A Gathering of Straw People
There are a couple of discussions of Evolutionary Psychology at Twisty Sticks: one here and the other here. They’re interesting because of what appears to be a fairly unshakable assumption that all evolutionary psychologists have a right-wing agenda and that the agenda determines their conclusions. That’s probably true of some evolutionary psychologists – I think I’ve read one or two of those – but it’s not true of all of them. It’s a bit puzzling. It’s not easy to figure out why people are convinced that thinking natural selection might have played a part in making human nature what it is requires being a free marketeer. I’m not a free marketeer, and I think natural selection played a part in making human nature what it is. I don’t quite see how it could be otherwise, really. How could natural selection not have played such a part? How could we have evolved over millions of years in complete independence of selective pressures? How exactly would we go about doing that? With a little help from the God of the Gaps? Is that it? If not, then what? It would be a pretty good trick.
But, then again, maybe I just don’t understand the points that are being made. But I can’t help thinking I detect a good deal of rhetoric in play, a certain amount of deck-stacking. Though maybe I’m wrong. There’s a very interesting archived discussion of related issues among Steven Pinker, Janet Radcliffe Richards and John Gray on In Our Time in November 2002. Anticipating this discussion at Twisty Sticks, I happen to have re-listened to it a few days ago. And the email interview B&W did with Steven Pinker about a month before that is also relevant and worth a read.
Update. Our old friend the Anonymous One has made one of his classic comments, where the retort is so obvious it would be too cruel to make it. Too like aiming projectiles at a piscid in a water-containment device, as Chris said (using other words) of the sweet bit of Lacanian profundity I gave you the other day. So I’ll make it here, instead, where it’s not quite so cruel. He does have (Anonymous, not Chris) such a way of saying things that apply to himself better than they do to the people he’s saying them of. It’s quite hilarious really. A few weeks ago – this is really funny – he called Scott McLemee a talentless ankle-biter! I nearly fell onto the floor laughing at that one. Yo, dude, if you’re going to pee, don’t do it upwind because – oh too bad, too late. I’d run home and take a shower if I were you.
You should check out the wonderful passage from Higher Superstition where they argue that—just hypothetically, mind you—the science faculty at MIT could do a better job at teaching the humanities classes than vice versa, if it came down to that somehow…I certainly would feel better about the future with an English professor teaching differential equations than with Pinker teaching literature or philosophy. At the least the former would be humble enough to try to learn something about what they were doing beforehand, which Pinker rather plainly would not.
Right. Uh huh. And if the English professor is someone like, oh, say, you, which I’m afraid it all too probably is – well, I know which of the two I would rather be taught by. But I won’t say which one that is; it would be too cruel.
I just got back from listening to your BBC selection. Really, John Gray?
Objections to ev psych do not amount to simply a denial of the natural origins of so-called “human nature”. To the contrary, there are any number of objections to be raised to the terms of its construction of such and to the scope it grants itself and its constructions. To begin with, the gene selection theorem is not such a momentous innovation in biological science that its underwrites a revision of all of our previous conceptions. It is actually a small contribution to evolutionary theory that sits beside others. Further, genes are not wholely determinative of biological reality; the complexity of their functions and their interactions with other biological organizations is coming into view- (as witness the “disappointment” of the human genome project, in which the number of genes found was about half of that expected, such that the genetic difference between a human and a fruit fly is less than previously supposed.) So the deployment of biological ideas outside of the proper domain of their application is bound to raise suspicions and questions as to how they function outside their integumen. (I myself regard evolutionary theory as an “idea of reason” in Kant’s sense,- though Kant would have had no such notion and I myself reject Kant’s recourse to transcendental constructions,-that is, as at once a ground for the critical testing of ideas and as a source of “transcendental illusion”.)
Just consider ev psych “explanations” about violence,- (though aggression would be the more proper biological topic, even aside from problems of definition.) Consider that only human beings, because they possess or are possessed by a language, which renders possible a horizon of otherness, as well as a horizon of temporal expectation, are capable of paranoia.
Speaking for myself, my objections about EP are not so much the idea, as the execution. In particular, EP is very much a social science from my exposure to it, but dressed up in Darwinian garb. So the standard of evidence is pretty poor and the theories close to unfalsifiable. Alternative and highly plausible social (i.e. non-biological/evolutionary) explanations are ignored if they contradict the thesis or deployed if they help to explain away contrary evidence. To some extent this is a limitation of this kind of reasoning in evolutionary behavioural ecology (a field often attacked for ‘just-so’ stories) but it is a lot worse in EP, I think because of the preponderence of soft psychologists and the subject matter.
