New APPS

Art, Politics, Philosophy, Science. Mostly Philosophy.

recent posts

about

    Quick thoughts on pluralism in philosophy of biology

    Philosophy of biology is pluralistic, or so my friends tell me.  Or maybe it would be more accurate to say that many philosophers of biology believe that biology is pluralistic.  One friend recently used the phrase "irreducibly pluralistic."  But I am not so sure.

    There seem to be at least two sources of this pluralism.  One is John Beatty's excellent essay, "The Evolutionary Contingency Thesis." Beatty points out that all generalizations in biology are the outcomes of evolution, and as such, are subject to exceptions (either in the present or the future), and could be otherwise (due to the chance order of mutations and the possibility of solving the same selective challenge in different ways).  And if there are no universal generalizations necessitated by nature in biology (i.e., no laws), then we should expect pluralism.  Beatty highlights pluralism of theories in particular, such as between neutralist and selectionist theories in evolution, or alternative theories about the origin of sex.

    Another source of pluralistic thinking in philosophy of biology surely results from decades of debate over species concepts.  Many philosophers of biology think that different areas of biology reasonably use different species concepts, e.g., one might rely on a morphological definition of species while another might rely on an interbreeding one.  As Marc Ereshefsky so elegantly argues in "Eliminative Pluralism," this is due to the multiplicity of evolutionary processes that divide up the natural world in multiple ways. 

    I am persuaded by both Beatty and Ereshefsky.  But if Beatty and Ereshefsky are right, does that mean that, when faced with a new area of biological inquiry, we should assume pluralism from the outset?  Is it a mistake not to do so?  Old fashioned, even?

    Recently, I have been involved discussions over the concept of "population," which, surprisingly, has not received nearly the philosophical attention that "species" has.  I defended a particular concept of population, which I dubbed the causal interactionist population concept.  In other words, I took the position of a monist. Is that just wrong-headed from the start? 

    I don't think that it is.  My view is that we should try to see if one conception, or one theory, or one process, can do all the explaining for a given domain, even if we think pluralism is likely.  Maybe my monist conception will go down in flames, but if it does, we are likely to learn something from the attempt.  Or maybe it won't.  Just because pluralism is prevalent doesn't mean it is omnipresent.  I even think it is possible that a defensible monist conception of species might still emerge (I know of at least one person who is working on one).

    To me, it is just common sense, or maybe prudence, not to simply assume pluralism.  Then my friends shake their heads at me.   So clearly my sense isn't common.  Should it be?

    Posted in ,

    24 responses to “Quick thoughts on pluralism in philosophy of biology”

    1. John Basl Avatar
      John Basl

      Hey Roberta, nice post. Do you think that your methodological principle of starting from monism for a given type of biological inquiry merely methodological or pragmatic? Or, do you think there’s some evidential reason to favor monism? If there’s a monistic conceptual framework that’s to be preferred for non-pragmatic reasons? Or, is it just that we can hope to get more out of our concepts and working through a problem together if we try to make monism work first?

      Like

      Reply
    2. Roberta L. Millstein Avatar
      Roberta L. Millstein

      Good questions, John. I think I am making a claim about methodology here, i.e., about how to do philosophy of biology in general. However, when it comes to a particular monist claim, then I think that there ought to be more than methodological considerations at work. I tried to provide those for “population,” e.g., by pointing out the consquences for our understanding of selection if populations could be delineated in any way that we choose. So, there might be evidential reasons in a particular case to be a monist. Does that make sense? (I’m still in the process of thinking this through myself, I guess).

      Like

      Reply
    3. John Basl Avatar
      John Basl

      Yes. I thought you were asking something like the following: “When we are approaching a new area of biological inquiry and working out the meanings or criteria of application of various concepts, should we start out under the assumption that a single concept or set of concepts is adequate for our purposes in that area of inquiry?” As it turns out, it looks like the answer is no for certain domains of inquiry (what’s a species? are there universal generalizations in biology?). It seems plausible that for particular domains of inquiry we might argue that monism is appropriate (sounds like that’s what you are up to with respect to populations, but I’d wonder if we can talk of populations in a way that’s useful in some contexts but they aren’t the populations of interest to those thinking about selection). But, that seems different than a default assumption of monism until need is shown for pluralism. I was sort of wondering whether you thought monism was a default evidential position of some kind. I’m skeptical of that, but definitely not of the idea that it’s prudent to start with monism for pragmatic reasons or when there are reasons to be skeptical of pluralism in some domain. Does that make sense?

