I was recently pointed to this post by Edward Clint which purports to show Rebecca Watson using the 5 tactics of science denialism during her talk “How Girls Evolved to Shop” which was critical of evolutionary psychology at Skepticon.
I watched her talk, found it entertaining, informative, wondered why I haven’t been invited to Skepticon, and I found I agreed with many of her examples of really bad pop psychology nonsense that’s filtered into the media through both scientists, press-release journalism, and marketing disguised as science. In particular the “pink is for girls” idiocy, which when I wrote about it at the time I came to the same conclusions as Watson that it was a stupid interpretation of the data, and the researcher who was actually promoting this glib, incorrect, and historically-bogus interpretation was a fool. It was unusual in that it was an example of the scientist herself, not even the media, disastrously misinterpreting the data to make it meld with a specific societal bias about females.
The problem with this talk was that Watson used specific examples, especially those made prominent by the media, as indicative of the entire field of evolutionary psychology, and thus may have over-generalized about the field as a whole. Even though at the end when asked if there are any good evolutionary biology papers, she suggests there likely are but that’s not what makes it into the media because they’re probably boring (lies are often more entertaining), it was too late. The thrust of her talk probably was too one-sided, and suggested the nonsense that idiot journalists latch onto, and some of the more oddball researchers are indicative of an entire field, which is unfair. Edward Clint takes this as a sign of science denialism, however, and tries to fit the 5 tactics to her talk. While I agree that Watson may have over-generalized, this isn’t denialism. Let’s go over his points and discuss why I don’t think her talk crosses this line.
The denialism brought to Skepticon was to the field of evolutionary psychology, a thriving social science with roots going back to Charles Darwin himself. The critic was internet pundit and self-described feminist and skeptic Rebecca Watson. Watson is known for her blog website, as co-host of a popular skeptic podcast, and for speaking at secular and skeptic conferences. But Watson holds no scientific training or experience. The charge of science denialism is a serious one, and I will support the claim with a preponderance of evidence.
Ok, first of all, you don’t need to be a scientist or an expert in a particular field to be critical of it. At no point does Watson suggest she’s an expert, which would have been the only reason why such a critique is relevant. A layperson is perfectly entitled to research a field, and then give a talk such as this critical of a systematic bias towards women present in the field. I think she actually makes a compelling argument that there is a bias problem in the interpretation of the data coming out of these papers, and a big PR problem for evolutionary psych in that it’s especially the biased, stupid, and inane studies the media latches onto and amplifies for lay consumption. She doesn’t say it exactly like that, but that’s how I interpreted her talk.
The main points Watson wants to drive home are that evolutionary psychology isn’t science (as indicated by the quotes in the subtitle), and that researchers involved in it work deliberately to reinforce stereotypes and to oppress women. Watson frequently makes overly broad claims about the “they” or “it” of evolutionary psychology without further specificity, leading her audience to assume she simply refers to the entirety of the field, or to a large majority of it.
This is an unfair evaluation of her talk. I don’t think at any point Watson indicates this behavior is deliberate, malicious, or dishonest. It’s clear that she’s exposing a systematic bias in the interpretation of the data from these studies. She is not suggesting fabrication, tweaking, or dishonesty, just stupid conclusions, and flawed study designs, and I agree with her that in these examples, she makes the case, these particular researchers are either idiots or blind to bias.
Now we may ask, how would an (apparently) expert skeptic investigate the domain of evolutionary psychology to reach and support the conclusions that Watson has? The first step should be having a firm grasp on what evolutionary psychology is, and to have a working familiarity with the subject. Since we are talking about a scientific field, this at least would mean reading some papers, or maybe at a minimum, some scholarly reviews and meta-analyses. And they should be typical of the field, meaning from reputable journals and mainstream researchers. It would be silly to call biologists creationists and religiously motivated while pointing to Michael Behe and Francis Collins as examples of biologists as a whole.