But the thing I really hate is the use of ridiculously complex adaptationist arguments (e.g. for rape) when there are often nice simple adaptationist arguments (i.e. sex) that can explain the same phenomena. And often the reason for this is adherence to the philosophically and scientifically indefensible concept of modularity inherited from cognitive neuropsychology.
But I’ll go and have a look at those discussions and see what I think, been a while since I studied it…
Yes, I tend to think of evolutionary pyschology as consisting to a large extent of just-so stories that lack evidence but which match a set theory. It’s all very interesting and suggestive, but far from conclusive. The most egregious example would be Pinker’s views on art, which seem to me to be essentially motivated by prejudice.
Of course, there are left-wing Darwinians (I don’t know about EP acolytes specifically) like Peter Singer and historically the left was more interested in genetics. The problem was that that tended to manifest itself as using eugenics for social improvement; unsurprisingly the left dropped that one after WW2. But on the whole, EP does seem very conservative to me – it often seems to assume human nature to be cast in stone and that change is not possible (this may be due to the EP writers I’ve come across more than the field in general); wasn’t that exactly what Hirschman considered to be the defining characteristic of conservatism?
Anyone read Singer’s ‘A Darwinian Left’? I have to agree with his motivation, the left has had a knee-jerk opposition to biological research in human behaviour for too long. Unfortunately it is really bad, another case of a good idea, shame about the execution. He really doesn’t understand his material, and his stuff about how the Prisoner’s Dilemma can tell us about free-riders is laughable. Still, at least it’s short.
Regarding the second CT article you link to — it seems like an exclusive critique of badly done evolutionary psych. The author does not comment one way or the other on properly done EP. I agree with the author because there are a lot (and I do mean a lot) of armchair EP hacks who think they can come up with sound EP arguments for everything under the sun, even though they know nothing about hard science and external validity and internal validity and hypothesis formulation.
Hmmm, I’ve just had another look at the CT debate. I take the point about Dawkins being a Darwinian descriptively and very much an anti-Darwinian normatively, but I would have to point out that my understanding of his reason for that is the division between genetic and memetic evolution, the latter working in a Lamarckian rather than Darwinian fashion, thereby allowing greater scope for progress in human nature. The problem is that if there is any evidence for such a thing as a meme to exist, then I must have missed it.
As for Singer, yes, I have read it. A worthy attempt and important as a corrective to some right-wing elements of EP, but unsatisfactory in many other respects.
Unfortunately the inherent plausibility of their stated aims, and of the individual explanations they offer makes it difficult to point out that it is still bad science. I’m sure there must be some good EP out there but as a field it is pretty bad.
Then again, I think a lot of cognitive psychology is pretty bad too, but it doesn’t quite get the same media attention – I don’t think your average member of the liberal intelligentsia has many opinions on connectionism or language-of-thought.
Guys
I’d be grateful if you didn’t include html code in your postings. Because if you mess it up (as one of you did in this thread), it causes problems.
I don’t have time at the moment to amend the scripts which drive this ‘blog’ (to disallow html), so it’d be good if you could just self police.
Thanks.
Oops.
I’ve never heard Dawkins talk much about EP, anyone else? I always assumed he wasn’t much impressed by it, since it does distort and misuse a lot of his sort of biology. Although he used to do some lectures in a course that included some teaching of EP as part of it (not taught by him) at Oxford. The jacket for Singer’s book includes an endorsement by Dawkins for the series, if not the actual book.
Sorry OB, I wasn’t trying to put words into your mouth but to try and braodly mark out the different camps in the debate.
PM
You’re quite wrong about this EP thing. Just thought you might like to know that. :-)
Actually it was I who used html earlier. Sorry for the messup. I tried to correct it with my next posting but Jerry seems to have fixed my original comment.
Anyway there’s a good article here:
Caporael, L. R. (2001). “Evolutionary psychology: Toward a unifying theory and a hybrid science.”
Annual Review of Psychology, 52, 607-628.
Unfortunately it’s not online
JS-about Dawkins’ position or about EP being a bit rubbish in general?
It’s me again. There are a couple of pages from that article available online here and here.
PM
I think you’re probably right about Dawkins (for example, he says explicitly in several different places that he just isn’t very interested in human beings).