      Like

      Reply
    4. Charles Pence Avatar
      Charles Pence

      Good stuff, and something I’ve been meaning to think about more. One thought for now — one argument that I’ve always supported for pluralism in biology (and, by extension, in phil. bio.) was best articulated by Ken Waters in his review of Okasha’s book, and which he called “toolbox theorizing”. That is, I take something like the following to be incredibly common in biological practice:

      Biologist has collection of data
      Biologist consults array (“toolbox”) of theories/models/tools/etc. that might be used to analyze that data
      Biologist selects the one she likes best, based on convenience, available data, perspicacity, cheapness, availability in the lab, sexiness, likelihood to get published, disciplinary convention, etc. Importantly — she does not select it because she has any argument that it’s “the most correct” or “best” or “genuinely correct” theory/model/tool for the job
      Lather, rinse, repeat

      Now, you know well that I’ve argued that there can be interesting claims about the interrelationship of those models that might ground more monistic-looking claims about them (e.g., if we can get some grounding or fundamentality relations going, that starts to give the monist a toe-hold).
      But I think, to tie in to you and John’s discussion (hey John!), the prevalence of this kind of theorizing in biology might well lead us to a methodological approach as philosophers that means that you need a good argument that monism is the right way to go — pluralism is, and probably ought to be, the default frame of mind. Thoughts?

      Like

      Reply
    5. Roberta L. Millstein Avatar
      Roberta L. Millstein

      John, yes, that makes sense. I don’t think I’m claiming that in certain domains we should assume monism until a need for pluralism is shown. I think my view is rather that at the outset (or even later – more below) we shouldn’t be assuming monism or pluralism. I further think that, methodologically speaking, it can be advantageous to take a stab at defending a monist conception (iff there seems to be reason to do so, of course) to see how it succeeds and to see if other contenders arise. But I take it that this methodological claim is different from an assumption of monism.
      Even for species, even given all of the debate and the persuasiveness of arguments like Ereshefsky’s, I still would not say that it is a mistake for someone to defend a monist conception. It might turn out that a single conception can, after all, address the concerns raised.
      As for populations, I should be clearer about my monist position — it is only monist within the domain of evolution and ecology. Other areas of inquiry, notably social science, clearly and legitimately use a different concept of population than the one I have defended. But some of my friends don’t think that is pluralistic enough. 🙂

      Like

      Reply
    6. jay odenbaugh Avatar
      jay odenbaugh

      Roberta, Interesting post. I like you, have defended a “causal interactionist” approach to populations, food webs, and ecosystems. However, I think whether it is monistic or pluralistic is a level of resolution issue. Each of these concrete objects are individuated in terms of causal interactions. The specific type of causal interaction is what differentiates these objects as types or kinds of objects. For example, populations might be individuated in terms of interbreeding or selection regimes; food webs would be individuated in terms of patterns of trophic interactions; ecosystems would be individuated in terms of energy flows and nutrient cycling. Thus, they are both characterized in terms of causal interactions and the specific type of characteristic causal interaction. This must be so otherwise we would have to count the entirety of spacetime as a “population” because it is individuated causally through gravitational attraction (supposing that is causal).

      Like

      Reply
    7. Roberta L. Millstein Avatar
      Roberta L. Millstein

      Charles, I agree that you need a good argument that monism is the right way to go. But I would also claim that you need a good argument that pluralism is the right way to go. I am not sure why the prevalence of pluralist practices among biologists, especially when these are largely based on pragmatic considerations, ought to convince me otherwise.

      Like

      Reply
    8. Roberta L. Millstein Avatar
      Roberta L. Millstein

      Jay, I think we are largely in agreement. So yes: there are different “individuals” (or objects, if you prefer — I seem to recall that you differentiate between objects and individuals) at different levels in the biological “hierarchy” (scare quotes because the supposed hierarchy is a bit messy), individuated by different causal interactions, and that it is important to specify the particular type of causal interactions relevant for a purported individual at a particular level.
      Is that a type of pluralism? I wouldn’t have thought so, but I suppose I’m not opposed to thinking of it that way. I don’t think I’ve fully wrapped my head around it, though. Is it a pluralism about the sorts of biological individuals (or objects) that there are?