As far as Watson’s over-generalization of her findings to the field I agree with this criticism, however, my interpretation of the talk as a whole was about how when it came to ascribing differences in behavior due to sex that evolutionary psych has some big problems with systematic bias towards affirming societal stereotypes about women. I think she makes a compelling case for this, but it is possible, of course, that the cases she listed are the glaring exceptions. Clearly with regard to Kanazawa, the guy is a crackpot, but she also had some pretty deadly critiques of other more legitimate researcher’s conclusions.
However, Watson seems to have only the most superficial understanding of evolutionary psychology and it isn’t clear that she’s read even one paper in the field.
This is unfair and disproven by the talk in which she provides specific critiques and interpretations of data where they conflict with the author’s conclusions. It’s very hard to do this without reading the paper.
There are many reasons to think this. She cited no sources during her 48-minute talk beyond what is mentioned in newspapers and other media or publicly available abstracts. While she derided media distortion in one part of the talk, she implicitly trusted media reports for the bulk of it, and rather uncritically.
I don’t understand this because it’s clear from the video that her slides actually have several of the papers up and clearly visible. I also don’t think she blindly trusted media reports either, as she cites specific instances, like the “pink is for girls” study, in which the media cooverage, and the author’s own conclusions differed from the data.
At the end of her talk, an audience member asks Watson if there is any “good evolutionary psychology”. Watson throws up her hands while saying “prooobably? I’m guessing yes, but it’s so boring.. because you can only make it interesting if you make up everything. [...] if there is good evolutionary psychology, it’s not in the media[...]” (see index 47:30)
Setting aside the striking anti-science attitude that only media-hypable science can be interesting, as well as the jarring ignorance that a scientific field composed of thousands of researchers working for decades and publishing in numerous reputable science journals only “probably” has some good work being done, Watson clearly reveals that she is only familiar with evolutionary psychology in the “media,” having moments before shown incontrovertibly how unreliable the media is.
I don’t think she expresses the attitude that media hype is only sign of interesting science. I think her talk should have been narrowed, however, to specifically address how evolutionary psychology has major bias problems when it attempts to explain differences between male and female behavior.
The first work she mentions in her talk is important because it sets the tone and is, presumably, important to her thesis that evolutionary psychology is pseudoscientific and sexist. She cites a Telegraph article referring to a study done by one Dr. David Holmes about the psychology of shopping. However, this is an unpublished, non-peer-reviewed study conducted by a non-evolutionary psychologist paid for by a business to help them sell things better. This has no relevance to Watson’s thesis, unless it’s also true that Colgate’s “9 out of 10 dentists recommend you give us your toothpaste money” studies prove that dental science is bunk.
Again this is an unfair criticism, because she specifically addressed that this was marketing disguised as science. Watson states:
“all of the best studies I think are commissioned by shopping centers, so no this is actually marketing disguised as science, which is a trend that is becoming more and more popular as mainstream new outlets phase out any and all support for actual journalists that understand science.”
The strength of her point was how she moved from the obvious, BS, marketing-driven science and compared it directly to actual academic evolutionary psych purporting to show the exact same thing.
Supporting the extraordinary claims that a large scientific domain is sexist in general and methodologically bereft requires extraordinary evidence. It should entail a very serious, careful look at the nuts and bolts. How is peer-review accomplished? How well does it function? Are many awful studies passing it? How many? How easily? How is it that thousands of people, women and men, in dozens of countries across decades of time are all morally compromised in the same way? Did she speak to even one person who actually does evolutionary psychology?
I agree with Clint here that she needs more evidence before she castigates the entire field, however, I do think that she makes a compelling argument that (1) evolutionary psych has issues with injecting societal bias towards women into its conclusions – and this is actually not an extraordinary idea given the long history of psych and bias towards women, non-whites, immigrants etc (I would suggest a read of “Mismeasure of Man”) . If it been completely eradicated, I’d be shocked. Her failing was she generalized this flaw to evolutionary psych as a whole, and not just this subset of papers dealing with sex differences in behavior in which the findings always seem to conform with the most recent societal biases. (2) I think she shows, and this is not in dispute, that findings which reinforce a stereotype about women are widely circulated in a credulous media, and this is harmful.
