Steve Pinker weighs in on the “evolutionary psychology is impossible” paper

May 19, 2020 • 1:30 pm

After I wrote my critique of Subrena Smith’s anti-evolutionary-psychology paper this morning—hers titled “Is evolutionary psychology possible?”—I sent the link to Steve Pinker, who’s quarantining on Cape Cod. He wrote back with some nice words of approbation, but added a few points. I thought these points were good, relevant and, as the first two weren’t made by either me or critic Edward Hagen, I asked Steve if I could quote him. He said yes. So add these criticisms to those leveled this morning. (Steve’s words are indented.)

All your points are exactly right. The motive seems to be the slipshod politicizing I exposed 18 years ago in The Blank Slate: if we’re blank slates, there can’t be differences between races, which would make racism impossible; therefore to combat racism we must believe that humans are blank slates. It fails both in philosophical coherence (racism is not an empirical hypothesis that might be shown to be true or false) and in accuracy — most evolutionary psychologists argue for a universal human nature. Also, a philosophical argument against evolutionary explanations in psychology ultimately falls apart when it unwittingly “refutes” even the most unexceptionable evolutionary explanations, such as sexual desire or protectiveness of children. And ironically, though this argument claims to be based in the philosophy of science, it seems unaware that within that field “argument to the best explanation” is generally considered the only means by which science of any kind is done — science never makes apodictic pronouncements based on a prior list of methodological precepts.

A couple of other observations [about Smith’s argument]:

—“Massive modularity” [JAC: a term Smith uses as an inherent part of “Evolutionary Psychology”] is not a term that was ever, to my knowledge, used by an evolutionary psychologist. As far as I know it was coined by that foe of natural selection, the late Jerry Fodor, as a term of abuse (and partly to protect the brand of Modularity, which Fodor introduced to philosophy and cognitive psychology in 1983). I’ve always argued that “module” is a misleading metaphor; we should be thinking about specialization. The brain isn’t spam, but the specialization of information-processing systems, like that of organ systems, needn’t be into encapsulated modules (Fodor’s definition); the systems have to talk to each other, share subroutines, pass information back and forth, and so on, just like the circulatory system and the digestive system do.

—This is a point in the fine print of EP, forgotten even by many of its practitioners, but the Environment of Evolutionary Adaptedness should not be confused with the Pleistocene or the hunter-gatherer lifestyle — it’s a time-weighted composite of selection pressures right up to the present (since natural selection never sleeps). We’ve spent more of our time as humans in a hunter-gatherer lifestyle, and, ironically for the PC police, the simplifying assumption that “EEA = Pleistocene” guarantees human biological equality, since we were all hunter-gatherers then, whereas different groups transitioned to agriculture and civilization at different times subsequent to then. And so if we continued to evolve since then, some groups could be more biologically adapted than others to literacy, settled life, governance, and so on. So the Pleistocene-EEA idealization is simple and convenient. But it’s far from necessary, and I think is an impediment, because it requires being more gratuitously committed to savannah-caveman scenarios. Many selection pressures are more or less unchanged since the dawn of the Holocene — paternity uncertainty, infant care, avoiding toxic foods, securing allies, etc. And for many others the most dramatic dividing line is not Pleistocene-Holocene or hunter-gatherer/agriculture, but urban, educated modernity (knowledge of science and statistics, professional employment, universal information, access to the legal system, modern medical care) versus every other lifestyle. An American 19th century farming village, or a favela or peasant society in the developing world, has pretty much the same face-to-face social network, with a pre-scientific knowledge base and a remoteness from legal and professional institutions, as a hunter-gatherer society 10K years ago. What’s key is avoiding the assumption that humanity is adapted to a modern urban or suburban college-educated lifestyle, not assuming that we have the same brains as hunter-gatherers (whether or not we in fact do).

—I think that one important direction of evolutionary psychology (and medicine, and biology in general) is to look for statistical signs of selection in the genome. This showed that the modern version of FOXP2, involved in language and speech, was a target of selection (though not on the recent rapid sweep originally hypothesized, probably earlier). But this method works best in the rare cases where single genes with large effects can be identified. Now that we know that virtually all behavioral variation is under the control of many genes with minuscule individual effects, I wonder whether there are techniques that can assess signs of selection across an entire vector of genes rather than individual genes.


31 thoughts on “Steve Pinker weighs in on the “evolutionary psychology is impossible” paper

  1. This stuff is all above my pay grade, but Pinker’s and PCC(E)’s points suggest that Smith’s assertion is somewhat shaky, and certainly not the killer blow to evolutionary psychology that she believes.

