Skip to content

Darwin’s motivation

For some time now I have been convinced that Darwin’s original and most pressing problem was not adaptation. It was the existence of taxonomic diversity. I have thought that the debates over what was a natural classification amongst the unjustly derided Quinarians William Sharp Macleay and William Swainson were the jumping off point for Darwin’s theory of common descent. Michael Barton has pointed me to this recently added transcription at the Darwin Correspondence Project, of a note he wrote in February of 1841:

Feb /.41/

There is such disputes about affinity, linear, circular arrangement &c &c definition of species — that I cannot, as in geographical distribution & — as in relation of fossil to recent, or state my facts & deduce consequences, — but I must show, how what light the conclusions deduced from other parts bear on those vexed questions — After showing what a ‘species’ is add remark that question about origin of all our dogs became less material — it is simply whether nature or man, has transported, exposed to different climates & selected the offspring from one common stock. — or whether these have been effected by natural or artificial means

The language here is clearly influenced by the discussions of the Quinarians, in particular Swainson’s Preliminary Discourse of 1834. At this time it was a widely held view that local geography caused the changes in varieties and local species from a common stock, a view proposed by Buffon in the previous century and to an extent revived by Geoffroy.

“Affinity” is a misunderstood term. Often it is read as meaning, as O’Hara 1991 put it, “a relationship based on some sort of essential similarity”. He rightly goes on to note a distinction with homology: “While the later and somewhat related concept of homology was rarely discussed in the purely systematic literature (homology was a relation that obtained among characters, in contrast to affinity which obtained among taxa), discussions of affinity pervaded that literature.” But affinity was simply a summary of the overall homological relations between taxa, and the idea that something was essential was never, to my knowledge, asserted at the time. In short, affinity is a kind of “parsimony analysis” of homological relations.

Darwin is motivated to explain these affinities. He does so by assuming that for all intents and purposes the taxonomic scheme that is current is a good approximation to the natural system, at least at some levels. And he explains these relations by appealing to shared ancestry. As I have written before, homology precedes ancestry. That is to say, we know the homological affinities of taxa before we explain them by appealing to shared ancestry; homology is not identical to shared ancestry. Darwin knew this.

O’Hara, Robert J. 1991. Representations of the natural system in the nineteenth century. Biology and Philosophy 6 (2):255–274.


  1. Bob O'H Bob O'H

    Those O’Haras get everywhere, don’t they?

    • John S. Wilkins John S. Wilkins

      These Robert O’Hara’s get everywhere. Like chewing gum on pavements…

      Fortunately there are only a few dozen John Wilkinses.

  2. John Harshman John Harshman

    I think I must disagree. A concept of homology predates ancestry, but the modern concept of homology is all about ancestry. What began as an explanation has become the definition, and rightly so in my opinion. Instead of asking “What causes this homology?” we now ask “Is this similarity homologous?”

    I see no reason why we have to reconcile all potential meanings and adopt some sort of core compromise. Patterson’s formulation, “Homology is synapomorphy” seems quite enough for me, and anything else is counterproductive.

    Or did you mean something much cleverer than that?

    • John S. Wilkins John S. Wilkins

      A good point. I would be the last person to challenge the idea that scientific concepts can evolve, and of course I like Patterson’s definition. But “synapomorphy” is not necessarily an evolutionary term (even Patterson thought that it wasn’t); some read it as being an evolutionary state, while others treat it as the distribution of character-states in a cladogram where that could as easily be a Venn diagram or, as it was in Darwin’s day, an indented list with braces.

      But I think that a lot of confusion arises from the fact that these terms are being used as synonyms for several related ideas. The older sense remains, but newer senses have arisen.

      • John Harshman John Harshman

        Can you support your interpretation of Patterson’s definition as being “not necessarily an evolutionary term”? Whether it was Patterson’s view or not, I agree it may be the view of at least some transformed cladists. But I would also consider that view pathological.