I think you’re over-generalising about EP. For example, I’ve got a transcript here where Steve Pinker says exactly what constitutes good EP and bad EP, what the criteria are for falsifiability, what kinds of EP explanations fall under the “just-so” story suspicion, etc.
Also, your juxtaposition of the social and biological would, I suspect, have Pinker spinning in his grave were he dead.
There is bad EP, of course, but hey there is plenty of bad sociology, for example. (I should know having been responsible for some of it.)
It just seems that the most prominent stuff is the bad stuff, or at least the stuff they made me study. I’m not saying some of their explanations mightn’t be right, its just the standard of evidence isn’t quite what I might like. As I said, its partly the limitations of the field, human and nonhuman – but the biologists usually have more of a chance to argue that the environment hasn’t changed much from the EEA, and to take more reliable measures of evolutionary fitness.
Incidentally, does Pinker do much EP work now? I mean last time I came across his academic work it was some odd psycholinguistics stuff and he was just a populariser of EP, does he concentrate on EP now?
PM
Well, Pinker is convinced that we know plenty about the environmental circumstances in which the brain evolved. When I chatted with him about this (sorry that’s a bit name-dropping, but it’s the reason that I can’t give you references to back up what I’m claiming), he spent about ten minutes listing what we did and couldn’t know.
I think he’s still mainly working on linguistics, but, of course, that doesn’t preclude a kind of EP approach. He gives a graduate seminar on EP, I think (or at least he did when he was at MIT).
Well Pinker may or may not be right about the EEA (I’m not really qualified to comment, although I’m not convinced) but the problem is testing these kinds of evolutionary arguments. Most of the usual tools of the biologist are missing so it is very difficult to assess experimentally whether the adaptationist argument is right or just plausible. For a start, measures of offspring number (usually used as a proxy for fitness) are gonna run into some limitations with humans and our fertility manipulations (e.g. education is maladaptive for women in terms of fitness), not to mention the distance between the EEA and now. Now, as I mentioned before, even studies of this kind in animals are derided for being too limited but at least in animals you can manipulate the putative adapted trait to see what happens to fitness, or you can look at animals lacking it. I just don’t see that the necessary tools are there to make EP a science – at best it is a social science.
And don’t even get me started on modularity!
And as I mentioned in the main comment, Pinker also talks about Just So stories and falsifiability in his interview with B&W. So that’s one reference we can give.
“As for Dawkins – no, I don’t think so. I’ve heard and read him frame the matter very much in ethical terms.”
Doubtless. Ethics being a memetic concept to large extent.
“Pinker also talks about Just So stories and falsifiability in his interview with B&W.”
Hilarious!
“at best it is a social science.”
Hmmm. Well I think you misunderstand the kinds of reasoning and evidence which are applicable to EP. But anyhow, here’s Pinker on the point about EP being a social science (please don’t quote this anywhere else, not that you’d be particularly inclined to, I’m sure):
“The right comparison is not between evolutionary psychology and physics, it’s between evolutionary psychology and non-evolutionary psychology… You can’t compare water and the human brain. You can compare the study of the human brain, whilst walling off the body of knowledge we call evolutionary biology from that which does not incorporate it. That’s the relevant comparison.”
Ah. I’ve figured out why Dawkins’ anti-Darwinism in a moral sense may not be as widely known as I expected it to be: the most explicit statement of it that I’ve seen turns out to be the first essay in A Devil’s Chaplain which I thought had been published elsewhere, like the others in that book, but in fact it hadn’t. It’s an excellent essay – you’ll all just have to rush out and buy the book. You should anyway. It’s good. And it’s edited by Latha Menon, who wrote that wonderful essay for us.
I’ll have to give some quotations from the essay.
Dawkins also calls on people to rebel against the tyranny of their selfish genes at the end of The Selfish Gene.
And in the interview I did with him explicitly talks about the Is/Ought gap, etc.
And in his article attacking Midgley’s Gene Juggling says that a society based on Darwinian principles would be pretty horrible.
To be blunt, if people don’t realise this about Dawkins, then they’d really be better off keeping quiet about this stuff.
But then most people who talk about this stuff – including the usual suspects on Crooked Nonsense – would be better off keeping quiet about it.
Ha. Yeah. You got that in there while I was dutifully typing. Anyway there it is now.
I’m pretty sure there are some articles at World of Dawkins where he says this sort of thing, but it may take me a little time to find them. Curious readers should look.