Finally, let’s address Clint’s critiques that this actually represents the 5 tactics.
In 2007 Scienceblogs writer Mark Hoofnagle wrote an oft-cited essay about 5 general tactics used by denialists to sow confusion. John Cook distilled these a bit for an article in 2010 which discusses climate science denial.
It is useful to cite Hoofnagle here because Rebecca Watson demonstrates all five of these in a single presentation and because climate science and evolutionary psychology have a lot in common.
Watson’s denialist tactics
1. Conspiracy theories
Watson frequently spoke of a shadowy, diffuse “they” of evolutionary psychology. When she cited researchers by name, they were held as examples of the they, and not distinguished as a subclass. She also often spoke to their devious, immoral intentions. Not just that they’re mistaken about their claim or that their method is flawed, but that they actively wanted to oppress women and reinforce harmful stereotypes. Thousands of people in dozens of countries, women and men all working together toward goals such as defending rape as “natural” and therefore good (see video indices 20:07, 22:43, 23:41, 35:40, 36:08, 38:40). No evidence was presented which could establish these ulterior motives in such a large group, and as I shall explain, they are entirely false. Mark Hoofnagle wrote the following on Scienceblogs about conspiracy theories; not Watson’s, but his words fit equally well here:
[...] But how could it be possible, for instance, for every nearly every scientist in a field be working together to promote a falsehood? People who believe this is possible simply have no practical understanding of how science works as a discipline.
The problem with Clint’s analysis is that at no point does Watson ascribe conspiratorial behavior to these scientists typical of a denialist argument. I think she’s ascribing a systematic bias towards women, and given the issues that science has had in the past with systematic bias towards less-valued groups in society, this is not either out of the realm of possibility or even surprising that it’s still persistent in psychology. This is where a reading of SJ Gould’s “Mismeasure of Man” would come in handy to understand how these biases are propagated. What was amazing was how Gould, in his description of the science behind alleged-differences in races, showed that the researchers weren’t fabricating or being outright deceptive, but were led by bias into over-interpreting data, throwing out inconsistent data, and methodological errors that would affirm their prior conclusions. Conspiracy in science is frankly absurd, but bias in science is a constant struggle, and one should, if anything, suspect its presence until proven otherwise. Contrast this to the global warming conspiracism of cranks such as Inhofe, who describe the entire field as a “hoax”, which suggests active deception for an alterior motive.
Denialist conspiracy theories are non-parsimonious. That is they raise more questions than they answer, because they’re generally being used to explain the absence of data, rather than fit together existing data into an explanation of reality. This is why it’s so absurd when denialists talk about actual conspiracies, like criminal conspiracies, or the assassination of Abraham Lincoln. Those are not “conspiracy theories” in the modern parlance, because they provide an explanation that fits the data, the results of investigation, motives, etc. They don’t create more questions, like, “how could all those thousands of people keep quiet.” The answer is they can’t. Just ask Lance Armstrong, the tobacco companies, or any gangster that’s had their operation undone by a snitch. Secrets are pretty hard to keep.
Watson is not proposing a non-parsimonious conspiracy theory here, instead she’s demonstrating examples in which authors are clearly overinterpreting their data to conform to societal assumptions about women. This is far from an extraordinary claim about psychology, it’s been demonstrated in the past, and is something psychologists should be on constant guard against, because it is more likely than not that at some point bias will enter their interpretation of data. Watson’s case is pretty solid, in regards to these examples, that the bias is plain to see.
