats his presupposition that natural processes in the brain are one thing and learned actions (which he claims are “not a natural process” and are of a “different category”) are another thing (p. 221). [Adams, pp. 448-449.]

And where did Adams get his “as Ryle and Wittgenstein before him”? Ryle distinguishes actions and biological processes in a couple of places but there is only one Ryle passage I know of where Ryle says that actions are different from natural processes and that thinking is an action, not a natural process. I quote it on page 12 of The Act of Thinking. (Incidentally, Ryle is not presupposing that thinking is an action here, he is arguing for this.) But this passage is exceptional. Everywhere else this question arises, Ryle is adamant thinking is not an action. So, what passages in Ryle does Adams have in mind? I can’t claim to have read the entire published Wittgenstein. Heaven forbid. But I’ve read quite a bit, and I’ve never read Wittgenstein claiming either that natural processes are different from actions, or that thinking is an action and not a natural process. I’d be very surprised too, if either Ryle or Witgenstein ever published anything concerning “information processing events that go on in one’s brain”. Can Adams supply chapter and verse here? Or is it just that he thinks what I say is the sort of thing ‘Ryle and Wittgenstein’ might have said?

My theory of thinking

Chapters 3-7 of The Act of Thinking contain a hundred pages of theory and empirical findings relating to how thinking is learned. Arguably, ‘how thinking is learned’ is what the book is primarily about. But Adams’ only acknowledgement of this core part of the book is a parenthetical aside, denying it exists:

(By the way, how does one learn to think? I know one can learn to improve thought. I know one learns to play the piano… by thinking and practicing, but how does one learn to think, if not in part by thinking? Beats me. And it must beat Melser too, for he never explains, though he makes the claim that thinking is an action one learns.) [Adams, p.449, no omission.]

For a reviewer to inform would-be readers that an author “never explains” something is to imply not just that the author fails to explain that something satisfactorily but that he or she doesn’t try to explain it. When there are a hundred pages in The Act of Thinking specifically dedicated to explaining how one learns to think, it is difficult to avoid the conclusion that Adams either didn’t read or even see chapters 3-7 (or the Vygotsky or Hampshire sections in Chapter 2), or is deliberately misleading Mind’s readers as to my book’s contents.

What of Adams’ allegation that the whole book is a rehash of ideas found in Ryle, Wittgenstein, Lakoff and Johnson and/or ordinary language philosophy? Could this apply to the theory of thinking put forward in Chapters 3-7? Well, no. It would be apparent to any reasonably attentive reader of Chapters 2-7 that the main avatar of my theory is Vygotsky. Vygotsky was not an ordinary language philosopher. However, by the same token, none of the theories of thinking (from the behaviourists, the physiological abbreviationists, Ryle, Vygotsky, Hampshire, Sarbin, Mead, and de Laguna) that I review and criticise in the first two chapters of my book has any essential connection to ordinary language philosophy. And my own theory of thinking in chapters 3-7, which is significantly different from any of the theories I review and criticise in the historical chapters, has nothing whatsoever to do with ordinary language philosophy. Calling my theory ‘ordinary language philosophy’ is just odd.

It is particularly difficult to understand Adams’ insistence that my book is derivative of “Ryle and Wittgenstein”. In the hundred pages of theorising in my chapters 3-7, Wittgenstein is mentioned just once (on page 80). In his three-page review of The Act of Thinking, Adams mentions Wittgenstein four times. But, to my knowledge (pace Hacker), Wittgenstein never produced anything that could be called a theory of thinking, or a theory of ‘mind’. By contrast, Ryle produced at least four different theories of what thinking is. However, nearly all my discussion of Ryle is in the two historical chapters. In those chapters, apart from the page and a half I devote to Ryle’s incarnation as a logical behaviourist, I concentrate on his ‘adverbial’ and ‘refraining’ theories – only fragments of which are to be found in The Concept of Mind. Understanding why Ryle’s refraining theory (from his 1979 book On Thinking) is wrong was a big help to me in developing my own theory. But does this mean my theory is ‘warmed-over’ Ryle? In any event, Ryle always (except in the passage I quote on page 12 of my book) denied what is the main plank of my own theory – that is, that thinking is an activity in its own right, one which people must learn how to perform. Consequently, my account of thinking is very different from any of Ryle’s four accounts.

It is interesting that Adams’ mentions of Wittgenstein are all accompanied by mentions of Ryle, as if the two were inseparable. Is Adams telling us that Ryle and Wittgenstein co-authored the adverbial and refraining theories? Is he saying that Ryle and Wittgenstein were in communication with Vygotsky?