Hmm, still see EP as more like the social sciences my self. In particular I really don’t get the comparison between EP and psychology – for a start ‘psychology’ is a bit of a wide term, lets say he means cognitive psychology, now I don’t see how the study of the phenomenology of human cognition and microcognition (which is what cognitive psychology is, I don’t really see it as all that explanatory) is similar to evolutionary/adaptive explanations for more global, even folk psychological cognitive processes (hypothetical cognitive processes at that).
“which is what cognitive psychology”
No, it’s not. At least not in any straightforward sense the *phenomenology* of those things you mention.
Anyway, the significant point is that Pinker doesn’t think EP is the same kind of thing as the hard sciences. Psychology isn’t normally considered a hard science.
But psychology isn’t usually considered as soft as sociology on the other hand. And really psychology is a diverse assortment of disciplines ranging in scientificity (made that word up ;)) from psychobiology to social psychology. But most of psychology has some adherence to the experimental method.
As to what cognitive psychology is, perhaps my choice of words offends the philosophical ear, what I mean is that it examines the phenomena, is reaction time speeded or slowed under these conditions etc. rather than being explanatory, reaction times are slowed under these conditions for this reason…
PM
Well, I just don’t think you’re right about what cognitive psychologists do. What you’re describing sounds like some kind of variant of behaviourism.
Have a look at some of the work which cognitive psychologists have done on memory, for example. It is explanatory, not simply descriptive.
The other thing is that I think you’re underestimating the extent to which it is possible to specify a rigorous methodology for EP. Have a look at the stuff Pinker has written about “reverse engineering”, for example.
Do you think that positing putative functions for cognitive/behavioural traits is a rigorous methodology? Isn’t it just another way of saying ‘just-so’ stories? In the past these sort of stories were just speculations that accompanied more substantial research programs within a particular field. It is only recently that anyone has thought that you could combine these speculations together and call them a science.
To say that the mind is a collection of ‘organs’ of cognition, each evolved for a different function, is way out there scientifically speaking (and a bit dualistic to boot), just ask any neuroscientist or psychologist – some might believe it but you are using sneaky rhetorical moves if you want to imply rejecting these assumptions is somehow anti-Darwinian or even contrary to neo-Darwinian thinking a la Dawkins.
I reiterate, evolutionary explanations in animal behaviour can call on comparative methods (within and between species), optimality modelling, experimental manipulation and actual measures of fitness. Evolutionary explanations of human morphology can at least call on the fossil record and comparative anatomy. EP cannot fulfil any of these criteria, it doesn’t even have very good characterisations of the putative cognitive modules it is ‘explaining’, so it is poorer than cognitive psychology in this respect.
PM
The trouble is that you’re talking as if there is no evidence at all for the “modular” theory of the mind (not that Pinker is happy with that expression).
But there is evidence. The Wason selection tasks, for example. The fact that autism selectively impairs people’s cognitive abilities. The fact that a subset of Williams syndrome – a subset associated with genetic difference – results in an inability to process complex 3-d arrangements, but has no other effect on cognitive ability.
As far as “just-so” stories are concerned. Positing putative functions for a cognitive/behavioural trait should be the starting point of EP analysis, not the end point (though, of course, as Pinker readily admits, with bad examples of EP, it can be the end point). But you can develop falsifiable hypotheses. (The classic example, I think – though it may not be the kind of EP that you’re talking about – is the difference in the rate at which parent’s murder their natural children, as opposed to their adoptive children (which holds across cultures, and which even arch critics of EP, like John Dupre, accept is quite striking).
As far as the Darwinian point, well I agree. I mean part of what distinguishes sociobiology from EP is the extent to which it results in a commitment to the idea that the brain comprises particular organs, etc.
But isn’t that the point, we know from various sources that the brain is to some extent ‘modular’ – but you can’t just go on to assume that anything you wish to call a module is one. I’d be impressed if you could find me an illness where the rape module is selectively impaired! The problem with this a priori assumption of cognitive modularity is that, like some cognitive neuropsychologists, EPers don’t show that the putative module actually is one first. And on the other hand, there are reading ‘modules’ in teh brain too, but that doesn’t mean they have been under selective pressure! I think Annette Karmiloff-Smith talks about this kind of thing.
The problem with arguments like the infanticide one (isn’t that pre-EP?) is that they are not testable, you come up with your hypothesis based on the evidence but can’t then go on to test much else because your hypothesis is essentially untestable – you can’t see if men who kill more kids have greater fitness like you might be able to in animals.
