About Rationally Speaking

Rationally Speaking is a blog maintained by Prof. Massimo Pigliucci, a philosopher at the City University of New York. The blog reflects the Enlightenment figure Marquis de Condorcet's idea of what a public intellectual (yes, we know, that's such a bad word) ought to be: someone who devotes himself to "the tracking down of prejudices in the hiding places where priests, the schools, the government, and all long-established institutions had gathered and protected them." You're welcome. Please notice that the contents of this blog can be reprinted under the standard Creative Commons license.

Friday, December 07, 2007

Do you believe in human nature?

My professional field of study is gene-environment interactions, or what philosophers and social scientists call nature vs. nurture. That is why, whenever I’m asked if I believe there is such a thing as “human nature” my answer is yes and no. Yes, in the sense that the genetic makeup of the human species is distinct enough from that of other species to make us different in a variety of non-trivial respects (of which having developed blog technology is just one). No, because human beings are genetically contiguous with other primates (which means there are very few things that are truly uniquely human), and also because we know that genotypes can produce very different outcomes depending on what the environmental conditions are (a phenomenon known in biology as phenotypic plasticity).

And yet, it is amazing how polarizing this sort of question can be, and has historically been. Moreover, it is interesting to find predictable (though not absolute) differences between practitioners in different fields. A philosopher colleague of mine, for instance, was recently incredulous that I would even mention the term “human nature” in a class we were teaching jointly. For her, it was simply laughable that any sensible person would entertain such a clearly silly notion. Surely, her argument went, decades of cultural anthropological studies have shown that human behaviors are pliable at will, that there are no human universals, and therefore no “human nature.”

Au contraire. It is quite obvious that there are limits to the range of both human physical and behavioral attributes. Going to the gym and eating less will make you lose weight, but it will not get you six-pack abs unless you are genetically predisposed to have them (in which case, you’ll get them even if you don’t go to the gym and eat cheeseburger for breakfast). It’s unfair, but so is life. Analogously, most of us can learn to play a musical instrument by practicing, but -- contrary to the famous quip -- it takes a lot more than practice to get to Carnegie Hall. As for human universals, there are quite a few, some of which we actually share with other animals, such as the fight-or-flight response, or the expression of many of our emotions, as elegantly documented already by Darwin.

On the other hand, I get equally annoyed when I hear some of my colleagues (Steven Pinker comes to mind) going on and on about how we are not a “tabula rasa” because our genes encode much of what we are and do. Once again: genes encode proteins, and absolutely nothing else. From proteins to complex traits such as cognitive ones there is a vast unknown ocean referred to as “epigenetics” (literally, beyond genetics), the surface of which we are barely beginning to scratch. Our genetic makeup certainly poses limits to what we can and cannot do, but how ample those limits are is currently largely beyond the scope of human biology, partly because we cannot do the right experiments that would settle the matter (it is both impractical and unethical to breed human beings and raise them under controlled environmental conditions, which is what we do with other animals and with plants when we wish to study gene-environment interactions).

Even among philosophers, of course, the question has hardly been settled. While John Lock orginated the phrase tabula rasa, Aristotle famously thought that the essence (the nature, we would say) of humanity consists in our ability to reason. Judging from the history of the intervening 23 centuries, he probably overestimated how sapiens we really are. Nonetheless, humans are the only species with a developed language, art, science and technology. And surely those things ought to count as evidence of the existence of a human nature. The problem with trying to be a reasonable skeptic is that one easily makes enemies on both fronts of any debate: you acknowledge that genetics does set limits to human characteristics, and you get accused of being a genetic determinist and possibly encouraging eugenics. You grant that the environment plays a sometimes major role and you are ridiculed as an anti-scientific fuzzy thinker. Wake up, ladies and gentlemen on both fronts: the reality is both more complex and more fascinating than either caricature would allow. It is neither nature nor nurture, it is -- as the title of an unusually balanced book by Matt Ridley puts it -- nature via nurture.


  1. How odd. I thought the consensus has long been that "both" was the answer to the nature/nurture question.

    What is it that makes people insist that there must be a clear "either/or"?

