Re-read review: Evolution’s Rainbow, by Joan Roughgarden
This is a popular-science book that is both intensively researched and intensely personal. In 1997, the author began her transition to living as a woman, not knowing if she’d be able to keep her job as an evolutionary biology professor. She did – though she was asked to step down from administrative work – and in 2003 published the first edition of this book, and then a second edition in 2013. It explores variation in gender and sexuality in human and non-human organisms, and muses on the role of diversity as a whole in nature and evolution. I re-read this book trying to decide whether to recommend this book to students, since I recalled both parts I really liked and parts I thought were a bit iffy. That impression persisted on this second reading, and I’ll do my best to outline the pluses and minuses below.
Part 1 – Animal Rainbows: Very good (with a few minor caveats)
The first part of the book is essentially a catalog of what was known at the time of writing about variation in sexual behavior, male and female social roles, and family structure in animals. Given that Dr. Roughgarden’s professional background is the evolutionary biology of animals – particularly marine organisms – it is unsurprising that this is the strongest part of the book. She also presents an argument for why we should stop using the term “sexual selection” (coined by Darwin) and start talking about “social selection” instead. It is an interesting proposal, which I’ll discuss more in a minute.
Roughgarden notes that sex cells (sperm and egg) represent one of the few true binaries in nature. Few organisms have only one type of sex cell, most have two, and none seem to have more than that. They are produced by a wide array of bodies. Those that produce only eggs we call females, those that produce only sperm we call males…but many species produce both at the same time1, and some – like clownfish - can switch between being male or female. Males are not always bigger, more colorful, or more aggressive than females and they can not only play active roles in childcare but some (eg. the seahorse) even give birth. Females don’t always prefer stronger or more aggressive males, either; in species where both parents are involved in rearing young, factors that suggest a male will be effective at nest-building or nurturing can be very important. One fun example of sex role variation comes from white-throated sparrows: white striped individuals of both sexes are more aggressive and territorial, tan-striped ones are less aggressive but provide better parental care. The optimum team for rearing chicks is white + tan, but it doesn’t matter which sex fills which role! Similarly, most hummingbird species include “masculine” females with a colorful throat patch and “feminine” males (without). In at least one, instead of looking for a breeding territory where flower resources are defended by a male, masculine female hummingbirds defend their own territories.
Just as there are many ways for a biological sex and body type to be arranged, there are many different animal family types, from pairs to polygamous groupings to extended families where older children or other relatives help with rearing the young. Socially monogamous pairs – like swallows – can also engage in “extra-pair copulations”. This is often described in terms of cheating or even theft, but Roughgarden points out that animals may not have the same sense of private property as those human writers, and we certainly can’t assume they apply that thinking to each other. This seems to be borne out by swallow society, where most interactions occur out in the open yet swallows don’t seem to react much to their mates’ infidelity. Also, some of those extra-pair copulations are between males.
This of course leads into observations of same-sex sexual behavior in nature, which are a lot more common than most people think! A few examples include all-female populations of lizards that don’t need males to reproduce, but still go through courtship and mating with each other; female squirrels who clearly DID mate with males but who team up to raise the babies, and seal that partnership with sex; the 10% or so of male domesticated sheep that only court each other; bighorn sheep, where most males have sex with each other as well as with females and those that DON’T are the ones labeled “feminine” (because they pee like girls, apparently); and of course our cousins the bonobos, the bisexual hippies of the ape world, who use sex (very effectively) for conflict resolution. I agree with Roughgarden that it is kind of painful and silly the way some scholars have bent over backward to argue that this sort of thing only occurs because animals are too dumb to tell males from females, and I like that she highlights the social roles that sex can have apart from reproduction.
I wish she had talked about the inclusive fitness hypotheses for this type of behavior here, though. Well, it does get brought up in the human section of the book (see below), but only dismissively. “Inclusive fitness” is based on the fact that your genes are not just in your body; there are copies of the same gene variants in your relatives, too. Thus, if your relative has high fitness (high survival and reproduction leading to more copies of their genes in the next generation) YOU effectively have higher fitness too. This concept is used to explain phenomena ranging from altruistic behavior (helping another at a potential fitness cost to oneself) to menopause. In the context of sexual behavior, if there are genetic variants that increase the probability of exclusive same-sex attraction - which would usually reduce individual-level fitness - they may not be strongly selected against if A) those same alleles in the opposite sex boost fitness or B) gay individuals are more likely to help their relatives rear offspring. Thus, the genetic variants persist in the population. I understand the complaint – voiced in a quote from a gay scientist, if I remember correctly – that gay people are not worker bees devoted to raising relative’s offspring…but they don’t have to be for this phenomenon to help explain why exclusive homosexuality might exist at low but persistent levels. Humans have historically often lived in extended families; those who, for whatever reason, never married or had kids (or whose kids are grown up, as in the "grandmother hypothesis" for menopause) often volunteer or get called upon for babysitting duties and – even if they live separately – may contribute resources to the family. That’s all that has to happen for the “passing on genes by proxy” to work on a population level! It doesn’t mean that one is obligated to do this to prove one’s worth any more than a male-female couple is obligated to pop out as many kids as possible.