        I think you are confusing representation with intent, too. Whether we represent a cladogram as a tree, Venn diagram, or indented list, the question is what we mean by it. If it’s just a handy summary of some character states, who would care? But if it’s intended to reflect real evolutionary relationships, that’s important. Similarly, if a synapomorphy is just a similarity, even a patterned one, who cares?

  3. Jim Thomerson Jim Thomerson

    I think Darwin’s barnacle experience; being a working taxonomist, was very important in guiding his thinking.

  4. Robin Craw Robin Craw

    While the architects of the new systematics (e.g. Mayr) have ridiculed Macleay’s ideas both David Hull (1988, Science as a Process, pp.92 – 96) and Nelson & Platnick (1981, Systematics & Biogeography, pp. 110-118) provide reasoned discussion and reproduce figures from Macleay’s (Hull, N & P) and Swainson’s (N & P ) work. More recently Williams & Ebach (2008, Foundations of Systematics & Biogeography, Chapter 9) provide quite an extensive discussion and many references to the Quinarian System and its influence. It hasn’t been that hard done by, and Williams & Ebach (2008, p. 159) note : ” Macleay’s separation of affinity and analogy had a marked effect on many naturalists”. Macleay ended up in Australia and Swainson emigrated to New Zealand.
    That Darwin’s thinking was heavily influenced by his taxonomic work on barnacles is well covered in Michael Ghiselin’s book Triumph of the Darwinian Method

    • John S. Wilkins John S. Wilkins

      Yes, the Williams and Ebach book is one of my general guides, along with Nelson and Platnick 81.

      But even O’Hara noted that too much emphasis is placed upon the numerology of the Quinarians, and too little to the excellent discussions they made about “natural classification”. And their influence was upon contemporary naturalists; which is what I claim here. They are not widely read these days.

      A couple of points about the barnacle work. One is that Darwin explicitly stated that his thinking about evolution did not materially affect it (apart from the embryological affinities with other crustaceans, which could have as easily been done by a Geoffroyan), and in any case he treated the work as standard Linnaean classification. I defy anyone to find either a precursor to evolutionary thinking (which he had finalised the core theses of by 1844, when the Cirripedia books were published in 1851) or an implementation in them. Of course, I haven’t read them through.

      The other is that the barnacles were about the group, which had already been identified as “natural”; in sum, he was no more influenced in his evolutionary thinking (i.e., his phylogenetic thinking) by the barnacle work than he was by Linnaean classifications in general. That’s the wrong level to conceptualise it.

      I have been to the Macleay Museum in Sydney (and donated a copy of Omphalos to them); unfortunately Macleay is not greatly featured there. But I did get a copy of his Horae Entomologicae.

      Oh, and Williams and Ebach are friends and collaborators.

  5. Robin Craw Robin Craw

    As well Ghiselin published a paper on Darwin’s barnacle monograph :
    M.T. Ghiselin 1973 Phylogenetic classification in Darwin’s monograph on the subclass Cirripedia. Systematic Zoology 22(2): 132 – 140.

  6. Robin Craw Robin Craw

    Not much from that period (1820-1835) on systematics is widely read these days – also both Macleay’s and Swainson’s work is hard to track down at least in New Zealand. I don’t get the point you’re trying to make John – do you disagree with Ghiselin’s interpretation of Darwin’s barnacle work ?
    On a lighter note I have found a potential famous ancestor for you who should appeal – Timothy Wilkins (d. 1671) an Epicurean.

    • John S. Wilkins John S. Wilkins

      It’s been a while since I read it (about 20 years). I’ll get back to you on that.

      I have no famous ancestors, but several interesting precursors. As an Epicurean myself I like Timothy. As someone who had an impact I like Bp John, first secretary of the Royal Society and the sponsor of Newton’s Principia (but a neo-Platonist, and hence mad). As an Australian I like Hubert, Antarctic explorer. As a philosopher of biology I like Maurice. But none of them are in any useful way my forebears. I think someone on my mother’s side was hung, drawn, and quartered as a highway robber once (but only once).