      Like

      Reply
    9. Roberta L. Millstein Avatar
      Roberta L. Millstein

      Charles, I should also add that I think my position is consistent with Ken Waters’ pluralism. As I understand his position, he does think that there are some ontological reasons to expect pluralism, but that one should be open to finding either monism or pluralism. That is, one should be pluralist in one’s methodology about monism and pluralism. 🙂

      Like

      Reply
    10. John S. Wilkins Avatar
      John S. Wilkins

      Suppose there is no singular monistic notion of species. What would that mean for biology as a science? It seems to me that biologists will still name species the way geographers name mountains in the absence of a monism of mountain building.
      Part of the issue seems to me that we expect there is a theoretical monism for these objects, when I do not see that there has ever been established any case for that expectation. We need something like “organism” or “individual” (except when we do not, as in ramets), something like “population” (except when we do not, as in clonal populations), and something like “gene” (except when that fails to denote), but why do we need a theoretical category for “species”?
      I’ve argued (and will make this explicit in a forthcoming paper with Brent MIshler) that “species” is a phenomenological notion, not a theoretical one – they call for an explanation rather than offering one. Some species are causally connected, while others are not. In the end, the genealogical connections cause species and so the genealogy (clades) is prior to “species”. If it turns out that a species is not “good” (that is, it fails to be a population, a lineage, and a clade) then the species is dissolved into various inclusive clades. But “species” is not a natural kind in my opinion; it is a pattern in the data that is explained, or not, but genealogy and all that goes with it.
      Hence “species” is a pluralist notion, because it is the outcome of plural causal relations, although I expect that these causal relations will usually be shared locally on the phylogenetic tree, since they rely upon shared derived developmental properties.

      Like

      Reply
    11. Charles Pence Avatar
      Charles Pence

      That’s exactly the issue that I’d like to think more about — and I suspect you’ve thought about it some already, as I know your philosophical inclination is similar to mine, so I’d be interested to hear your thoughts.
      I know on the one hand, that as somebody who wants to do philosophy heavily informed by biological science, that I have some obligations toward the details of how biologists do what they do, especially as regards forming my prima facie beliefs about how we ought to approach some part of biology. (I take that to be doubly-relevant here since we’re talking exactly about how to formulate prima facie methodological orientations.)
      But on the other hand, I know just how loose biologists are with exactly this kind of stuff — namely, things like ontological or metaphysical implications, or questions about theory-world relationship. So it also seems like you’ve got good reasons to ignore what biologists are telling you when it comes to a philosophically loaded question like monism vs. pluralism.
      I’m not sure how to balance these — either in general or even in specific cases like this one — but it’s been weighing on me lately…
      (One more thing: I think a lot of people think that we do have a good argument in favor of pluralism. Something like “I know that biologists have all these different models/approaches, and I can’t see a way in which they could all actually be related.” You and I both think that’s a bad argument, but it’s important, I think, to say that it’s probably behind a decent bit of pushback against monistic views.)

      Like

      Reply
    12. jay odenbaugh Avatar
      jay odenbaugh

      Roberta, My preferred way of framing it would be this. Suppose for the sake of argument that nominalism is true — all there is are concrete objects (or individuals). How many types of things are there? One, because everything object is is composed by causal interactions. Many, because objects are composed by different types of causal interactions. On this way of thinking, biology is monistic in one respect and pluralistic in another. Species would be more complicated. Suppose species are populations. However, they are individuated by different types (species) of a type (population) of causal interaction. So, monism and pluralism depends on the level or merelogical nest of organization.

