I am a student of evolutionary biology, and doing undergrad at Melbourne Uni a couple of years ago, one of my lecturers was a bit of an enthusiast for something called "evolutionary psychology" (basically, the idea that human behavioral traits are evolutionary adaptations; e.g. "men like sleeping around because they invest less in their sperm than women do in their eggs"). It didn't strike me as having much scientific merit, and as a reader of Steven Jay Gould (a prominent evolutionary biologist who never liked EvPsych), i quickly discovered that the underwhelming impression EvPsych had on me was not just a result of being an ignorant student. Some professionals were also... less than whelmed. And yes, the more i read about it, the more confident i became that my own criticisms where not only valid, they had been pointed out by many professionals in related fields before i ever heard of it.

One of the texts that were set during our lectures was a book called "The Blank Slate: The Modern Denial of Human Nature", by Steven Pinker, who is apparently interested in language and cognition, and had written another book suggested by the same lecturer, "The Language Instinct", that was supposed to be good. I have had vague plans to read The Blank Slate for a while; the things it supposedly said have never accorded with my understanding of human nature, evolution, or science, and i had the impression that i needed to at least read the book, and be able to counter it's arguments, in order to have any confidence in my own beliefs. So this post was originally going to be a big, scientifically minded critique (or exploration) of The Blank Slate.

Well, I've gotten my hands on a copy, and i have to say, the book is as underwhelming as the field it defends. It's really not worth the great big chapter-by-chapter breakdown that i was considering, but there are some points i wanted to record for future reference so i don't have to read through it again. So, instead of a refined critique, this post will be a series of points about Pinker's book that i don't want to let slip away into my memory (you never know when they'll come in handy). I'll divide it into two lists:
*The writing itself,
*The science.

(Background info: For those who are not familiar with the book, The Blank Slate is based on Pinker's idea that three particular themes...
1."The Blank Slate" (that we all start off equal, mentally speaking),
2."The Noble Savage" (that underneath culture, people are good and idyllic),
3."The Ghost In The Machine" (that our consciousness is an abstract "soul", that mind and brain are distinct),
...have taken over science, the arts, and the humanities.

Pinker's book is devoted to explaining how, in his opinion:
1. All these ideas are wrong, because of:
- discrete modular mental traits existing,
- a genetic basis to behaviors and thoughts,
- the idea that human behaviors are the product of evolution (with culture, society, and volition coming secondarily, on top of our evolved minds)
2. All the people who disagree with Pinker are actually just deluded by various unwarranted philosophical and political fears (e.g. a fear of eugenics and racism),
And:
3. By ignoring the science Pinker advocates, we ceate dysfunctional societies, and both modernism and postmodernism suck as a result, bringing us crappy art that only snotty intellectuals will ever like.)

*straw men: Pinker's rhetoric depends very heavily on Straw Men (e.g. creating a caricature of what your oppponent says in order to more easily defeat them). The book is peppered with very poor summaries of possible alternate arguments. (This makes sense to an extent, as the entire thrust of the book is a negative argument against "the blank slate", "the noble savage" and "the ghost in the machine".) (Philosopher Simon Blackburn made a similar observation the centre of his aptly titled review of Pinker; "Meet The Flintstones").

*Anonymous friends: He quotes an anonymous friend on 'innate' gender differences: "Look I know that males and females are not identical. I see it in my kids, I see it in myself, I know about the research. I can't explain it, but when i read claims about sex differences, steam comes out of my ears." (p.351) Why did he not examine the actual arguments and debate regarding gender differences (e.g. Ruth Bleier's early critique is still valid)? Why the need for an anonymous quote? Is this anyones actual words, or is it just what Pinker imagines someone would say? Would someone with "steam coming out of their ears" over the issue really be so unaware of the debate? Is it possible that it is just Pinker that is unaware of any of the arguments against an innate reading of gender differences; so he invents the only position that he can concieve that an intelligent feminist scientist would take and then ascribes it to a "colleague"?

*Marginalising opponents views: Opponents views are treated as either discredited or accepted for unsound reasons (p.135), even though they take part in active debate on the issues within the fields he discusses. The only real debate, in Pinkers eyes, is between him and the others who largely agree with him, debating issues like what manner of adaptation "Art" represents (p.405). A fairly unconvincing attitude; and maybe a sign that he isn't trying to convert any opponents so much as appeal to a pop science audience.