In relation to Adams’ ‘looking backwards’ charge, there is another misapprehension to be skirted. I have mentioned the behaviourists above, and Ryle’s reputed ‘logical behaviourism’. And there are three and a half pages in my historical chapters devoted to explaining and rejecting behaviourism. Perhaps because of these three and a half pages, or possibly because I talk about actions a lot, and call my theory an ‘actional’ theory, other reviewers have suggested – what Adams, with his introductory “the thesis of the book is that mental talk is really about action (behaviour) covert or overt” [Adams, p. 447] also suggests – that I am a behaviourist and backward-looking for that reason. However, I am not a behaviourist or a neo-behaviourist. Furthermore, apart from behaviourism itself, none of the theories of thinking that I review in The Act of Thinking is a behaviourist or neo-behaviourist theory or has any essential connection with behaviourism. I reject behaviourism, as many others do, on the grounds that it denies there is any identifiable process or activity that can be called ‘thinking’. But I also reject it on the ground that it presupposes, what is false – and what I devote Chapter 11 to arguing against – that simply re-labelling people’s actions ‘behaviour’ makes them observable and specifiable in a purely scientific, purely objective, way. ‘Behaviour’ in this putative scientific sense is a fiction. Try to imagine a science with ‘the things people do’ as its subject matter.

The metaphorical-origin theory

Well, if the ‘Ryle, Wittgenstein, Lakoff and Johnson, ordinary language philosophy’ charge does not apply to the theorising in chapters 3-7, does it apply to the account of metaphor and its role in folk psychology in my chapters 8-10? Actually, no again. My ‘metaphorical-origin theory’ is substantially original. W.G. Lycan has called it “…a novel error theory, indeed a novel kind of error theory”. Despite what Adams says, my theory doesn’t come from Ryle. You could fit everything Ryle said about metaphor on one page. Nor does the metaphorical-origin theory come from Wittgenstein. Wittgenstein muttered darkly about ‘pictures in our language’ a few times, but that’s about all. Neither Ryle nor Wittgenstein offered anything other than very brief remarks concerning the role that metaphor plays in our everyday talk about thinking. Nor was metaphor’s influence in folk psychology and philosophy of mind discussed at greater length by other ‘ordinary language’ philosophers. Has Adams evidence to the contrary? Dimly aware, perhaps, that sourcing the metaphorical-origin theory in ‘Ryle and Wittgenstein’ is too far-fetched, Adams finds someone else for me to have borrowed from:

Of course, the book is not only about ideas found in Ryle or Wittgenstein. The author also blends in some ideas about metaphor first argued for in Lakoff & Johnson (1980) – such as that metaphorical use of language distorts and mistakenly nominalizes words like ‘mind’ making us think there are things such as minds. [Adams, p.448.]

But my metaphorical-origin theory doesn’t come from Lakoff and Johnson either. In their 1980 book, Metaphors We Live By, Lakoff and Johnson never discuss the noun mind and the metaphors that surround it. Adams should have a look. Nor do Lakoff and Johnson say (any more than I say) that metaphor “distorts and mistakenly nominalizes words like ‘mind’”. For a start, they always talk about concepts rather than words – an important distinction in this area – and they never talk about metaphors either ‘distorting’ or ‘mistakenly nominalising’ concepts. They talk about metaphors ‘structuring’ concepts. In fact, the Lakoff and Johnson line applied to the ‘mind’ concept would look very much like the line Berkeley (and Hegel, Hampshire, and others) took: there is something called ‘mind’ but we rely on metaphors to help us understand it. According to my research, my more radical proposal — that the notion of mind is entirely a product of the metaphors – is nascent only in the work of Thomas Reid. Reid wrote two or three pages on the topic. But I didn’t pinch the ideas in Chapters 8-10 from anyone. Not even from Derrida – if that is what Adams is now thinking.

However, all this may be academic. Adams has done research of his own which he believes rebuts the metaphorical-origin theory.

While I will not have time to lay this out here, I have compared Melser’s list of metaphorical uses of the word ‘mind’ to similar uses of the word ‘heart’. If Melser has shown that ‘mind’ does not refer because of metaphor then he has also shown that ‘heart’ does not refer — and by exact parity of reasoning. What this tells me is that the reasoning is flawed, since there are hearts (and there are minds too, by the way.) [Adams, p. 448.]