      Like

      Reply
    13. Matthew Barker Avatar
      Matthew Barker

      When I’ve thought about this issue, I’ve been attracted to meta-pluralism. This is the view that both monistic and pluralistic options should be tried. Roberta, I think that you’re response to John that says “we shouldn’t be assuming monism or pluralism” aligns with this. But I think both that response and my meta-pluralism don’t align so well with something you said in your original post: “we should try to see if one conception, or one theory, or one process, can do all the explaining for a given domain”. (Well, they don’t align if by “we” you meant something like “all of us working in this area”. But you probably just mean “some of us working in this area”.)
      Even if we just stick with “some of us should try monism”, then like John I don’t think there is evidence that favors this view as a very general one. But there may be evidence for it in some domains of inquiry, like certain areas of chemistry. Also, it depends what you mean by “evidence”. On some views of evidence, evolutionary considerations probably provide some evidence to think pluralism will be correct in most areas of biology, but evolutionary considerations are not the only ones that apply in those areas and it is hard to say what the evidence favors on balance. One thought is that post-Darwinian biology is awfully young and it may be too early to weight evidence on either side very heavily. That may be a reason to stick with meta-pluralism as a methodological principle for now.
      On species: After teaching it a couple of times (again!) recently, I’m pretty sure the main argument in the Ereshefsky paper you cite is flawed. The argument starts with the last paragraph on p.676. It depends on the premise that different evolutionary forces have split organisms into different types of lineage. A monist can just seize on “lineage” there, saying either that “being a lineage” is a theoretically interesting higher level property that all types of species share in common, which gives the monist an out); or that “being a lineage” is not theoretically interesting in that way, which I think would beg the question in this argument. It would be fun to talk about ways to fix the argument. More generally, if you consider just a small set of data as evidence in this case–say, all the disagreements about the meaning of “species” since the 1930s–then perhaps pluralism looks tantalizing. If instead you widen your data net to include cases of theoretical unification in other areas of science, where centuries were needed before we struck upon unifying concepts, then calling it quits on species right now may start to look like antsy failure of imagination. Especially when there are tools out there like the HPC view of categories.
      On populations: Joel and I consider your population concept in “Deep Conventionalism about Evolutionary Groups”, and argue that a new sort of pluralism applies to it. Jacob Stegenga has a nice new paper coming out in BJPS that argues for population pluralism too. It overlaps in ways with the one by Joel and I.
      On pluralism in biology more generally: you might be interested in section 4 of Rob and I’s SEP entry called “Biological Notion of Individual”. It shows some of the many ways in which pluralisms and monisms can be compatible. (In line with Jay’s comment.) There are many sorts of each, after all.

      Like

      Reply
    14. Roberta L. Millstein Avatar
      Roberta L. Millstein

      John, sounds like an interesting argument — I look forward to reading it. But are you disagreeing with something I’ve said here? It doesn’t seem to me as though you are.

      Like

      Reply
    15. Roberta L. Millstein Avatar
      Roberta L. Millstein

      Charles, I agree that as philosophers we have an obligation to pay close attention to the details of how biologists do what they do. But when some biologists are monists (and some are), there is no way to consistently satisfy them all. So, my view is that we need philosophical arguments (which can themselves be informed by biology) to make the case for pluralism or monism in a particular domain. I would thus urge you not to feel too weighed down. 🙂
      And yes, I think you’re right in pointing to another reason why some philosophers of biology are pluralists, and yes, I agree it’s not a very good reason.

      Like

      Reply
    16. Roberta L. Millstein Avatar
      Roberta L. Millstein

      Ok, I think I see what you’re saying now, and that sounds right to me, although it seems like this is a less pluralistic pluralism than most philosophers of biology want. (I resist the view that species are populations, except in the special case where a species is composed of one and only population rather than multiple populations).

      Like

      Reply
    17. Roberta L. Millstein Avatar
      Roberta L. Millstein

      Matthew, meta-pluralism, yes. And yes, I did not mean that we should all be monists at the outset of a new inquiry, although I think we should all resist automatic (I am tempted to say “knee-jerk”) rejections of monism. Considering the possibility of monism, even if one is not pursuing a monist argument oneself, is all that I am claiming that we ought to do.
      I certainly was not claiming that there is general evidence for monism; as I think I said, I actually find the arguments for thinking that pluralism is pervasive to be persuasive. But pervasive is not omnipresent (and now I repeat myself).
      On Marc’s paper – given that he says that the lineages formed by the different evolutionary processes are not the same, there is no real comfort for the monist there, it seems to me. Anyway, I like the paper because it gives specific, reasonable grounds for pluralism. However, as I said above, I don’t think it’s necessarily the last word on whether monism or pluralism characterizes species concepts.
      Interesting suggestion to think about the short vs long view with regard to biologists’ pluralism. I think you’re right that things could certainly change in that regard.
      I’ll take a look at your paper with Joel (I think I saw an early version?) I asked Jacob to send me his paper, but he has not done so. And I’ll look at your SEP entry. 🙂

      Like

      Reply
    18. David Duffy Avatar
      David Duffy

      As a geneticist, I don’t see conceptions of population differing much from the original statistical sense, alluded to, for example, in the citations in Sober’s “Evolution, Population Thinking, and Essentialism”, (1980). In population genetics, one happily models sub- and sub-sub- populations as necessary (as hierarchically nested sources of variation); admixture, introgression, migration between “discrete” populations; as well as continuous smooth models (say clines in frequency of an allele by geographical location). The mathematical concept is already infinitely flexible, and the other uses in biology partake of that.