  2. The concept “module” is ok, but needs to be defined in a functional sense. So a “module” is “those aspects of the brain responsible for a given behavioural trait”.

    Of course modules are not localised regions of hardware, they very much overlap.

    {Also, goes running for a dictionary to look up “apodictic “.]

    1. I don’t see the value of using the word when it requires redefinition since the generally accepted usage includes the idea of “part” or “unit”. The word “function” would seem perfectly adequate when talking about brain stuff. We don’t understand the xxx function of brains but…

      1. But it can be “part” in a functional sense. Producing a given behavioural trait is a “part” of what the brain does.

        If we talk about “modules” that make up a degree course, or software “modules” that combine into a complex package, we’re not necessarily saying that they are physically localised and discrete. Rather, it’s just a way of thinking about and referring to parts from the point of view of what they contribute to the whole.

        1. Software modules are “parts”, discrete and bounded bits of code that get combined for some purpose. Similarly, degree courses are bounded “things” comprised of books, lectures, exams, and so forth.

          There’s no particular value in using the word here. All it does is confuse the matter because people in the conversation will justifiably think you’re talking about areas in a brain. And you end up arguing semantics. We get enough of that in discussions of free will. And as I said before, “function” works perfectly well without potential confusion.

          1. But the “module” of the brain is not the function, it’s the thing that has the function. So, for example, the “visual system” is the module or system that has the function “seeing”.

            We apply that systems/functional thinking to other parts of the body (“respiratory system”, “cardiovascular system”, “immune system”), and the term “module” is just adopting the same approach to the brain.

          2. OK, you don’t like “function”, use “system” instead. It, too, is perfectly reasonable and doesn’t lead to confusion because it isn’t a word being re-defined.

  3. >Now that we know that virtually all behavioral variation is under the control of many genes with minuscule individual effects, I wonder whether there are techniques that can assess signs of selection across an entire vector of genes rather than individual genes.

    Isn’t this what GWAS (Genome-wide association study) is for? And isn’t David Reich doing work related to that with the Allen Institute.

    And I believe a fair amount of work has been done with genetic changes around the time the transition to farming.

  4. A thought experiment would be to ask: If a huge disaster were to wipe most of us out, and the 21st century survivors were made to live in a new Stone Age without the ability to use metal or agriculture, would our species then go extinct, or would it survive?

    1. I would say survive, but that sort of depends on the disaster. Martin Rees (Templeton winner see some old WEIT pages) has written a couple of good books on that – if I recall he was not so confident but he explores various scenarios…

    2. Assuming the ideas of tools and of growing things didn’t miraculously disappear, some survivors should do well, and pretty rapidly redevelop a fairly secure lifestyle.

      It would help, of course, to be situated in an area with a mild climate and secure food resources — I’d bid on property at the mouth of a nice salmon stream in the northwest US/Canada. And acres of clams. And camas fields within walking distance would be nice. Of course there’d be war with neighbors with design on the same property, but a generation or two on, the inheritors would be fine.

      Unless you’re going really sci-fi-y, it’s hard to imagine going full paleolithic. We might not be competent to smelt metals, but there would be an endless sea of ferric and other metals that anyone with knowledge of the “before” would be able to forge and grind into a nice tool kit. And ancestral knowledge of agriculture would guide gardening of local [or feral] food plants.

      Further, you wouldn’t need to be a country-fair artisan to know that basketry and pottery and sewing and weaving are things, and find time to make your own.

      I wouldn’t count on this world being saner than ours. The demon-ridden human brain would come up with gods, witches, taboos every time.

    3. Likely perish. Different cultures are adapted to different environments, not through their genes necessarily, but through environmental info acquired and stored in their head (i.e., culture). Thus if modern humans were transplanted into a “stone age environment” we wouldn’t know how to get by. We know this to be true because many times in history it has happened. The first chapter of Joe Henrich’s amazing book “The Secret of our Success” documents many instances where explorers traverse an unfamiliar environment (e.g., Australian outback) and fail, while the culturally-adapted indigenous population thrives in the same environment…not due to genes, but due to the cultural accumulation of survival knowledge.

  5. I find the domestication of humans fascinating. That, to me, is what this is all about. Agriculture, civilisation, have between them had a huge effect on behaviour & seen the start of dense population levels which have turned us from grasshoppers into locusts. It forces behavioural changes leading to complex rules-based societies. I would say!

    1. Speaking of Locusts: the U.S.A. is supposed to be on the verge of emergence of 17 year cycle Cicadas. (Having been on the east coast during the 13 year Cicada emergence, I know firsthand that they are quite large and loud, and you don’t want them sharing a trailer with you. I considered trying to eat some, but wasn’t brave enough.) Nice addition to Covid-19. Which of the Biblical ten plagues is next? Should we be putting blood on our doorposts?