1. Such animals are officially, and in this book, called “hermaphrodites.” However, some intersex people dislike this as that term has been used as an insult toward them. We can’t call these animals “intersex” because that already means something biological (eg. like the female deer that grow antlers). In botany, flowers that have both male and female parts are referred to as “perfect” or “bisexual” flowers… which I kind of enjoy, but which is also potentially confusing (and unnecessarily insulting toward single-sex flowers). Probably something like “bi-gametic” would get the point across better.
The thing I have the biggest argument with is the way Dr. Roughgarden uses the term “gender”. In attempting to expand the concept beyond humans, I think she muddies the waters of an already contentious topic in a way that could (ironically) be inadvertently harmful to trans/non-binary people. For example, she refers to fish that have one type of female but three types of males with different appearances and mating strategies as having four genders. However, those differences seem to be entirely biological/genetic…which is the basis on which transphobes are currently arguing that trans women are actually men or trans men are confused lesbians. Now, there is a behavioral component to this proposed definition as well; Even so, if applied to humans it would tend to place trans people into separate categories within their assigned-at-birth gender rather than into the gender category they actually identify with. There are a ton of components to human gender identity and expression that animals don’t seem to have simply because you need a certain level of cultural complexity (eg. clothing, makeup, careers) to express them. I find it far less confusing and less potentially problematic to just keep talking about male sunfish as having three alternative mating strategies!
I have mixed feelings about the proposal that we scrap the concept of sexual selection because in textbooks it is often presented in the very traditional way Darwin first formulated (females are choosy, males display/compete for them, etc.) when in fact sex roles and what animals look for in a partner are much more variable. I agree that we shouldn’t be teaching the super narrow, inaccurate, and literally Victorian version; I’ve deliberately added material (including examples from this book) to my lecture on the subject for that reason. However, to say that “the need to attract a mate/select a suitable mate forms an important selective pressure” it isn’t a useful concept because there are wide range of outcomes is a bit odd considering that natural selection itself has outcomes that are contingent on the situation. Will an allele providing antibiotic resistance spread through the population? That depends on if there are antibiotics in the environment (which would make that allele advantageous) or not. Similarly, “Will males or females be more brightly colored?” can depend on which sex is the one that secures a nesting territory and advertises that fact. In eclectus parrots the red and blue female defends the nesting tree, while the male is a duller green. I also don’t personally know any biologists of my generation who use the concept in the narrow way described, though I’m sure there are some. However, I do like the broader concept of social selection, as there can be many ways that individuals exert selective pressures on one another that are not directly tied to reproduction.
Part 2 – Human Rainbows: OK
As with the first part, the section of this part that I have the least quarrel with is the listing of biological facts. Chapter 10 illustrates the process of development from the perspective of “my egg part”, “my sperm part”, and the resulting embryo – a creative narrative device! There are some mind-boggling bits here. For instance, I always forget that the germ-line cells that give rise to sperm and eggs originate when the body containing them is itself still an embryo. The next chapter focuses on sex determination…though there is again a confusing usage of “gender” here to describe gonadal differentiation, which I would A) definitely put as part of sex determination and B) is a usage that works against the author! In any case, Roughgarden does describe the complexity that goes into building a body that can produce sperm or eggs – it isn’t all down to X or Y chromosomes, much less single genes, but rather the interaction of a whole set of developmental genes. For instance, SRY is a gene on the Y chromosome that has an important role in specifying maleness. But some mammal species have no SRY, or even no Y chromosome, and still have males; if SRY is introduced into XX mice, 30% (not all or none!) produce testes. Hormones are discussed next, including the fact that everyone has both testosterone and estrogen in different levels, and that both hormone exposure and presence of the appropriate hormone receptors are important in development2. Chapter 13 gets into gender identity properly. After establishing that actual biological differences between male and female brains are probably much more limited than people often think, Roughgarden notes that, of the “three rice-grains of brain in and around the hypothalamus” that are sexually dimorphic, BSTc might be involved in gender identity. The BSTc in cis men (gay or straight) is about 150% the size of the structure in straight cis women; the 6 trans women tested matched the cis women, while the 1 trans man matched the cis men. Then there is some discussion of how whatever causes gender identity likely happens after physical sex determination but before birth, since genital development pre-dates full brain development, but sex reassignment following injury after birth tends not to work3!