      Intellectually, of course, I am Cuvier’s descendent, but then so is everybody.

    • John S. Wilkins John S. Wilkins

      Oh, and both Macleay and Swainson are available online with some searching. I start with and then if they link to Google Books I click on the “http” link, which will often give the PDF directly.

      Ghiselin’s comment that Quinarianism “was, in effect, a numerological system, and therefore incompatible with the theory of evolution” (104) is misleading. It was the discussion of “the natural system” that was crucial for Darwin, not the specifics of the representations (although I think he may have considered how to represent evolution as a tree or a coral based on reflecting on Macleay and Swainson’s osculating circles). But I do agree that the key issue with the barnacles was the anatomy: comparative anatomy underlies his use of taxonomy.

      Ghiselin’s treatment of the German morphologists is in line with the received view, and I think underplays the empiricism of the morphologists. I have no direct claim to make about the morphologists, except that they were almost unconcerned with taxonomy (Oken and Goethe especially). I think their influence, especially upon Owen, is crucial for the intellectual fervour in which Quinarianism arose.

      There was no stress upon “essential similarity” in that literature, IMO. That has been imposed upon the metaphysics of the day by those who have an axe to grind. “Essential” meant a diagnostic, not constitutive, set of properties, and everyone knew this was subject to natural variation.

  7. Robin Craw Robin Craw

    I’ve just checked Darwin’s 1854 monograph (I know there are earlier barnacle works) and he uses the terms affinity, analogy and homologies- it would require a very close reading of his barnacle works to sort all this out. Re Cuvier as intellectual ancestor are you pushing the Bachelard/Foucault epistemological break line here, i.e. 18th century natural history vs 19th century biology ?
    That Darwin’s concern was taxonomic diversity rather than adaptation I wouldn’t disagree with but then he also wrote:
    “…no one has hardly a right to examine the question of species who has not minutely described many” (Darwin to Hooker Sept 10 1845)

    • John S. Wilkins John S. Wilkins

      No; I think there is no such epistemic break. So far as I can tell the interests change relatively slowly, and such sudden shifts are usually constructions of the historians in my experience. Natural classification was as serious a concern in the 1750s as in the 1840s. Peter Stevens’ work indicates how that went down (an excellent book!).

      “Affinity” is a term that gets most of its traction from the period around 1820-1840, after which it gets supplanted by “homology”. What I can’t track down is to what extent “homology” was in use before Owen’s glossary entry. He uses it as if everyone knew what it meant, but his “Limbs” is the first use in print in English I can locate. On the other hand, I don’t work at the NHM like Dave Williams does, with access to that library.

      I would not attempt to define “species” myself, having minutely failed to describe any, but I can speak to how others have.

  8. Robin Craw Robin Craw

    Most of the people who have described “many species minutely” over the past 250 years don’t even worry about defining species,etc – describing them is so labour intensive and exhausting that most taxonomists seem to go with the idea that a species is what a competent expert in the group says it is (Tate Regan said this more succinctly in 1926).

    • John S. Wilkins John S. Wilkins

      Yes. I discuss that definition in my books 🙂

  9. Robin Craw Robin Craw

    I’ve always thought that systematics/taxonomy should cease struggling for
    “scientific” status and accept that at heart it is an intellectual, intuitive and aesthetic discipline – consider the analogies with art history and writing : catalogues, connoisseurship, expertise, historic collections and literature, importance of a “good eye”, provenance of biological type and other specimens/art objects, etc.
    As regards the intuitive level I’ve always liked the following quote from Charles Lyell :
    “By a kind of innate perception, an intuitive sense, an instinct, even children recognise species; to abandon this creed is like giving up a belief in the existence of the material world because we cannot disprove Berkley’s argument.
    A great naturalist like Agassiz is aware that he cannot always decide the demarcation between some species and others, still he believes in them.”
    (L.G. Wilson (ed) 1970 Sir Charles Lyell’s Scientific Journals on the Species Question, p. 254).
    A recent popular book on systematics/taxonomy addresses this instinctual and intuitive aspect:
    Carol Kaesuk Yoon 2009 Naming Nature : The clash between instinct and science. W.W. Norton & Co., New York & London.
    Darwin’s motivation ? An aesthetic, intuitive and instinctual drive to grapple with the sheer diversity of life.