*Inconsistent history: One moment, the evil believers in "the Blank Slate etc.." are taking over the world and dominating all intellectual pursuits. A few pages later, they are "obsolete", outdated or discredited (e.g. pages 133-135). He varies between these two positions in a way that doesn't make a lot of sense upon reflection.

*Disingenuous quoting: Possibly a result of Pinker failing to carefully read his sources. (Louis Menand's critique "What Comes Naturally" points out a substantial misquote and misintepretation of a line by Virginia Woolf, in which Pinker apparently (and seemingly without knowing it) reverses Woolf's meaning.) He only partly quotes Gould, and Rose, Lewontin and Kamin. A reading of the original sources shows that these writers have views which differ from what Pinker claims they believe. (Pages 126 and 128 provide two clear examples, when you read the original essays by Steven Jay Gould which he cites there.)

*Andrew Bolt-ian level of Arts scholarship: His problem with the Arts and Humanities seems to boil down to "I don't really get it, so it must be stoopid.", but he takes pains to write as if he is reflecting some wider consensus. His criticism of the humanities rests entirely on an appeal to the reader to fail to comprehend what its practitioners are doing. He never engages their ideas, he just points and says "Look, they think they're better than us!" He quotes a passage from Judith Butler, utterly out of context, to make fun of all the big scary words she uses (p.415). Her arguments and beliefs are not important. His critique might have more weight if he showed any sign of even wanting to understand that which he criticises; but he doesn't want to, and he probably doesn't need to: He isn't appealing to our reason, he's appealing to knee-jerk anti-intellectualism. (Louis Menand points out Pinkers' misidentification of Piss Christ as "a crucifix in a jar of urine" and writes: "Pinker thinks that modern art is all ideas because it is only as ideas that he can experience it." i.e. because he has not seen what he criticises, he's only read about it, and thought it sounded dumb.)

*Heritability: Pinker misunderstands (and misuses) the concept of "heritability". He defines it on p.374: "Heritability is the proportion of variance in a trait that correlates with genetic differences." and on p.377: "To say the heritability of intelligence is .5 ... implies only that half of the variation among people is inherited". This is close to what heritability means; it is a measure of the variation in some characteristic in a population that can be attributed to genetics. It is not a measure of the 'amount' of a trait that is genetic. A better definition of heritability would be to say that it is a measure of the degree to which breeding can alter a trait within a particular population at a particular time, ranging from 0 (100% of variation is environmental, and breeding programs fail) to 1 (100% of variation is genetic, and breeding programs are a rousing success).

Something can be completely genetically determined, and have a heritability of zero (e.g. if all the people in a population of humans who have missing toes lack their toes because of accidents, then 100% of the variation in number of toes is caused by the environment, and the presence or absence of toes has a heritability of zero). This point is explained quite clearly in one particular book, "Not In Our Genes", by Richard Lewontin, Leon Kamin and Steven Rose (1984). Pinker references this book (derisively), but appears unaware of it's intellectual content.

If a trait has a heritability of 0.5, it does not mean that the trait is 50% caused by genes and 50% caused by environment, it simply means that 50% of the variation in the trait in a population is determined by genes. The country next door might show a heritability of 0.2 for the same trait, even if they're substantially the same genetically. Their children might live under some oppressive regime where everybody is forced to act the same way, and they may find a heritability of 0.9, even if they're all genetic clones of their parents. To claim that a 0.5 heritability means that a trait is 50% "made up of" genetic effects is to make the mistake of jumping from a general empirical statement about a population-specific observation of the way a trait changes across individuals, to a cause-and-effect statement about the underlying reason for a trait in those individuals. Pinker says something vaguely similar (see the quote from page 377, above), but he doesn't seem to be able to apply this definition in practise.

For example; he uses "heritable" and "inherited" interchangeably. The paragraph at the bottom of page 50 is a representative example, wherein he refers to traits being "heritable" when he seems to mean "inherited". On page 373 he claims "all human behavioural traits are heritable". In the index on p.498, the entry "heritability" directs us to several pages where he discusses an "inherited-ness" to traits, and never mentions heritability. On page 379 he talks about manipulating "heritability estimates" as if they could be switched around between studies, in reality, heritability estimates are population (and situation) specific.