Fake experts are not featured prominently in Watson’s talk. However, at the end Watson cites several fake experts whose opinions on the science are inconsistent with established, uncontroversial knowledge. She implores the audience to read Cordelia Fine’s Delusions of Gender, a book seeking to justify a radical social constructionist view of gender differences. While Fine makes some reasonable points about some flawed studies, scholarly reviews have criticized Fine for cherry-picking studies as examples which are amenable to her conclusion and ignoring the rest
Watson goes on to suggest Greg Laden’s blog. Laden is a bioanthropologist who is on record uttering unscientific opinions such as that men are testosterone-damaged women.
Clint acknowledges these examples are weak, and in particular picking on Greg is really just a smear. I think it’s hard to interpret his post on “men as testosterone-damaged women” as serious, as he himself says:
e. Or whatever. Other people were more thoughtful about it and objected to the statement because it is wrong. Well, that’s good, because it is in a way wrong, because it is an oversimplification. But it was not meant to be a description of the biological and cultural processes associated with the development of individual personality, culture, and society. I am a little surprised that people thought it was such a statement, because it is so obviously a remark designed to poke certain men in the eye.
It was a shock-statement, not a serious statement of scientific fact, and it’s unfair of Clint to be dismissive of Laden over such a triviality. Only the MRAs seem to take that statement seriously, and they, as a group, should be ignored whenever possible. As far as Cordelia Fine, I have a great deal of trouble speaking with any confidence on her position in the field as a non-expert myself. However, reading Diane Halpern’s review in Science (no denialist rag) I find it to be more-nuanced that Clint’s quote suggested. Halpern writes:
Cleverly written with engaging prose, Delusions of Gender and Brain Storm contain enough citations and end notes to signal that they are also serious academic books. Fine and Jordan-Young ferret out exaggerated, unreplicated claims and other silliness regarding research on sex differences. The books are strongest in exposing research conclusions that are closer to fiction than science. They are weakest in failing to also point out differences that are supported by a body of carefully conducted and well-replicated research.
I think a book described by an expert reviewer as a “serious academic book” but flawed in one regard shouldn’t be so easily dismissed, as this reviewer in Science, while critical, was mostly positive about her book. I think the fake expert moniker should not be applied to either of these two, and frankly, considering true fake experts out there like Monckton, the assertion is somewhat laughable.
3. Cherry picking
As outlined in part II, Watson restricted her citations to stories that appear in the general media and critical popular science books. She focused on some of the worst possible examples that could be found, such as the interviews (not publications) with the disgraced Satoshi Kanazawa, instead of focusing on mainstream, reputable researchers. She also limited her citations to the sub-topic of sex and gender differences. While it is understandable that she may choose a narrow topic to present to a conference, she frequently makes her claims about the field in general, not merely as it pertains to sex and gender differences. For example, she rehashes Stephen Jay Gould’s “just so stories” criticism, (long debunked by biologists and others), but then uses as examples only sex and gender claims.
Now here I agree with Clint, Watson should have limited her remarks to evolutionary psych and the “sub-topic” of sex gender differences, as it’s clear that there is more to evolutionary psych than this idiotic “girls like pink” crap. But I’m also going to disagree with him that Stephen Jay Gould’s criticism has been “debunked” based on his provided link I actually agree more with Gould than I do the author. While Gould was clearly proven wrong in a few instances, I think his criticism of “just-so stories” is actually quite-compelling, and is an attempt to try to avoid a biased understanding of evolutionary mechanisms to try to find a purpose to every behavior, or every evolutionary modification. This criticism reads truer to me than many of the post-hoc explanations I’ve seen in evolutionary biology, and if anything should be internalized by researchers in this field. To reject the possibility that one is telling a “just-so story” without adequate evidence is to reject the null hypothesis prematurely. While it is clear from the essay that this evolutionary psych can have its hypotheses tested, and even that Gould was wrong in one instance, doesn’t mean that it’s a tendency in the field and one that needs to be addressed.