2. Weird tidbit: Clicking sounds can be generated by the inner ear which can in turn be recorded with a microphone. Women’s ears “click” more than men’s on average, but the ears of women with a twin brother (who would have exposed her to more testosterone in utero) don’t. Lesbian and bisexual women also have lower click frequencies. Neat, if somewhat useless, right? Although, given Roughgarden’s not entirely unjustified paranoia about genetic engineering and psychiatry, I’m kind of surprised she wasn’t worrying about eugenicists/homophobes sticking microphones in people’s ears to try to classify them. I certainly have an urge to try and record my own ear sounds now...
3. Technically, presumably, in the 95%+ cases where the kid was initially assigned “boy” and then raised as a girl but was in fact still a boy.
Then we get into the sexual orientation chapter and, oof, we get a lot of accidentally weird statements! Although Roughgarden states that an environment that stigmatizes homosexuality may influence if and how people express such behavior or how they identify themselves4, she seems to keep forgetting that when interpreting/explaining studies. For instance, she says: “One anthropological study tabulates same-sex sexual practices from twenty-one cultures, and in fifteen of these homosexual practice was concurrent with heterosexual practice”. Um, do you mean “…as an accepted thing?” Because otherwise I’m going to hazard a guess that it is really 100%! Or do you mean “concurrent for the same people”? That’s possible - some societies, such as Ancient Greece, have assumed bisexuality is the norm, at least for one gender – but, if so, you should say that! I mean, I appreciate the almost sarcastic treatment of the study that tried to dismiss existence of bisexuality while ignoring the pressures to “pick a side”…but then she mentions “lesbians who are bisexual.” That makes no more sense than “bisexuals who are straight” would. We’re neither, damn it! That’s the whole point!
4. To which I can personally testify. Jeez, the facepalming every time I remember something I said or thought in college that should have had a “no heterosexual explanation for this” sticker slapped on it! But no, it took me over a decade more to figure that out…which was much less about the time or personal maturity, per se, than the shifting social environment.
The remaining three chapters focus on why we shouldn’t over-pathologize human diversity. I absolutely agree with that – particularly chapter 16’s outline of criteria for what is actually a “disease” - but I suspect negative experiences may be leading the author to be a little too harsh on psychiatry and genetic engineering. Both are tools which may be applied in ways that are helpful or harmful. For instance, I know quite a few people who struggle with depression or anxiety who have found the options of medication and therapy very helpful. And a therapist who is well trained can be similarly helpful to a LGBT+ person in dealing with the stress that comes with discrimination or just the soul-searching that is a part of discovering/redefining yourself. The issue is that many are not trained in that way. However, I do appreciate the description of how even psychiatric narratives that are attempting to be helpful – highlighting stories of gender identity emerging before puberty in a kind of sexless way, for instance – can erase the variety of trans experiences (eg. maybe for some people changing their body is sort of a fetish…but is that automatically bad?). Likewise, genetic engineering companies – however they might talk about curing disease or feeding the world – are out to make a profit, and I agree that cloning and genetic engineering can be used in ways that reduce the diversity of populations in ultimately harmful ways. However, it again doesn’t have to be used that way: I recently reviewed a grant that aimed to use genetic engineering to put resistance to chestnut blight – a disease that effectively led to ecological extinction for the American chestnut - into a broader variety genetic backgrounds, to try and recapture the adaptation to local environments that would otherwise be lost.
As evolutionary biologists, I also think we have to distinguish between “genetic diversity is good for a population” and “genetic variation is good for individuals” – natural selection works because heritable variation in traits leads to differences in survival and reproduction. Some allelic differences will kill you, and it is hard not to be on the side of someone who doesn’t want their kid to inherit that allele! My main concerns would be A) it should only be done carefully, for variants that have a wholly negative effect (eg. things like as Tay Sachs or cystic fibrosis) and B) whether people have equal access to such treatments. Admittedly, because of the tendency to pathologize anything that isn’t the average, allowing human germline editing at all could open a can of worms! At least when it comes to things like gender identity, sexuality, and much of neurodiversity (eg. autism), modern genetics studies suggest dozens to hundreds of loci and many potential variants at each locus are probably involved, making “gene therapy” for these non-diseases near-impossible to implement, however much some people might wish to “cure” them. And, since not everyone would want to eliminate alleles even for things like dwarfism that have a simpler genetic basis, so long as the editing is voluntary it would be unlikely to actually purge these variants from the population. Any such attempts would of course be very stressful to anyone who was witnessing an integral part of themselves be treated as a disease…but that happens right now anyway! That way of thinking underlies any harm that might be done to marginalized groups by these tools and will continue to do harm so long as it exists, whether a specific tool is used or not.