Is this right? Well, no, it isn’t. The colloquial metaphorical uses of the word heart — and yes there really are hearts, and thus literal uses of the word heart — are not comparable with the colloquial metaphorical uses of the word mind. Like the words meaning, time, truth, concept, reality and perhaps language, the word mind has a much more intimate and difficult relationship with its metaphorical hangers-on than heart has with its. The correct analogy here, which does make possible an “exact parity of reasoning”, is the one between metaphorical and literal uses of the word heart, and metaphorical and literal uses of the word brain. To be honest, I’d never thought of comparing the literal and metaphorical uses of heart with those of brain. But it is a perfect analogy, and it enables us to state very simply what cognitive science is about. I’d always assumed (following philosophical history, I suppose) that the idea of the brain as the venue or agent of thinking derived from the prior notion of the mind in this role – and that identifying the mind as the brain was a recent intellectual move. That’s how cognitive scientists consider it — as their discovery. But, of course, the vernacular has for centuries been identifying the brain as the agent or locus of thinking — just as (and we should be grateful to Adams for reminding us of this) the vernacular has for centuries been identifying the heart as the locus and/or agent of emotions.

Here is a sample of everyday ‘heart’ metaphors: he has a good heart, her heart was filled with sadness, it came from the heart, he could not find it in his heart to forgive her, a heart to heart talk, a heartwarming tale, a heart-rending decision, he eventually lost heart, I knew in my heart that…, I have your best interests at heart, he won her heart, he has a cold heart. These expressions figuratively construe the heart as the agent or locus of feeling and emotions (of one kind or another). Now here is a sample of ‘brain’ metaphors: use your brains, brain work, she’s very brainy, a brainless scheme, a hare-brained scheme, my brain said yes but my heart said no, he racked his brains, my brain couldn’t cope, she had sex on the brain, a real brainwave, he’d been brainwashed into believing…, he hasn’t got the brains he was born with, she was the brains of the outfit, if you had half a brain you’d realise that..., a bear of little brain. These expressions figuratively construe the brain as the agent or locus of thinking and thoughts (of one kind or another). One might grin at the idea of a branch of cardiology dedicated to investigating the emotions – an ‘affective science’ which equated ‘heart events’ and ‘heart states’ with feelings. But cognitive science has an exactly analogous role with respect to the brain and thinking. It takes the everyday brain metaphors (and one should add the ‘in the head’ metaphors) at face value. It assumes that thinking really does take place in people’s heads, in their brains. And it calls in scientists in white coats to investigate.

The empathy argument

Another implied criticism of cognitive science in The Act of Thinking is the empathy argument in Chapter 11. Adams says I botch the empathy argument and that what I meant to argue is what simulation theorists have already argued (much better than I could). And he chides me for ignoring the autism and mirror neurons literature:

Melser makes a mess of the empathy argument. What he wants to say is something like what simulation theorists such as Goldman, Gordon and Heal say. That is, he wants to say that to observe and understand an action one must simulate the actor, simulate one’s performing the action oneself. Now there is interesting literature on autism and on mirror neurons and many other exciting and new things to be learned and to be said about this basic idea that Melser could use to support such a claim. Melser talks about none of this literature” [Adams, p. 449].

Simulation theory is a theory about how, in everyday interpersonal situations, people determine what mental phenomena are occurring in the minds of other people. By contrast, my empathy argument claims that the observation methods we use, and need to use, for perceiving and understanding other people’s actions preclude the scientific study of actions. The empathy argument poses a practical, methodological problem for cognitive science. So simulation theory and my empathy argument are about two very different things. Adams should ask Goldman, Gordon and Heal whether they think my empathy argument is about what their simulation theories are about.

It is worth a brief digression here. Simulation theory arose in opposition to ‘the theory theory’. Both theory theory and simulation theory posit the existence of things called ‘mental states’ literally inside other people’s heads, which we as observers of those other people want to find out about. Where theory theory says we use proto-scientific theorising and inferences from observed behaviour to identify the posited mental states, simulation theory suggests that we find out about them by simulating or empathising them. The reason I ignore simulation theory in my book, and the reason why on the one occasion I do mention it I call it “a version of theory theory”, is that it shares theory theory’s basic assumption – that inside people’s heads there are things called mental phenomena, which other people want to know about. The Act of Thinking is dedicated to denying this assumption. I don’t believe that, outside a scientific or medical context, anyone has ever wanted to know what is going on inside the head of another person. On the other hand, we constantly want to know what others are thinking or feeling.