    • John S. Wilkins John S. Wilkins

      Like all cognitive activities, systematics has many aspects. Some of it really is algorithm-driven, while some might one day be, but is the outcome of classification systems in the head that are not reducible to Bayesian methods or parsimony. Some of it is pattern recognition, which we know can occur in ways not tractable with linear computational techniques. And as you say a lot of it depends on the standards of elegance and aesthetic judgement of the user.

      But that is a first order analysis. The real philosophical question regarding classification is whether it is something that is a process of learning about the world (leaving to one side the issues of conventionalism and psychologism). Those who say that it is not leave it there; it’s all about convenience and psychology for them. Those who say that it is, tend to fall into two classes, one very large, and one very small. The larger one holds that all natural classification is derived from theory, following Duhem. This is pretty well the bog-standard view amongst modern philosophers of science.

      The very small group consists of me and a few of my collaborators (although some do not know this about themselves yet). We argue that classification can be done in the absence of a direct local theory of the domain in question, and that it is a natural classification. We do not deny any of the above, only that they are sufficient. Some classification happens simply because we go out and observe the world, even if no theory tells us what we should be seeing.

      Whether or not this is innate as Lyell said, or there is some other epistemic process in play, is something we can debate. I think innateness is appealed to when we cannot find another justification for what we do, but it is post hoc and unnecessary, I think.

      As to Kaesuk Yoon’s book, I was very underwhelmed with the argument, even as I was impressed at the historical overview.

  10. Robin Craw Robin Craw

    Yes Yoon’s book is light weight.
    So you and your collaborators should be looking towards folk taxonomies, Atran’s work,etc. Or have I got it wrong?

    • John S. Wilkins John S. Wilkins

      Not really. Folk taxonomies are well understood (as you note: Atran and his colleagues). I am interested in the role of classification in the natural sciences.

      • John Harshman John Harshman

        In your search for generalizations across fields, I think you may have gone a generalization too far. Biological classification (if you insist on calling what phylogeneticists do “classification”) is more natural than, for example, any classifications that geologists do, to the extent that there really is a tree of life, while “granite” or “quartz monzonite” are arbitrary divisions of a continuum. Then again, there is a periodic table, but I don’t see a whole lot of structural similarity between that and biological classifications.

        Oh, and I think Robin Craw is confusing alpha taxonomy, the sorting of individual into species, with the whole of systematics.

        • John S. Wilkins John S. Wilkins

          Actually there are several modes of classification in geology; some are compositional, yes, but some (especially the Russian system) rely upon causal history. As you note, some classifications are effectively phenetic in geology, but not all. We are doing a bit of research to exemplify elicidate these differences. Soil science is particularly interesting, and has major differences across national criteria.

          The periodic table is also very interesting. Initially it is simply an empirical summary (of atomic weights). Later, it is refined to be atomic number, due to the adoption of valency theory, which is devised to explain, among other things, the weights. Later still, it gets revamped as quantum mechanical and Bohrian.

          There is a complex relationship between what I call functional and etiological classifications. A functional class gives you inference only on the definienda of the set: if to be a reptile is to be ectothermic, amniotic, etc., then any new instance of the class can be inferred only to have these definienda. Nothing else can be inferred (consider “predator” as another example).

          An etiological classification like a phylogenetic taxon, however, is indefinitely extensible. If you know something is a member of a clade, but nothing else, you know a lot about it by induction; if it’s a member of Accipitriformes (your suggestion!), then you know its general anatomy, physiology, biochemistry, genetic structure, morphology, ecology and so on. You may be wrong about some of these inferences, because of evolution, but by and large the phylogenetic classification encodes a lot of information, based upon your sampling of that taxon.