      Like

      Reply
    19. Jonathan Birch Avatar
      Jonathan Birch

      I share David Duffy’s concern that there is a danger here of ‘over-theorizing’ a concept that is already well understood by demographers and statisticians, albeit in a minimalist way.
      The minimal notion of a ‘population’ is that of a set of individuals or objects of interest. Thus construed, you can have a population of RNA molecules, or genes in a cell, or cultural variants, or viruses, or elephants, or insect colonies, or even whole species. This is all good news for someone who thinks that, in principle, ‘population thinking’ is something that can and should be applied at all levels of the biological hierarchy.
      Moreover, embracing a minimal/deflated concept of ‘population’ does not mean the end of philosophical debate. On the contrary, it provides a useful starting point for substantive debates about what a population has to be like, in order to possess certain biologically important properties or undergo certain biologically important processes. For example, we can ask what a population has to be like in order to undergo cumulative adaptive evolution; in other words, we can ask what makes a population more or less ‘Darwinian’. This, of course, is the core of PGS’s project in Darwinian Populations.
      In general, I think a number of ‘monism vs pluralism’ debates can be deflated by embracing a minimal conception of the relevant notion, whether it is ‘population’, ‘individual’, ‘species’, ‘gene’, etc. I also think deflating these debates can be a good thing, if it helps us move on to more substantive questions about the nature of biological processes.

      Like

      Reply
    20. Matthew Barker Avatar
      Matthew Barker

      Hey Roberta–ya, the paper by Joel and I is one you saw an earlier draft of, and before it was published we reframed partly due your helpful comments.
      Don’t get me wrong, I really like that paper of Marc’s too. That’s why I continue to teach it. My point was just a very restricted one concerning his main argument. The premise I singled out cannot say that the different forces produce lineages that have nothing interesting in common, because that would be question begging–that’s what the conclusion is supposed to say, on the basis of some independent reason. To avoid the circularity, the premise will have to intend that the lineages are different in some ways while leaving open that they are interestingly similar in others. (Hammers and saws are different in important ways, but both are tools, which is interesting in some contexts.) But like you, I don’t think this saves monism in general or makes pluralism hopeless. It would be fun to try fixing the argument because I think it is an important one.
      David’s mention of Sober’s 1980 paper reminds me that at the end of that paper, I think Sober allows for a kind of essentialist natural state model with respect to the concept “population” (though of course not with respect to “species”, which the paper had been mostly about). That may comfort monists about “population”.

      Like

      Reply
    21. Roberta L. Millstein Avatar
      Roberta L. Millstein

      David and Jonathan, I raise concerns for a more flexible concept of populations in the following two papers:
      http://home.comcast.net/~rmillstein1/papers/Millstein2009populations.pdf
      http://home.comcast.net/~rmillstein1/papers/Millstein2010.pdf
      Since this thread is about pluralism and not about the concept of populations per se, I’d be happy to hear any reactions you had to those arguments over email. I do discuss PGS’s view briefly in the latter paper.
      But yes, deflating a concept to its most minimal is one way of achieving monism. My concern is that a deflated concept no longer does the job that we need it to do for evolution and ecology, and so we still need a more specific concept in any case.

      Like

      Reply
    22. Roberta L. Millstein Avatar
      Roberta L. Millstein

      I definitely agree that there is more to be said about the work that “lineage” is doing in Marc’s paper. That could be an interesting response.

      Like

      Reply
    23. Arnon Levy Avatar
      Arnon Levy

      Speaking to the general issue in the initial post (i.e. not specifically about populations and/or species): What about the (familiar and not biology-specific) idea that one ought to seek a unified/monistic* theory for cognitive and epistemic reasons? Cognitive: all else equal, unified/monistic theories are easier to understand and use. Epistemic: all else equal, unified/monistic theories are more likely to be true. Myself, I am somewhat skeptical of the epistemic rationale, but I place considerable weight on the cognitive rationale. I tend to think that, even in a fairly messy domain like biology, it motivates a “defeasible presumption in favor of unification” as Sober puts it in his “Two Uses of Unification” (although he there defends the epistemic rationale.)
      * I use “unity/monism” to highlight the connection to general discussions of model selection etc.

      Like

      Reply
    24. Roberta L. Millstein Avatar
      Roberta L. Millstein

      Interesting. I agree with your skepticism about the epistemic reason. The cognitive one is a bit more appealing. If it seemed as though a monistic conception was likely to be able to handle a large number of cases, it would gain even more traction.

      Like

      Reply

    Leave a comment