The theory I put up in opposition to theory theory (and simulation theory) is the metaphorical-origin theory, which claims there are no such things as ‘mental states’ in people’s heads – there are only various acts of thinking, imagining, etc., they are performing, which colloquial metaphors persuade us to construe as ‘mental states’ or whatever. When we see other people saying or doing things, or listen to descriptions of others’ thinking couched in the everyday mentalist metaphors, we are not speculating about (and/or experimentally simulating for heuristic purposes) what is going on in those others’ heads. We are simply trying to empathise or imitate the imagining they are doing. As I say, colloquial figures of speech persuade us to construe these empathisings of ours as imagined witnessings of things going on in people’s heads. But really, intracranial goings-on don’t come into it. We’re just trying to do the same imagining they are. (See The Act of Thinking, pp.217-220.)

Like the simulation literature, the autism literature is constrained by the above ‘theory of mind’ assumption – that people have minds in their heads and other people want to know what is happening in them. The idea is that autism sufferers are no good at identifying others’ intracranial happenings. As for mirror neurons, the Rizzolatti findings were interesting ten years ago but, as no-one has yet specified just what the relation is between the firing of mirror neurons in P’s brain and P’s perceiving and/or empathising an action someone else is performing, my interest is waning. Perhaps someone has discovered the relation but the news hasn’t yet reached New Zealand.

Adams has another reason for dismissing my empathy argument. He thinks it “does not face the challenge” that the cognitivist orthodoxy poses to a purely actional approach to thinking. As a sample of the received views he outlines a version of token identity theory:

...any materialist about persons must take the person to be a composite of sub-personal events. Even if my mental doings are actions of mine, actions at a personal level of description, when I do something, my doing (my causing) something cannot be more than a collection of the proper parts of my brain’s causing something too. My doing is the relevant portions of my brain’s causing the relevant events. The personal level of description may employ mental terms and actional language to describe it, but the very same physical events that make my actions possible have to be able to be described in sub-personal terms. Indeed on a token-identity theory, they are the very same events variously described at different levels (something any enlightened materialist accepts as true). [Adams, p. 448.]

However, the empathy argument does constitute a challenge, a serious one, to the orthodox views, and to token identity theory in particular. The empathy argument is that (1) actions have to be empathised in order to be identified, or even perceived; that (2) this empathising precludes the kind of objectivity of observation that science requires (for identifying brain events, say); and that (3) this makes actions unsuitable as candidates for scientific scrutiny. One obvious corollary of this argument is that there can be no viable procedure for correlating ‘mental phenomena’ (or my acts or states of thinking) with brain events. This is a purely practical problem. Every time one shifts one’s gaze from an item on the ‘mental’ (or actional) list to an item on the ‘brain event’ list, one has to change one’s observation method. You can’t keep both lists simultaneously in view. This means there is no do-able correlation procedure. The claim that mental events (or actions) and neural events are “the very same events” is vacuous if there is no possible procedure for demonstrating the identity.

But Adams is having none of this nonsense:

Frankly, I find it very hard to believe that your perceiving my scratching my nose requires empathy... But this is what we ‘learn’ from the empathy argument. Now how is that a reply to my point about token-identity of actions with physical sets of events? Clearly, it is not… What could be a more objective, physical event than my scratching my nose? It can take place at a time (06.02.05 at 10:00 a.m. EST). It can be witnessed (by my office staff). It can be photographed. And so on. How much more physical and objective can something get? Well, that is the empathy argument. [Adams, p.449.]

How is this a reply to the empathy argument? How does Adams imagine one would set about determining the physiological correlate of such commonplace actions as, say, ‘accepting a dinner invitation for Thursday, then changing one’s mind and suggesting Saturday’, or ‘selling the old mare for a good price’ or even ‘scratching one’s nose’? Where would you start?

Adams has a third card to play. He accuses me of the fallacy of arguing from the epistemological to the metaphysical: this means, arguing from a fact about how we know something to a fact about what it is really like:

Worse yet, Melser’s proclaimed point is supposed to be about observation of actions… an epistemological point. Instead it comes out metaphysical: ‘Actions require empathy’ (p.222). Surely, scratching my nose when it itches is an action. Surely scratching my nose does not require empathy… He thinks that if we have to simulate actions to observe them, then they cannot be objective, physical events. He hints that we cannot even refer to them (p. 245). But this does not follow. It mistakes the epistemological for the metaphysical (as noted). [Adams, p. 449.]

Certainly, my Chapter 11 is about the epistemology of actions. The title of the chapter is ‘Our knowledge of actions’. And the empathy argument operates entirely on the epistemological level. It contrasts how we know actions with how science knows its subject matters. Adams’ accusation – that I illegitimately infer from facts about how we know actions to what actions are really like – is based entirely on one sentence, “Actions require empathy”. This quoted on its own appears to be a claim about actions per se. It seems to be saying that actions somehow ‘have empathy in them’ or some such. But if one now adds in the sentence preceding it on page 222, one’s eyebrows lift again. What I wrote here is:

I argue that we are unable to correlate people’s actions with physiological events, because the two subject matters require different heuristic methods for their individuation and observation. Actions require empathy. [The Act of Thinking, p.222.]