Also, these dissociations, where some disease/head injury/developmental disorder has relatively selective sparing of some cognitive tasks versus others are a lot more complex when you actually go and look at the literature – i.e. they are actually not normally as unimpaired in other tasks as is made out (I don’t know about the examples you give though), they are actually not all that impaired at the supposedly impaired task (i.e. its only a small effect) or there are good task specific reasons (one is harder than the other) for the difference. Let alone the complications developmental abnormalities with their perturbed developmental timecourse being used to infer adult mental modules.
Hey, while I’m monopolising this board, what do you mean by Wason selection tasks as evidence for modularity of mind? I know what it is but I don’t see what it tells us about modularity. I think, someone in EP, but I don’t remember who, claimed that some of the task facilitation you can get by changing the context proved some module or other exists but that would be very indirect evidence for modularity indeed, especially when we know that context helps -in general-, not just whatever context she(?) was using (cheater detection?).
Of course, again, even if this was evidence for some module in the brain -for- whatever, it wouldn’t be evidence for cognitive modularity in general
Last post for now – I promise.
Autism. Not exactly selective deficits here. Are we talking evidence for a theory of mind module?
Delayed and disordered language, impaired social interaction, abnormal responses to sensory stimuli, events and objects, poor eye contact, an insistence on sameness, an unusual capacity for rote memory, repetitive and stereotypic behaviour – not very selective at all.
And even if all this was down to say damage to the amygdala it doesn’t mean that the amygdala -does- theory of mind or whatever, just as an area of the brain lighting up in brain imaging studies doesn’t mean that part of the brain -does- whatever task you were using.
Its all too complicated to make these kinds of glib pronouncements, I might as well put all the theory of mind deficits in autism down to impaired facial emotion processing (amygdala damage) leading to a lack of early reward/punishment cues in early life and thus a failure to develop normal social interactions, rather than that the amygdala -does- theory of mind and that autistics -lack- theory of mind.
PM
Sorry, but I just don’t have time to keep up with all this.
I’ll try to deal with some of your points over the next 24 hours or so.
Very quickly – Wason selection task. Have a look at “So you think you’re logical” on http://www.philosophers.co.uk.
The point is that whilst giving some general context does increase scores, it doesn’t do so anywhere near to the extent that giving a cheating scenario context does.
I’m uneasy with your talk of “modules” since that tends to suggest distinct brain regions (and which your response about autism suggests you’re talking about), which isn’t what Pinker is on about. But I wonder how you would explain the Wason task/cheating scenario thing without recourse to this kind of theory. An interesting thing about it: how well you score is not influenced by your education level (given certain level of education, obviously).
I imagine degrees in philosophy and particularly psychology are a bit of a spoiler though ;-)
I think this kind of thing can be explained in myriad ways. Off the top of my head, how about cheating contexts being the most familiar/common contexts where we come across this kind of rule?
I think the mental modules/brain modules distinction is a very important one – I find talk of mental modules, as distinct from brain modules, a bit dualist and I’m not convinced it makes a lot of sense to posit selecting at the level of mental modules when EPers are allowed to avoid defining them or showing that they exist.
In support of my argument that it is due to familiarity; if we look at the figures the biking one is quite high and even the filing one versus circles and squares in the non-cheating context, whereas the really high scores are found in drinking and surfing in the cheating context versus tattoos and cassava.
Oh, the “modularity of mind” debate again… I think the best thing that can be said about “modularity of mind” in relation to evolutionary psychology is that it’s unnecessary; even a system without modular structure can be subjected to evolutionary pressure and change accordingly. We have a lot of models showing how the human mind COULD be structured into modules, and a lot more models that show how the same behavior can be obtained without formal modular separation. I can dig up references if anyone cares, some of them my own.
EP suffers from a similar problem; we have an excuse of how something COULD have arisen, but no real evidence to support the claim THAT it did. Possible explanations can be a good start for an inquiry, but they’re a terrible finish.
PM (and anybody else interested)
Just to be clear about what Pinker thinks. He doesn’t like the word “modular” because he doesn’t think that the “organs” of the mind correspond to a particular “blob like” location in the brain, with a specified number of inputs and outputs, but he does think that the organs of the mind are directly instantiated in the brain, and he expects that we will find their physical implementation when we know enough about the brain so that we can parse it in functional terms, “to track circuits, as opposed to blobs” as he put it to me.