          A hybrid class, on the other hand, licenses neither definitional inference nor taxonomic induction. If you try to make a functional class phylogenetic, and it happens that some convergence or paraphyly has occurred, then you cannot say for sure very much at all about an as-yet undescribed member of the class. Evolutionary systematics is hybrid classification, IMO.

          Now how this plays out in physics or chemistry is complex, because at some point functional and etiological classifications coincide. In the “general” or “universal” sciences, they are effectively identical (and are replaced, when suitable theory is available, with theory-derived classes). But in the “special” or “historical” sciences, which Whewell (my hero of the day) called paletiological sciences, there is a distinction. General scientific classes do not specify the object entirely, and so we need to appeal to causal histories and their outcomes.

          I take under advisement that I am overgeneralising. As I go through this material, I am revising my claims constantly, but it seems to me that classification does indeed come in the two flavours, and biological systematics is not privileged as the model of classification simply because it started an elaborate scheme first (which, in any case, it did not; the geological/mineralogical schemes are at least as old if not older).

  11. Robin Craw Robin Craw

    No I’m not confusing them-to me species are just one sort of natural group -aesthetic,etc considerations come into play whatever the rank, and whether one is a Quinarian or a cladist.

    • John S. Wilkins John S. Wilkins

      Gary Nelson wrote a piece in which he noted that species are just taxa, back in the 1970s. I agree with him on that. Once you abandon the rankedness of taxa, you no longer make such distinctions.

  12. Robin Craw Robin Craw

    No I’m not confusing them-to me species are just one sort of natural group -aesthetic,etc considerations come into play whatever the rank, and whether one is a Quinarian or a cladist. Theoreticians of systematics seem always to be overanxious about being seen to be “scientific” and “philosophical” – think all those dense and overwritten papers in the journals justifying Hennig to start with.

    • John Harshman John Harshman

      I can’t agree. Systematics these days is a rigorous science, not an art. That day is gone. Aesthetics doesn’t enter into it, other than, perhaps, choosing what name to give a group. And I’m not talking about all that pseudo-Popperian nonsense, but real, operational science. Either you are not familiar with recent systematics literature or you are choosing to ignore the last 40 years or so.

      Further, species are not just another taxon. They’re a special one, because there is a special criterion by which we define and recognize them (though what the criterion is depends on species concept). There is no such criterion for genera, families, or any other ranks. Species also, depending on concept, may not even be clades.

  13. Robin Craw Robin Craw

    I agree with most of what you say JW. And geologists do use the term “genetic”. Sorry about duplication.