  2. While I agree with the general gist of your post, Massimo, I think you exaggerate there a bit... While it's clear that Pinker favors the innate components, the way you talk about it paints him as a classic genetic determinist, which I would say he hardly is, from what I remember reading in "The Blank Slate". And, technically speaking, saying that there is no tabula rasa is completely correct, because obvious: as long as one little tiny thing is innate (even ignoring the most basic instincts and reflexes), the tabula is not rasa anymore. So what would really be the problem? As Christine said, only the uninformed (even if majority) still fight over this.

    But the worst, for a genomicist, was this fine example of a straw man:
    Once again: genes encode proteins, and absolutely nothing else.

    All right, it's just proteins, they don't do anything of interest, let's just move on... (my turn to straw man) First of all, your statement is incorrect by modern biology standards. The old "one gene, one peptide" idea has long been abandoned -- or should be, by those who haven't been following the subject more closely. The concept of gene is much more nuanced, I'd say.

    And, more important, even if it were correct that "genes encode proteins" and that's it as you say, you gotta remember a genome is mostly NOT genes. And the "junk" portion shrinks a bit every once in a while, although I don't think it will really get anywhere close to disappearing. I'm sure you must have heard of the regulatory activities of the non-coding areas of the genome. What about micro RNAs and other non-coding, highly conserved sequences whose function is mostly unknown?

    More important than the parts is how they are organized -- and look out for those pesky emergent properties. Complexity science 101. :-)

  3. "Analogously, most of us can learn to play a musical instrument by practicing, but -- contrary to the famous quip -- it takes a lot more than practice to get to Carnegie Hall."

    What about becoming a scientist?

  4. "Once again: genes encode proteins, and absolutely nothing else."

    Uh, well proteins that may be expressed in many different forms depending on post-transcriptional processing, and untranslated RNAs that control which other genes are activated, and who knows what else.

    The view of "genes" as merely instructions for making specific proteins is unnecessarily limited.

  5. J,

    c'mon, you don't really give me enough credit here. I am a professional biologist, so I do know about genes indirectly affecting all sorts of other things other than proteins, and I do know about iRNA and the rest. But my point remains: most genes "do" only proteins, and I never said that makes them irrelevant. Everything else comes through a complex series of indirect interactions that must include environmental signals and all sorts of epigenetic effects. Which means that "nature" cannot be the major determinant of things.

    As for being uncharitable to Pinker, I simply mentioned his name in passing, I did not provide an in-depth analysis of his positions. However, many "naturists" are careful at acknowledging complexity, environmental effects, etc., but then turn around and relegate them to secondary roles. Not even E.O. Wilson is a straight genetic determinist anymore. I guess that's progress.

  6. j,

    As an aside, I wouldn't expect too much insight about what makes us human to come out of more research into 'junk' DNA. Much (if not most?) junk DNA is known to be nonconserved repetitive DNA and selfish genetic elements that insert themselves around the genome. Only on the very rarest occasions might be they be exapted for the benefit of the organism (though I know of no such case). As for RNAs that don't encode protein, their function can still be readily described in terms of gene regulation, enzymatic activity, etc. Thus, I don't think that bridging the genotype-phenotype divide will arise from a more nuanced understanding of a gene. Rather, it will be something like continuing to figure out gene function and the complex gene-gene and gene-environment interactions. Or something like that...

  7. Hey, Massimo (and Chris)

    c'mon, you don't really give me enough credit here

    Sorry it sounded that way. Of course, I was 99% sure you did know those things, and sure more. But your *readership* is not composed exclusively of professional biologists though, so I thought a little discussion might be in order...

    However, many "naturists" are careful at acknowledging complexity, environmental effects, etc., but then turn around and relegate them to secondary roles.

    That is very true, and I did think that while reading some passages of Pinker's book. Actually, I've seen that in many seminars by many researchers in diverse areas. I guess it's natural that people would do that. Human nature? ;-)

    But I still have some problems here. You're right, in a sense, with the gene-protein thing. I sympathize with that, since in the couple genomes we annotated and the others we're working on "gene" is what codes for proteins, plus a few tRNA and rRNA "exceptions". But that is just our perceptional bias at work! It's a circular thing: it's much easier to find protein-coding genes, therefore we do find them almost exclusively, which leads us to say that genes practically only make proteins. It is easy to find protein-coding genes, and almost impossible to find non-protein-coding ones (with the honorable tRNA and rRNA exceptions mentioned). So, that's what we do. In a bacterium, fine; >90% of the genome encodes for proteins anyway. But things get muddier in eukaryotes, even unicellular ones sometimes.