With this contexting it is obvious I am saying that actions require empathy for their individuation and observation. This is an epistemological claim. It is only Adams’ omission of my ‘for their individuation and observation’ qualification that allows the impression that what I say “…mistakes the epistemological for the metaphysical”. I don’t make this mistake, here or elsewhere, and it is trickery on Adams’ part to withhold necessary context in order to create the impression that I do.


Adams ends his review with a list of things I say in my book that, he says, “leave one stunned”. It is clear Adams thinks it is incomprehension or incredulity rather than, say, admiration or gratitude doing the stunning. He admits the quotes are taken out of context – “…but, even so”, he says. Here they are:

“…mind is the name of something abstract” (p. 179), “…we do not have, independent of the metaphors, any concept of the mind at all” (p. 181), “metaphorical-origin theory suggest that we are largely ignorant as to what kind of activity thinking is…” (p. 182). “A word is essentially an action…” (p. 210), “…things cannot be about things” (p. 210), “thinking is a learned action and not a natural process” (p. 221), “Actions require empathy” (p. 222), “…simulation theory [is] a version of theory theory that claims…” (p. 225). “One’s eyes are not literally… something one uses to see ‘with’…” (p.233). [Adams, p. 450.]

It is revealing to go through these, supplying the context Adams leaves out.

“...mind is the name of something abstract...” On page 179 I am discussing what other philosophers (including Berkeley, Hegel and Hampshire) have said about the fact that, in everyday speech, the noun mind is always used in association with some metaphor or other. I venture that:

It is almost universally assumed – both by philosophers who think that the mind metaphors are important and those who don’t – that mind is the name of something abstract, which we are initially acquainted with, and form a concept of, without the metaphors. [The Act of Thinking, p. 179.]

Adams’ quoting just “…mind is the name of something abstract”, leaves the false impression that it is me who is saying that mind is the name of something abstract. He forgets to italicise mind too.

“...we do not have, independent of the metaphors, any concept of the mind at all” On page 181, I am introducing my metaphorical-origin theory and contrasting it with theory theory. I say:

I claim that the traditional view outlined above is misguided and that we do not have, independent of the metaphors, any concept of the mind at all. I say there is no sense in which the mind is ‘there first’. Our notion of mind is entirely a product of the metaphors. The metaphors accompanying the word mind in everyday speech constitute, rather than just supplement, any understanding we have of mind. Furthermore, they do not furnish us with a concept, let alone a theory, of mind. They provide only numerous diverse and mostly incommensurable images, collectively undeserving even of the term notion. Collating a concept from such ill-assorted imaginings would be impossible. [The Act of Thinking, p. 181.]

I also say that my

...metaphorical-origin theory suggests that we are largely ignorant as to what kind of activity thinking is and that the imaginings we indulge in connection with the figures of speech in the colloquial thinking vocabulary allow us the impression that we do know roughly what thinking is. [The Act of Thinking, p. 182.]

The metaphorical-origin theory is presented as an alternative to theory theory. It takes three chapters to spell it out and, because it is a new and rather radical proposal, one-liners abstracted from it, without their three chapters of supporting argument, are always going to look bald. Bald isn’t such a big deal, though. Is Adams implying that my “we are largely ignorant as to what kind of activity thinking is” is not true?

A word is essentially an action...” and “...things cannot be about things”, from page 210, look quite sensible and interesting in their context too. I am discussing intentionality – in the philosophical sense of ‘aboutness’. Here are two paragraphs from that discussion.

If there is such a thing as intentionality, then in my story it is a function of the whole referring transaction – a transaction comprising two or more agents, perceptual activity, ostensive gestures, speech, and concerting procedures. The conventional notion of linguistic intentionality is problematic because it is specified as attaching not to whole transactions but to words or other linguistic items...

A word is essentially an action (or action component) and is thus inseparable from its wider transactional context. To imagine that words can somehow exist in their own right, as (linguistic) objects in the world, is to contemplate the bizarre notion that they can by themselves reach out and ‘mean’ other things in the world. However, as I said earlier in connection with the fancy of ‘words referring to things’, it is only by synecdoche that one ingredient in an activity (a vocal utterance, say, or a mark on paper) can accomplish what the whole activity accomplishes. The referring transaction as a whole can perhaps be said to have intentionality. Referring sessions can be about things. Other actions or activities can be about or directed toward things too. But things cannot be about things. [The Act of Thinking, p. 210.]