Pinker's confusion regarding the meaning of heritability becomes more serious when we apply it to one of his central arguments: that the mind is not a blank slate. Heritability is important to Pinker; he uses "heritability estimates" generated by behavioural geneticists as if they were measures of the degree to which a trait is a result of genes (the estimates themselves are dubious, but thats a seperate issue). And he uses this to claim that we are not blank slates, we are partly determined - partly written on - before we ever encounter the environmental influences that supposedly can only sway us the other 50% of the way.

But this is not what heritability means at all. If the heritability for IQ in a human population was 0.5, it would not mean that education, for example, can only aspire to an equal 0.5 effect to genes. Heritability could be 0.5 for some measured population at some given place and time, and education could still theoretically sway it so completely that the genetic effect may as well never have existed.

Like a blank slate.

Well, an eraseable slate, anyway. The simple fact is, heritability has nothing to do with the truth or falsity of a "blank slate" view of human nature, and Pinker's mistaken preoccupation with the concept does not support his thesis.

In fact, a discussion of the proper interpretation of "heritability" only highlights flaws in the basic assumption that "inherited-ness" itself even matters as much as Pinker claims: Maybe we do have innate "human natures" that fix our mental attributes at birth. And maybe nobody since the beginning of conscious society has ever actually expressed any of that genotype, except where it accidentally coincides with what they've decided/been told to/learned to do.

*Interactions: Pinker claims to support an "interactionist" position (e.g. p.127, p.379), where genes and environment interact with each other to produce behaviors. However, when he provides his breakdowns of the contribution of genes and environment to personality traits, he writes: "Genes 50 percent, Shared Environment 0 percent, Unique Environment 50 percent." (p380) There is no sign of an interaction term (e.g. "Genes X Environment"). This is not an isolated error, Pinkers' writing throughout the book seems to be based on a belief that one can quantify "Y Percent Genetic Influence and Z Percent Environmental Influence" for human behavioral traits. For all his claims to interactionism, he either doesn't fully grasp the concept, or he doesn't really believe it. Indeed, he seems to believe that "50% genes and 50% environment" is an interactionist statement, even though it contains no interaction term (and doesn't leave any room for one, either).

*Just-So Stories: One of the chief criticisms of EvPsych is the use of "Just-So stories" to explain how a behaviour arose, in evolutionary terms. Pinker frequently references Robert Trivers, a man roundly criticised for his use of this technique, however Pinker does not seem to recognise this criticism at all. He repeatedly makes statements to the effect that a given human behavioral trait is "exactly what would have been predicted from evolutionary psychology." The point that Pinker misses here is that any trait you can imagine would be "exactly what evpsych predicts". Even if you took a trait that was the exact opposite of something that evpsych predicts, it would still be something "which evpsych predicts", due to the flexible nature of a Just-So story. You can create a just-so story to explain almost any trait, and it will rest on nothing beyond its own internal logic, but you can't then claim that the existence of the trait is evidence for the validity of the story.

*An entirely hypothetical science: It is worth noting that evolutionary psychology has never, to my knowledge, demonstrated the existence of what it studies. Geochemists have rocks to study, cultural theorists have media to study, and ornithologists have birds to study. But noone has ever empirically shown that a given behavioral trait in humans is really a result of adaptive evolution. Unlike most (well, all...) other areas of science, evolutionary psychology is an entirely hypothetical discipline. It's a bit like cryptozoology; by definition, there's a lot of stuff for cryptozoologists to concentrate on (yetis, lake monsters, chupacabras), and you can do a lot of rigorous scientific speculation on these study subjects, but cryptozoology has never yet demonstrated the empirical existence of its units of study. Evolutionary psychology is in the same state. And this has an impact on books like The Blank Slate that argue in favour of EvPsych: All Pinker even tries to offer us as an explicit defense of EvPsych is that particular behaviors seem to resemble behaviours that have been found to be adaptive in other animals, or seem to resemble what game theory predicts would be the outcome of evolution on a behavioural trait. The problem is that the connection between the behaviour in humans and the behaviours in other animals is far from clear (and many evpsych assertions seem to violate the basic biological no-no of "anthropomorphism"), and when it comes to predictions of game theory in human behaviour, cultural, political and philosophical explanations for the behaviour work exactly as well as evolutionary ones (though they dont require the added assumptions of an evolutionary explanation, so they're really favoured by Occams Razor). To put it another way, the entire research agenda of evolutionary psychology basically boils down to "Wouldn't it be cool if...?"