Part 3 – Cultural Rainbows: Good
The book concludes with some interesting case-studies of how variation in gender identity and sexuality is defined by and incorporated into various different cultures. It is by no means exhaustive; there are no examples from Africa, northern Asia, Australia, or pre-Christian Europe, for example. But those that are given serve to illustrate some interesting points. For one thing, even where the variation has some kind of official term associated with a role in society, the level of equality/acceptance implied can differ. The various versions of “two spirited” people recognized by Native American cultures were often highly integrated into the society as a whole, respected as warriors, artists, etc. 5 – and this was an aspect of the culture that European colonizers were pretty brutal about suppressing! By contrast, the hijra of India are people assigned male at birth who mostly seem to identify themselves as women, but who are not accepted as such, and whose already limited traditional roles have narrowed, pushing many into sex work and/or homelessness6. Another interesting thing to observe is the way that someone’s self-identification may be shaped by the available vocabulary; you can see many of the interviewees mixing and matching terms in ways different from the way the anthropologist or the reader of this book might, as they try to describe their own experience. And there can be a conflict between traditional classifications and western ones, even if they are somewhat similar: the Polynesian concept of mahu versus the “imported” concept of travesti, for example – with many Tahitians viewing the latter as “over the top” and inauthentic compared to the former.
There’s then a good bit of focus on what the bible does or doesn’t say about gender-variant people or same-sex attraction. This is probably due to Roughgarden’s own Christian faith plus the frequency with which this debate happens in the US, as there is no similar deep dive for, say, Islam – though that would be quite interesting! There is then a switch back to contemporary societies, though I’m not sure why “Tomboi, vestidas, and guevedoche” are all in one chapter! The first seems like it would have been a better pairing with the mahu, the second with the hijra, and the third refers to an intersex condition - people whose genitals look more or less female until puberty, but who grew up to look (and usually identify as) male - that was so common in an isolated area of the Dominican Republic it got its own social category.
Finally, there is a discussion of trans politics in the US. The part about the need for LGBT unity – since homophobes/transphobes generally don’t care about accurately parsing identity! – feels pretty timely with the emergence (re-emergence?) of TERFs (trans-exclusionary radical feminists), a category of thought which unfortunately includes a number of vocal lesbians. The bit about “how many letters in the alphabet” doesn't hold up as well. Roughgarden confidently states that we’ve probably captured all the variation now…but she hasn’t mentioned asexuals at all and while briefly mentioning the concept of "gender-queer" sort of talks about it like one of these things “the youths” are experimenting with. This was last updated in 2013, remember! I do like that it ends on Roughgarden’s personal six point “trans agenda” (you know, since the gays and the trans are always accused of having one!), which can be summarized as:
1. Wanting to be cherished as a normal part of human diversity
2. Getting to tell their own stories
3. To be treated with courtesy and dignity
4. To not be murdered
5. To have equal participation in all social institutions
6. Medical plan coverage of transitions
Amen, sister!
5. Regarding Osh-Tisch, who modern western classifications would call a trans woman, one of her fellow Crows said: “Did the men ever tell you anything about a woman who fought?...She looked like a man, and yet she wore woman’s clothing; and she had the heart of a woman, Besides, she did a woman’s work…I felt proud…because she was brave.”
6. Here, though, potential effects of British colonization are not addressed, which seems weird to me.
Overall recommendation:
This book is excellent when it comes to scientific observations of biological and cultural diversity, which are meticulously foot-noted. The interpretations of some of those facts, on the other hand, might be argued with even by other scientists with similar political opinions (as I’ve done here). In science, we usually try to be as objective as possible, but our individual perspectives nevertheless influence which questions we choose to study and what interpretations we lean toward when discussing the data we’ve gathered. Dr. Roughgarden often states outright how her experiences affected the way she thinks about some of these issues, and I respect that transparency. Indeed, it would be nice if all scientists could be that reflective about their own potential biases!
