When do species begin? 11 Jun 200922 Jun 2018 Last night I attended a talk by Alexander Bird of Bristol on the metaphysics of natural kinds. I confess that a lot of it struck me as largely irrelevant to the actuality of the science as the level of metaphysics here, involving possible worlds and necessity, not to mention semantic properties, exceeds the metaphysical concerns of sciences. Biology, as I often say, has a very small metaphysics. No infinite denumerable lists or impossible worlds in biology. In the course of the talk, Bird made the following claim, about speciation [below the fold]. Take Ernst Mayr’s model of speciation by allopatry or peripatry: In between steps 2 and 3, according to Bird, there was a point at which there was a last breeding pair, after the death of one of which there was now a new species. So there is a moment at which a new natural kind appeared, with no intrinsic properties of the organisms changing. I find this rather odd. The property of reproductive isolation presumed by the conception of speciation here is a property of the population, not of the individuals that comprise it. And moreover, as Mayr and those who follow his work, such as Coyne and Orr make clear, what has changed is the probability (which we can read for now as either the frequency of interbreeding between the populations, or as the likelihood that any individual in one group will interbreed with any individual in the other) of interbreeding. On the frequentist account, that reduces to zero at the moment of the last interbreeding (which we can only know post hoc), and on the likelihoodist view, it never reduces to zero, but only approaches it. But that is not the problem. Instead, it is this: to be a species precedes the last interbreeding event by a long way. Interbreeding between species is rife, but they remain species. Coyne and Orr talk about lowered probability of interbreeding, not the absolute barriers between species. So we can see the two populations as being probability distributions about reproductive isolation. Each population has shared reproductive mechanisms from their common ancestor – the issue is whether or not interbreeding, and the subsequent gene flow (the technical term is introgression), is sufficient to disrupt the overall cohesion of the population or not. In many cases, where interbreeding post-speciation is rife (such as, oddly, primates), selection against intermediacy is sufficient to maintain the ecological and morphological, and most significantly the reproductive and developmental, typicality of the species. Novel genes add to the resources of the species, but they do not undercut it (in cases where they don’t. Stage 5 of the figure shows where they can). In the case of hominids, it looks like occasional interbreeeding with the lineages leading to Pan (chimps and bonobos) occurred up to one million years after the lineages were formed by a speciation event. In other words, there is never a zero probability of interbreeding after speciation, until the clade in which the species occur goes extinct. What this means for species as natural kinds, well, that’s for another time. But if philosophers are going to use biological examples (and they always have) it pays to use realistic examples. The idea that species appear at a single point in time is fundamentally mistaken – except (for there are always exceptions in biology) when species are formed by processes like polyploidy, when individuals are formed that have radically different genetic structures, which also often happens in hybridisation events. If species are natural kinds in the modern sense of Mill and Russell and Quine, and if they have essences, these are not at all like the usual philosophical kinds where it is all-or-nothing. Evolution Metaphysics Philosophy Science Species and systematics
Philosophy On Templeton money 4 Jun 2010 I work on the naturalising of religion: that is, the philosophical implications of religion being an evolved and natural human phenomenon; I’m even writing a book about it. I would love to get a grant to support that work, and indeed as an academic I am required to seek grant… Read More
Evolution How to get an improbable outcome 16 Jun 2008 Creationists and Darwinian skeptics often claim that natural selection could not produce the sort of improbability (often, for reasons that nobody is quite sure of, below 1 in 10 to the 500th power) that we see around us. So it comes as a pleasant surprise to find that UK skeptic… Read More
Philosophy Latest developments on the Synthese affair 6 Apr 2013 Readers will recall that a couple of years ago, Synthese published a special issue on evolution that carried a scandalous disclaimer by the editors in chief that the papers included were not up to par. I had a paper in that edition. One of the editors in chief, Gualteiro Piccini, has… Read More
John you have made a category failure in the administration of your blog. ‘When do species begin?’ is a post and not a comment!