“...thinking is a learned action and not a natural process” Before I introduce the empathy argument at the beginning of Chapter 11, I sum up what the book has attempted up to that point:

In the Introduction I argued that people’s learned actions and natural (in particular, biological) processes are two very different categories, and I give reasons for believing that thinking is a learned action and not a natural process. The chapters since then have continued the argument that thinking is a learned action. I have attempted to show, among other things, what kind of action thinking is and how it is learned. [The Act of Thinking, p. 221.]

That thinking is a learned and voluntary action like speaking or swimming and not a natural, internal-bodily process like digestion or conception is a serious philosophical contention – on behalf of which I argue at great length in The Act of Thinking. Adams seems unable to open his mind to this perfectly sensible alternative possibility. He seems unable to accept the possibility that thinking might be something people do. In a philosophy professor, that’s very surprising.

Actions require empathy” from page 222, we have already dealt with. Without the “…for their individuation and observation” from the previous sentence, it does look a mite incomprehensible. But who would quote it on its own?

“...simulation theory [is] a version of theory theory...” I have explained above too. As for the actual context, on page 225, I am talking about the arguments Dennett supplies for adopting ‘the intentional stance’ towards others, and I comment:

This argument is similar to that of simulation theory, a version of theory theory that claims that empathy – as opposed to logical and inductive inferences from observed behaviour – is our preferred means of determining what mental states prevail in the minds of others. Here again, empathy is thought to be justified on the grounds of heuristic necessity. [The Act of Thinking, p.225.]

I group theory theory and simulation theory together, and contrast them with my metaphor-based theory, on the basis they both assume there are minds and related mental phenomena inside people’s heads, whereas my theory posits subtle actions that people perform – and metaphors with which we (mis)construe these subtle actions.

One’s eyes are not literally… something one uses to see ‘with’...” This final stunner would boggle most people. One wants to ask what in God’s name we do use, then, to see with – and suggest other (possibly rude) body parts as alternative candidates. However, when you look at the context Adams has lifted this from, you find a remark most people would agree with, and some would find very interesting. On page 233 I am describing a kind of metaphor – one of the rhetorical tricks in the cognitivist kit – in which actions are construed in terms of the ‘using’ of relevant body parts. What I say is:

One’s eyes are not literally – as a telescope or an infrared sight is literally – something one uses to see ‘with’ or ‘through’. Use your eyes is not really an injunction to make use of equipment; it is only an injunction to look. In the normal case, simple looking requires no accessory hardware. [The Act of Thinking, p. 233.]

I mean, does a chap use his scrotum to carry his testicles around in? In the same sense that he might use a shopping bag to carry his tracksuit in? Would Adams say that? At any rate, this is another case of Adams’ omitting relevant supporting text to make what I say look odd.


Back to Descartes

Adams’ criticisms of The Act of Thinking suggest broader issues worth clarifying. What is or was ordinary language philosophy? Is there such a thing as philosophical progress, and regress? The tone of his review also raises questions. The review seems just a little too perfunctory, too scornful. It protests too much. Argument and counter-argument and continually going back to square one should be meat and drink for philosophers. But Adams is impatient, and indignant. What at? He might say, at such a poorly-argued book. I might say, at ideas so different from those presupposed by cognitive science. His tone implies he has, and expects from others, a moral commitment to cognitive science.

I would like to make it very clear that in the off-the-cuff speculations which follow, I am venturing well outside the ground covered in The Act of Thinking. My aim here is not to explain where the notion of mind comes from, or what thinking really is, but to offer a tentative explanation why a cognitive scientist might go off the deep end at the suggestion thinking is a learned and voluntary action that is derivative of, and still rooted in, concerted activity. I must concede Adams’ intuitions about what is going on between the lines, behind the scenes, in The Act of Thinking are in fact not too wide of the mark. However, his job as reviewer was to comment on what was in the book. I remember being very impressed as a child by the part in a Superman comic in which Clark Kent fails the eye-test in his Army medical exam by dint of inadvertently reading the chart in the room next door.

There are many interesting opinions about what is the subject matter of philosophy, and what the philosophers’ job is, or should be. My own view is that their subject matter is the fundamentals of culture. Their job is to devise and debate theories about the nature of certain culturally basic activities and institutions – verbal communication, thinking, concerted and cooperative action, solo action by individuals (personhood), our ways (such as science) of acquiring knowledge of the world, and ethics, art, law, religion and so on. Obviously people depend hugely, totally, on the cultures they live in. And philosophers call cultures, call the building blocks of cultures, into question. Philosophy is disturbing. Philosophical issues are sensitive, like religious issues – but perhaps deeper, even more difficult to comprehend.