    So, as usual, the problem is in our definitions and the way we think about things. If gene is defined (as I do) as everything that is transcribed in the genome, then we're VERY wrong in our gene-protein bias, according to the most recent data from the past few years. Apparently, MOST of the human genome is made into RNA, and only a tiny fraction of that codes for protein. Since I have smaller fish to fry, I don't follow this "big organism" stuff as closely as I should, but that seems to be where we're headed at.

    I just did a quick search for brief texts on this and you can take a look at these two: here (RNA study finds activity in ‘silent’ regions of human genome) and here (New Findings Challenge Established Views on Human Genome). That is supposed to answer your objections to a point too, Chris. The "junk" is quite mysterious, still. The fact that it is repetitive and not conserved does not mean much regarding it's functional status (or lack thereof).

    (Digression: the cover of Nature shown in one of the articles above, with the "Decoding the blueprint" line... The metaphors of genomes as "blueprints" or as "book of life" is so poor, I think. I like to use "software of life" in my little lecture in genomics; not perfect but much closer to reality, IMNSHO. Anyway...)

    Regarding looking to the DNA to figure out what "makes us human"... Well, I agree that strict reductionism is not the best tool here, given the nature of the system.

    Now, Massimo, a question about the epigenetic issue, since it's something whose literature I'm not following: where does it come from, in your opinion? What I mean is: epigenetics, as far as I remember, is regulation of the genome by methylation and/or acethylation (is that too narrow? would you say there's more to it?). So the environmental influences will give the signals. But where is the information telling, for example, the cellular machinery what to turn on/off and when? It most surely is not in the environment. Do you see where I'm going here, right?

  8. j,

    To answer backwards to front, there is evidence that some transgenerational epigenetics is environmentally induced. So, the 'effect' is on the genes, but the cause is the environment, if that answers your question.

    You're right, there are many examples of unexpected non-protein genes, but as of yet there is little evidence that the bulk of large eukaryotic genomes are composed of such functional elements. Conversely, there is much evidence that most of junk DNA is composed of genomic parasites that have a neutral if not slightly deleterious effect on their host. The fact that most of the genome is transcribed at a low level could probably be explained by spurious transcription caused by random sequences with low binding affinity to certain transcription factors. Thus, most junk DNA truly is junk, though it may rarely prove to be a beneficial mutation retained by natural selection.

  9. J,

    the field of heritable epigenetics is just about to explode. Look for a couple of major reviews to appear in the next few months (one from my lab, in Ecology Letters).

    Yes, the environment "only" triggers epigenetic markers (such as methylation sites), but there is good evidence of complex interactions between the genetic and epigenetic layers of information: after all, the ability to methylate is controlled by specific genes, but the methyl groups in turn control whether certain genes are on or off. The possibilities are not quite endless, but pretty numerous!

  10. Excellent post, I couldn't agree with you more. I guess Steven Pinker is just pragmatic on pushing the envelope a little bit, the truth is that to most people in the social sciences even the hint of genes determining something is a major heresy.

    (Plant biologist mode: on)

    I'm not sure I agree with you about the field of heritable epigenetics exploding soon. We know (roughly) what genes do and even more roughly what environmental cues do to phenotype. But the interface is quite blurry and I think the main drawback is experimental - do we have the power of analysis it takes to dissect G+E interactions yet? I've read heaps of your papers but still my feeling is that the answer is 'no'.

    Looking forward to your next review.


  11. Great post - I think you make some insightful points. As an editor for The Issue, a recently launched Blog Newspaper, I've decided to feature you on our site.

    You can find a brief excerpt and link back in the Science and Health Section of The Issue .


    Science&Health | The Issue

  12. I think Steven Pinker provided a much-needed critique of the all-environment legacy of Social Sciences in the last century. Its understandable that in knocking down the environmental causality bias it might appear that he went too far in the direction of genetic determinism. But to me it was a breath of fresh air. If you had ever sat through a womens studies class in the 1970's, you might have wished you had Steven Pinker with you.


Note: Only a member of this blog may post a comment.