*Reliance on anecdotal evidence: Anecdotal evidence is not given much weight in science. There are two key problems with anecdotes; 1) anecdotes are selected because they illustrate a point, so they are not an unbiased survey of the state of the world, and 2) anecdotes contain no controls whatsoever, and a range of conflicting and confounding influences on a situation may be bundled up in an anecdote - leaving the scientist with no ability to examine any of these influences. There are two spots in particular where Pinker falls back on anecdotal evidence. The first is in his discussion of twins. He happily describes twins with ridiculous levels of similarity to each other; he even writes: "Traits that are surprisingly specific turn out to be heritable [note the misuse of "heritable" again; he presumably means "inherited"] ... the identical twins seperated at birth who both grew up to be captains of their volunteer fire departments, who both twirled their necklaces when answering questions, and who both told the researcher picking them up that a wheel bearing in his car needed to be replaced." Many skeptics would start to hear alarm bells going off at such ludicrously precise examples, but Pinker's cognitive dissonance seems to have saved him from hearing those bells and thinking "maybe there's some 'selective reporting of coincidences' going on here...". Another place where an anecdote is treated as a genuine piece of scientific evidence is in the story of "Brenda/Bruce", the male child who was raised as a girl by his parents, and subsequently cracked it, declaring himself male after all. This is supposed to illustrate that boys and girls are innately different, their mindsets are determined by their underlying biology. But there are a million ways the case could have been unrepresentative of normal human behaviour (heck, maybe Brenda's parents were completely successful in raising him "as a girl": they just produced a transexual girl?), there are a million ways the the parents might not have even treated him as "a girl" (but, for example, as "a boy pretending to be a girl"), though Pinker doesnt seem aware of any such considerations at all.

*Misunderstanding of non-genetic influences on behaviour: Pinker seems unaware that there is more to psychology than "we all just passively soak up our roles from our environment" (eg. p.172). In fact, queer theorists speak about individuals actively pursuing identity, and psychologists tell us "It is not just our experiences that define us, but how we choose to interpret them" (psychologist Dorothy Rowe, paraphrased from memory). Most of the time, he refers to all non-genetic influences as "environment", collapsing and conflating, into "environment", a range of influences, from free will through to socioeconomic class. This has been accepted shorthand within the debate over the existence of genetically determined behaviours, but Pinker seems to turn it into a misunderstanding of the sheer variety of factors "environment" may cover, and this has an impact on his style of argument...

When discussing how genes and environment may interact, Pinker uses the example of attractive people being more assertive. Genes made them look a certain way, and this impacted their behaviour. But, he says, its not just genes for attractive bodies that cause assertiveness, because not all attractive people are assertive. Therefore, there must be another set of genes relating more directly to assertive personalities... The argument seems to be this: Create an "either/or" between two alternatives - "that genes control behavior or, environmental factor X controls behaviour". Environmental factor X clearly does not completely determine behavior, therefore, genes must control the rest. The problem is that Factor X is different every time. One moment, he's showing that good looks do not completely determine behaviour, so genes must do the rest. The next, he's showing that parental "personality type" does not completely determine behavior, so genes must do the rest. The next, he shows how sibling's birth order does not completely determine behaviour, so genes must do the rest. It is as if he feels that all te different environmental factors can just stand in for each other, as they're all "environment". If Pinker ever actually confronted all his Factor X's at once, he would find that it's almost impossible to make any useful statement at all about the effect of genes on human behavior, due to the massive range of confounding variables. His method of argument reminds one of a bad action movie where a single hero is able to fight off a group of thirty villains one at a time, because for some reason they all decide to line up in single file to get knocked out.