The social anthropologist Levi-Strauss says there are contradictions inherent in the rationale of any culture, and a need for myths to paper over cracks. If they are there, the deep contradictions, the imponderables and irrational coercive elements, they may have something to do with the fact that we evolved for small-group nomadic life and now live in settlements of thousands, millions, billions. At any rate, I believe there are modern myths. Rather than being communicated in explicit traditional narratives, myths are these days communicated and preserved almost subliminally, in habitual figures of speech. Innumerable unobtrusive metaphors make a fine safety net for our sense of rationality and our equanimity. Philosophers should talk to anthropologists and sociologists of knowledge (if there are still such) about these things.

Plausibly, the function of myths is to rationalise-over anomalies in the assumptions underpinning a culture. There is a need partly to obfuscate and partly to put a good, useful face on, the social basics and any difficulties therein. There is no mystery about metaphor’s role in myth-making. Metaphors were/are specifically devised to obfuscate some aspects of a subject matter, and to highlight and put a useful (perhaps distorted) face on other aspects.

Given cultural myths as our first line of defence against the inevitable lacunae in a culture’s rationale, against the inevitable cultural black holes, the philosophers’ job is perhaps best redefined. We can say that philosophers undertake repairs and maintenance, valuations, and occasional upgrading and modernising work, on myths. Like the myths themselves they are employed – by the culture they live in, so to speak – to protect our morale at a deep level, and our rationality. But these two aspects of the job description inevitably come into conflict. And this gives rise to two different kinds of philosopher – often in the same philosopher. The conservatives are concerned to make the existing myths look watertight and convincing, and thus maximise their morale-maintaining and rationalising potential – and thus maintain the cultural status quo. The other philosophers, the radicals, are more interested in dismantling old myths, or their incoherent elements, and exposing the deep forces and difficulties underneath – hopefully as a preliminary to cultural change, improvement. The antagonism between the two kinds of philosopher can be bitter. As I say, the whole game is disturbing.

If you squint a bit, you can see this drama being played out in Adams’ flare-up at my book. We can spell out the context, in terms of what I’ve just speculated, as follows. It is necessary in any modern culture that people act not only with others in groups but also, and mostly, alone and independently. These roles are in many ways incompatible. But both are necessary. To rationalise this necessity-in-incompatibility, we need a convincing concept of ‘the individual person’. And here is where the notion of mind is indispensable. The mind inside the individual as it were blesses the individual with the intelligence and authority to act autonomously. In the modern Western cultures the default philosophical theory of mind is Cartesian dualism, backed up by Cartesian mechanism. Descartes’ dualism and mechanism are founded on the same metaphors of internal agents and mechanisms (controlling people from within) that lay folk find, and for many hundred, possibly thousands, of years have found, so useful and so charming for talking about thinking. Descartes was the first to synthesise and formalise these metaphors, transforming them into a half-way-acceptable-looking doctrine.

Adjustments to this basic myth (or theory, or myth-cum-theory, or metaphor-spawn) of mind would both reflect and affect attitudes and morale throughout a culture. Thus, if the equation of mind with brain were to be generally accepted, together with a biological perspective – in which people are organisms dedicated to their own survival, their actions determined by mechanisms (which can be understood only by scientists) in their brains – this would tend to reinforce the traditional myth, to bolster it with science’s sanction. And this would tend in turn to reinforce certain existing attitudes and ways of behaving, certain social policies, ways of organising the economy, etc.

I do believe there is, or can be, progress in philosophy – that we can progressively dismantle and/or improve cultural myths without endangering morale. Regress is also possible, going back to the dark days of whatever. Sheer stalling, marking time, obfuscatory tactics on the conservatives’ part, is very possible also. That can go on for ages. It is salutary to remember that for centuries the best minds in Europe were employed in determining the physical properties of angels. Maybe this was a way of forestalling serious investigation of the moral and religious fundamentals of the time. In America now, Adams tells us, philosophers are “working on naturalizing meaning and mind”. The near-fifty years that cognitive science has been in vogue is not much in philosophical time. Very possibly, we can look forward to several more decades of stalling – awaiting impending exciting new discoveries about consciousness and the brain, imminent huge breakthroughs in science’s knowledge of mind, startling new research concerning the neural basis of emotions, significant advances in the bold new field of the biology of language...