4. Impossible expectations of what research can deliver
Some of Watson’s criticisms would un-make many sciences were we to take them seriously. For example she says (13:27) “they never tell us what genes” as if this is a grand indictment of evolutionary psychology. There are scientists making in-roads in this area, but tracing the path from genes to structures to behavior is difficult-to-impossible, except in the case of disease and disorder. Further, we certainly don’t hold any other sciences to that standard, even the ones for which genes and adaptation are critical. Does anyone know precisely which genes make a cheetah fast, and exactly how they accomplish that? The peacock’s feathers, the fish’s gills? Shall we toss out all the evolutionary biology for which we do not have genetic bases identified? I should think not. Cognitive science also focuses on models divorced from physical stuff like genes and even neurons, but no one doubts that genes and neurons make cognitive capabilities possible (which is why genetic illnesses can severely impact them).
While it’s true that it would be unreasonable to posit a genetic explanation for each trait since so many traits are polygenic, and we have a very incomplete understanding of the function of much of the genome, this criticism shouldn’t be dismissed so easily. Eventually this field will have to incorporate genome-wide analysis into our understanding of human behavior, although Clint is right, not every finding in biology that’s important or worth publishing about needs to be explained down to the last atom.
At 15:41 Watson derisively explained her view of the method of evolutionary psychology as picking a behavior, assuming it is evolved, and then find “anything” in the past that might be relevant to it. Setting aside the inaccuracy of her summary, she seemed to be balking that such an hypothesis is just totally made up. Yes, Ms. Watson, it is. That is how science works. It is not known what the answers are before starting, so a researcher makes as good a guess as they can and then tests it.
Yes, but the real criticism here is the absence of testing the null hypothesis, as I explained above. This should be a critical component of hypothesis testing. She also has a point that if there are too many explanations for the data, all of them consistent, the finding isn’t of particular value.
At 13:39 Watson says that we can’t know enough about the distant past to make assessments of what might have been adaptive. She refers to variation in climate and “environment” and that the lives of our ancestors also “varied”. In other words, evolutionary psychologists can’t make any assumptions. We can’t assume women got pregnant and men didn’t, or that predators needed to be avoided, or that sustenance needed to be secured through hunting or foraging; these are real assumptions evolutionary psychologists use. If we were to toss out evolutionary psychology for this reason, we must also toss out much of biology, archaeology as well as paleoanthropology. Much care must be used in deciding what can and can’t be assumed about the past, but archaeologists, paleoanthropologists, biologists and evolutionary psychologists know this quite well.
This is a valid point.
Last but not least:
5. Misrepresentations and logical fallacies
Please see section V. 25 False and misleading statements made by Watson. In that list, items 1, 4, 7, 8, 11, 12, 18, 19, 20 and 25 are misleading statements. This is not a comprehensive list. Watson makes liberal use of logical fallacies. I will describe just one for the sake of brevity.
The naturalistic fallacy. One can hardly find a more pristine example of this fallacy than in criticism of evolutionary psychology, and Watson’s remarks were no exception. She spelled it out clearly at 38:30 “men evolved to rape… it was used as a well it’s natural for men to rape”. The problem to Watson is that some evolutionary psychologists study the phenomena of rape as a potential adaptation, or a product of adaptations such as the use of violence to obtain what one wants. Watson assumes that if rape is about sex, and sex is good because sex is natural, then rape must be natural and therefore good. This is an absurdity of course; it’s every shade of wrong from the rainbow of ultimate wrongness.
Yes, but Watson was describing it as a natural fallacy herself! You two are actually agreeing with each other.
I also think that his list of false or misleading claims by Watson is worth reading and it really should have been the starting point for the discussion about Watson’s talk. They actually have a lot of common ground between them, and frankly evolutionary psych needs a wake up call to its public image problem. Instead Clint clumsily tries to fit the tactics of denialism to her talk, and in my opinion, fails. Yes there are problems here, and he raises valid points. But the presence of denialism is not one of them.