*Shifting the burden of proof: A scientific hypothesis must be tested appropriately; scientists are supposed to use appropriate control groups and avoid 'confounding variables' in their research. Studies of twins try to tease out genetic influences on behavior by finding twins that were raised apart from each other, or by comparing identical twins to fraternal (non-identical) twins. The most powerful criticism of twin studies is that they never actually manage to remove environmental effects. Identical twins, for example, actually experience different environments to fraternal twins - so difference between them may not be due to genes, but due to these environmental differences; and twins raised apart from each other are generally raised within their own extended family, and often have contact with each other for a significant period of their life. Pinker references these criticisms off-handedly, writing: "Skeptics have offered alternative explanations that try to push the effects of genes to zero" (p.47). What Pinker misses is that these are not merely "alternative explanations", they are confounding variables. Describing them as "alternative explanations" leads him into claiming that they are hypotheses that must be tested to be considered at all (and this is exactly what he goes on to assert); shifting the burden of proof away from those who claim to have established a genetic basis for behavior, and on to those who spot the "alternative explanations". But they are, in fact, not so much "alternative explanations" as flaws in the way in which the experiments have been designed.

*Shifting ground: On page 376, he writes about "placental factors" potentially confounding twin studies (he seems unaware of many other factors; in fact, placental factors are irrelevant to the most serious criticisms of twin behaviour research); and he says: "In any case, studies comparing adoptees with biological siblings don't look at twins at all, and they come to the same conclusions as the twin studies." Two pages later (p.380), he briefly acknowledges some of the confounding effects in adoption studies, and then writes: "And even if adoptive parents are unrepresentative in some other way, [these assertions about human nature] would survive because it emerges from large studies of twins as well." It's not the only place he uses this logic. On page 379 he writes, "the sheer consistency of the outcome across three completely different methods ... emboldens one to conclude that the pattern is real." The question is, what if they were all confounded, Pinker? Consistent results in bad science is still bad science. Indeed, consistent results would be expected if genes keep being confounded by the melange of factors Pinker refers to as 'environment', because the effects of 'environment' would confound them all in the same ways. Also, the fact that the same researchers using similarly poor methods produce such results is at least suspicious, because it is these researchers that pick and choose which results to make public and to highlight in their papers (and Pinker himself does a lot of picking-and-choosing of results).

*Failure to confront criticisms of the evidence: I noticed this most clearly in the chapter on "Gender", only because i became interested in the gender difference debate a couple of years back (it makes me wonder what other errors i'm missing because i'm less familiar with the debates he trots over in other parts of the book?). Here, Pinker takes us on a whirlwind tour of "gender differences", listing "a dozen kinds of evidence", many of which have been seriously criticised, some of which have been discredited, a couple of which depend on fairly ludicrous pieces of anecdotal evidence couched in rather faulty chains of logic (e.g. users of testosterone shots claim that it gives them a massive rush, therefore this must just be a sign of what testosterone does normally in men, therefore testosterone makes men and women different). In particular, the criticisms of studies of children with hormonal imbalances, treated in depth by Ruth Bleier in her work, "Science & Gender", pass directly beneath Pinker's radar. It seems likely that Pinker did no research beyond finding sources that say what he wants to say; the most extensive use he makes of opponent's published views (e.g. essays by Stephen Jay Gould, the aforementioned book by Lewontin, Rose & Kamin) is in a shallow critique of their political stances and a psychoanalysis of the "real" reasons for their opposition to what he promotes. (And as i've pointed out earlier, if he had read the Lewontin, Rose & Kamin book more carefully, he might not have made so many technical mistakes with "heritability".)

*Publicity of the scientific debate: On page 372, Pinker complains about the lack of attention paid to the topics of "The Blank Slate" by professionals. For some reason, most of the professional community, it seems, have failed to grasp the basic truths that are so obvious to an enlightened mind like Pinker's. In his words: "...most psychologists have not come to grips with them, and most intellectuals do not understand them, even when they have been explained in the cover stories of newsmagazines."

Even when they're explained in the cover stories of news magazines.

Good grief.

Now seems the right time to quote Beryl Lieff Benderly (quoted on p.17 of "The Bright Side of Human Nature", by Alfie Kohn)...

'The skeptics' ideas, though, made singularly poor copy. "Re-analysis Faults Finding of Differences" or "Painstaking Attempt Fails to Replicate Influential Study" headlined no late city editions. "What We Still Don't Know About The Brain" and "Faulty Assumptions Weaken Chain of Evolutionary Reasoning" enthralled the readers of no slick magazines. "Scientists Urge Caution In Interpreting Statistical Data" blared from no Sunday supplements.'