      Hope that our Cicada swarm isn’t as devastating as the recent Locust infestation in Africa.

      1. RE the cicadas. It would certainly be a “Pharaonic” plague [call of the dominant M. septidecim” is rendered as “phar-ooooh”. ] Complicating the matter, there are usually at least 2 species co-emerging in each brood, with different calls.

        Between the 17-year and 13-year Magicicadas, there could be 30 mostly reproductively isolated broods [isolated by year class, but most are also geographically isolated]. Currently 15 [12 of 17; only 3 of 13] of these exist, but the big ones are the Great Eastern Brood [17], which will emerge in 2021, and the Great Southern Brood [13], which emerged in 2011 and will again in 2024.

  6. Is the genetic analysis he suggests in the final paragraph a GWAS (genome wife association study)? If so, it already exists, and I know if at least one study using it to look at schizophrenia traits.

    1. GWAS exist, but they tend to identify statistically powerful enough loci (individual genes).

      Let me borrow from Wikipedia (since I haven’t studied it for a while):

      “Another trend has been towards the use of more narrowly defined phenotypes, such as blood lipids, proinsulin or similar biomarkers.”

      “A central point of debate on GWA studies has been that most of the SNP variations found by GWA studies are associated with only a small increased risk of the disease, and have only a small predictive value. The median odds ratio is 1.33 per risk-SNP, with only a few showing odds ratios above 3.0.[2][43] These magnitudes are considered small because they do not explain much of the heritable variation. This heritable variation is estimated from heritability studies based on monozygotic twins.[44] For example, it is known that 80-90% of variance in height can be explained by hereditary differences, but GWA studies only account for a minority of this variance.[44]”

      So here we are, but Pinker wants to do better. A laudable aim!

  7. An interesting response. Since a quick browsing the – to me new – concept of the Environment of Evolutionary Adaptedness seems to originate in psychoanalysis rather than psychology (e.g. reviews says earliest in Bowlby’s attachment theory) I did not think it was useful. Rather the idea to study signs of selection in the genome. (For an example I’ve seen, using wide window sizes so far, see “Impact of Mutation Rate and Selection at Linked Sites on DNA Variation across the Genomes of Humans and Other Homininae”, David Castellano, Adam Eyre-Walker, Kasper Munch, Genome Biology and Evolution, Volume 12, Issue 1, January 2020, Pages 3550–3561.)

    For what it’s worth, I’ve seen evolutionary psychology papers rely on “best explanation” correlations between observed data such as behavior and neural architecture rather than choosing any particular template coarse graining such as “modularity”. Much akin to genome study correlations (though admittedly GWAS tend to coarse grain too much in practice).

  8. I believe that in Skinner style behaviorism, there is a controversy over whether something like “zero trial learning” (I might have the name wrong, I can’t recall exactly and Google isn’t helping) can occur. The idea that a behavior really does just emerge spontaneously with no teaching.

    It seems to me that once you can establish that this takes place (and I think given the ubiquity of things like child development, you can, although that’s just my opinion – it seems highly unlikely that babies would hit the same milestones at the same time if these things were explicitly learned,) then you have to explain this one way or another, and evolution seems like the most likely candidate by far. I suppose you could propose other mechanisms, but if they don’t involve genes, it seems to me they would take on an almost mystic quality.

  9. I think one problem is that when folks attack EP, they are attacking some of the original literature and not keeping up with the changes to the field. Tooby/Cosmides wrote in 1995 “Our cognitive architecture resembles a confederation of hundreds or thousands of functionally dedicated computers designed to solve adaptive problems endemic to our hunter-gatherer ancestors.”

    It is not an irresponsible stretch to view this quote as being synonymous with “massive modularity” or that we are adapted to a Pleistocene (non-sedentary) way of life.

    John Bowlby, who came up with the EEA concept, talks specifically about a Pleistocene lifestyle: “…one that man inhabited for two million years until changes of the past few thousand years led to the extraordinary variety of habitats he occupies today.”

    Donald Symons also talks about how we are not currently under present day selection pressures, “it is generally agreed that insufficient time has elapsed since the invention of agriculture…for significant change to have occurred in human gene pools…Humans can thus be said to be genetically adapted to a hunting and gathering way of life.”

    And yet, EP as a field has moved away from these positions (somewhat) and/or modified them (e.g., we are still under strong selection for many traits), but researchers can still point to the above quotes to mount a critique.

Leave a Comment

Your email address will not be published. Required fields are marked *