Ordinary language philosophers were a radical bunch who claimed that philosophical theories can be vitiated, and the traditional ‘metaphysical’ theories such as Descartes’ theory of thinking were often chronically vitiated, by a previously unrecognised kind of problem. The problem lies in the assumption that everyday (‘ordinary language’) expressions, which have acquired their meanings through their historical usage for particular purposes in particular everyday social and practical contexts, will continue to mean the same when they are borrowed by the philosopher for use in his or her special technical contexts. The philosopher’s use of everyday terms out of their home contexts either constitutes or results in a special kind of logical or grammatical infelicity. Different ordinary language philosophers defined the infelicity, or attempted to define it, in different ways. Ryle’s ‘category mistake’ was one such (in my opinion failed) attempt. The idea of different ‘levels of description’, which cognitive science still relies on – see Adams’ rendition of token identity theory above – is another defunct relic of ordinary language philosophy.

I agree with ordinary language philosophers that philosophical theories are often (and the traditional metaphysical theories are characteristically) flawed by unacknowledged borrowing of expressions designed to be used in quite different (and ‘everyday’) ways and contexts. My view – and, to my knowledge, no other philosopher has proposed this – is that the expressions philosophers borrow from colloquial language (and naively expect to mean the same in high-flown philosophical discourse) are all simple figures of speech. In The Act of Thinking I concentrate on metaphor, but it’s not the only philosophically important figure of speech. The important ones are metaphor, synecdoche and nominalisation of verbs and adjectives. Hacker and Bennett, in their excellent book The Philosophical Foundations of Neuroscience, call synecdoche ‘the mereological fallacy’. My claim is that philosophers tend to take certain key families of figurative expressions literally, without realising they are doing so, and this is what gives rise to the infelicities that ordinary language philosophers were on about. Taking figures of speech literally produces all sorts of absurdities – not so much in what is said (because of the familiarity of the figure of speech, this often sounds quite plausible) but in what is presupposed by what is said.

If the above story is true, more or less, it may be that the innumerable culturally-influential figures of speech in colloquial language can never successfully be parleyed into coherent theory, even by a philosopher as clever as Descartes. They will always resist explicit formulation. Perhaps we can’t even talk about ‘myths’. Perhaps the metaphors are always too diverse to combine, and perhaps the obliqueness, the merely suggestive feature of metaphors, is indispensable. Perhaps it is hinting, intimating that is required for the cultural job, and precisely not explicit theorising.

At any rate, whether or not my views on taking figures of speech literally make me an ‘ordinary language philosopher’ – do Lakoff and Johnson qualify as ordinary language philosophers? – the taking of figures of speech literally certainly looks like a mistake. Against a charge of unacknowledged metaphor, synecdoche or nominalisation, a philosopher can argue either that he does not use the expressions in question in his philosophical theorising, or that these expressions are not figurative and/or not everyday-context-bound, or that they are figurative but taking them literally is not a mistake. None of these arguments is consumated by dismissive hand-waving about “being dragged back into the heyday of ordinary language philosophy”.

The way I see it, in the middle half of the twentieth century the default Western theory of mind, Cartesian dualism, was temporarily embarrassed into the background by, first, logical positivism and behaviourism, then ordinary language philosophy. When behaviourism and ordinary language philosophy bowed out in the seventies, Descartes’ story simply took up residence again. As Hacker and Bennett show, cognitive science is patently a re-mix of Cartesian mechanism and Cartesian dualism, with modern-looking handles on it.

The appeal that cognitive science has for professional philosophers and the public alike is the age-old appeal of these ‘inner agent’ metaphors – and to the new fancy of scientists in laboratories bringing the mysterious inner agents to light. But as a theory about what thinking is, cognitivism doesn’t rate. It is not so much a way of explaining thinking as of mystifying it. If philosophers really want to know what thinking is, they should read at least Vygotsky’s account, Hampshire’s theory of emotion, the work of Sarbin, Mead, de Laguna and others. They should at least acquaint themselves with the research work done by Bruner, Butterworth, Trevarthen, the Papouseks, Savage-Rumbaugh, Carpenter, and others. These people study children’s learning to think – their learning to do things in concert with others, learning to initiate sessions of concerted activity by token gestures (including speech), learning to perceive things in the world, learning to play pretending games, learning to re-enact whilst alone others’ speech and others’ actions, learning to rehearse actions (including demonstrative actions) covertly, learning to rehearse speech covertly, and so on. Philosophers interested in what thinking is could even ignore Professor Frederick Adams’ advice and read my book. Alternatively, so-called ‘philosophers of mind’ should explain what it is these children are learning – and spending such a huge amount of time and energy learning – if it is not the component skills of thinking.

— Derek Melser —