“If species are natural kinds in the modern sense of Mill and Russell and Quine, and if they have essences, these are not at all like the usual philosophical kinds where it is all-or-nothing.” This is leaving me hanging. Did Bird go on to develop this into: a) but species are actually very much like philosophical kinds… b) so philosophy can here learn from biology in terms of natural kinds… c) so we have to be pluralists about kinds, and adjust our notions of them for different domains… or d) so we’ve gotten SOMEthing wrong, SOMEwhere?
His argument was basically that if you are an essentialist about natural kinds you must be a realist about natural kinds. Several counterexamples were given, and the argument went some distance away from the nature of species.
All right. I can see how that follows, but did he defend why one should be an essentialist about natural kinds in the first place?
did he defend why one should be an essentialist about natural kinds in the first place? Ever since Kripke defended essentialism in the philosophy of language, especially with respect to universals, it has been an assumption that essentialism is the default position, as it were. This contrasts greatly with the consensus in biology, but it’s a different game in linguistic analytic philosophy.
For me visiting this blog just gets more tremendously exciting on each occasion I visit. I always thought that I was dealing with a subject area that owed more to art rather than science. That was an utterly foolish view. It’s the thought that understanding and applying this subject may not just explain the origin of Humanity from a biological perspective but that it can be applied to understanding the development of art and culture in human society and provide solutions to definitional problems that are somewhat vexing in my own subject area. Much of what is discussed in these pages is the most exciting material I have had the pleasure to chance upon. It is a most fruitful subject area which has utterly changed the way I look at my own subject and the methodology I construct to engage in research.
Even polyploids aren’t safe, you can have multiple independent origins of polyploids from the ancestral non-polyploid, and these independent polyploids can breed with each other but not their ancestral species. At least theoretically this is as one-way gene flow mechanism.
There is also respeciation in verts – to be precise, fishes – so you can have a new species which repeatedly evolves. Turner, G.F. 2002. Parallel speciation, despeciation and respeciation: implications for species definition. Fish and Fisheries 3 (3):225-229(220).
I’m not sure whether it is a faux pas for the subject of a post to respond. But here goes. First, the point I was making (not an especially original one—see Mohan Matthen “Chickens, Eggs, and Speciation,” Nous 43 (2009): 94-115) was somewhat tangential to my main line of argument. I just wanted to refute the Aristotelian view that an individual’s species is fixed and cannot be changed. The point is most vivid with a sudden speciation event, but even if speciation is spread out over time and is a matter of degree, the Aristotelian view is still refuted (since an individual will change the degree to which it belongs to a species). There is an interesting general point you make about the relationship between science and metaphysics. On the one hand I think it is absolutely necessary that a satisfactory metaphysics cohere with science. On the other hand, I don’t think that metaphysics gets its legitimacy solely from its relationship to science. There is more to metaphysics (even `metaphysics of science’) than can be read directly off from the science itself. (In some ways this is a bit like the relationship between mathematics and science, although I concede that maths is usually rather more useful to science.)
Alex, it is absolutely wonderful if the victim, sorry, subject, responds. You are perfectly right that “the Aristotelian” view is false, but it’s my argument that there never was such an Aristotelian view – in natural history. There certainly was in logic and metaphysics, though. I didn’t think it was appropriate to raise this in the Q&A, as it would have required ten minutes setting up the background, and would have detracted from the major point you were making and others were discussing. The metaphysics of biology is rather more limited than the metaphysics of modal logics and the like, as I noted. But I have seen that example used several times (Laporte used it at a conference I was at) and it is simply biologically unrealistic. That is not, in itself, reason to abandon it (it could be a toyworld example, or a schematic, so long as it is flagged as such). But when philosophers use biological examples, which again I noted they always have (“plucked chicken”), it’s really important that they do not import assumptions from folk biology, and the way you appealed to it did so, I fear. The reason why I blogged about it was to get myself clear on a talk I will one day give, on what hybridisation means for species concepts. This provided a nice hook. Thanks for the talk; it was not bad for a specialist in a minor discipline like me to hear.