  14. Robin Craw Robin Craw

    No I’m not ignoring the last 40 years of systematics literature- I lived through the so-called “cladistic revolution” and was studying that literature even as an undergraduate; as well I corresponded with Gary Nelson, David Hull, Michael Ghiselin and Leon Croizat; and I published papers (in Systematic Zoology[now Biology] between 1979-1989 relevant to systematics/biogeography interface. I have published a systematic monograph on a monophyletic clade (NZ Molytini- large endangered speargrass weevils) with a cladistic parsimony analysis. In 1999 I was senior author of Craw et al, Panbiogeography: Tracking the History of Life (Oxford University Press) that has a chapter on relationship between systematics and biogeography. I’m not posing as an “authority” – but my views are considered and informed.
    With regard to taxonomic diversity and natural groups it all comes down to the “cladistic parameter” (Nelson & Platnick 1981:p.169 et. seq.):
    “[Taxon] A and [Taxon] B are more closely related to each other than they to [Taxon] C”.
    Now Williams & Ebach (2008, Foundations of Sytematics and Biogeography), whom JW has cited as his guide in this discussion stream, have claimed that Goethe first clearly articulated the cladistic parameter pre-Darwin:
    “Consider the act of comparing a cat with a dog. The observer immediately relates both ‘objects’ and their qualities as ‘catness’ and ‘dogness’. Despite their differences, cats and dogs share many characteristics in common. It is, of course, possible to compare any two objects, listing how they are the same and how they differ. The list gains meaning only when a third object is introduced to compare the qualities of the relationships. Comparing a cat with a dog, for example, requires the introduction of a third creature to act not just as a reference point but also as a standard with which to orient the comparisons. For example, it is meaningful to observe ways in which a cat shares greater relationship with a panther than it does to a dog – the same cat-like forms ‘appear’ in the panther. It is not possible to meaningfully compare a cat with a dog unless another creature is present (or present from previous experience in our mind’s eye): “Thus we will not hesitate to suggest a third thing, intermediate between the two…” (Goethe 1995: 124). Goethe- probably for the first time-clearly ARTICULATES the Cladistic Parameter required to understand relationships (Nelson & Platnick 1981)”.
    To me this statement by Williams & Ebach that Goethe “clearly articulates the Cladistic Parameter” is not only historically inaccurate but illogical and nonsensical as well. Goethe is not saying that given two objects/taxa one needs to introduce a third object/taxon[cladistic parameter/three item statement] in order to resolve the question of relationship so that any 2 of the 3 can be shown to be more closely related[i.e. form a natural group] to each other than to the 3rd. Surely Goethe is emphasizing that the third taxon/object is INTERMEDIATE [being, situated or acting between two points, stages, things, taxa,etc]. So in terms of the example given of cat,dog
    and panther Goethe would say no,no…given cat and dog one introduces a third intermediate creature combining “catness” and “dogness”- a sort of cat/dog “hybrid” ,which gives a rigid trichotomous statement : [CAT, CAT/DOG (intermediate), DOG]. This is not the Cladistic Parameter!
    My published claim [1992, Margins of cladistics…, in P.Griffiths(ed.) Trees of Life, pp.68-73] is that the Cladistic Parameter [three taxon/item statement] was first clearly and explicitly stated by Fritz Mueller (most known amongst today’s biologists for discovery of Muellerian mimicry) in 1864 in his book “Fur Darwin” [English translation 1869 as “Facts and arguments for Darwin]. I challenge anyone to find an earlier statement of the Cladistic Parameter that is as clear and explicit as Mueller (he provides
    an empirical worked example from the amphipod genus Melita).

    • John Harshman John Harshman

      I’m glad that you aren’t ignoring systematics, but then I’m at a loss to explain your claims that it’s all an art rather than a science. And I’m afraid I do not understand the relevance to this discussion of your long disquisition on the “cladistic parameter”. Perhaps you can explain.

  15. Jim Thomerson Jim Thomerson

    It is intereting, and I think important, that we speak of speciation events, but not of genusization or familyization events.

  16. Robin Craw Robin Craw

    I understood this discussion to be about taxonomic diversity , classification and natural groups re Darwin’s motivation. The “cladistic parameter” is basic to any discussion of systematics/taxonomy (see e.g. Nelson & Platnick 1981, Williams & Ebach 2008). It was JW (6/12 @1.02) who introduced Goethe into the discussion: “Goethe and Oken almost unconcerned with taxonomy”. This is contra Williams & Ebach who claim (p.31) that Goethe “first clearly articulates
    the “cladistic parameter” pre-Darwin. I showed above that Goethe didn’t clearly articulate the “cladistic parameter”. My published claim is that Fritz Mueller first clearly and explicitly articulated the “cladistic parameter” post 1859 in 1864. Why is this not relevant to this discussion?
    Perhaps JH & JT could tell us what their idea of “a natural group” in taxonomy is, and why systematics is a science (I used to think it was).

    • John S. Wilkins John S. Wilkins

      Although Williams and Ebach are friends, I never quite understood what the cladistic parameter is. W&E have it that it is an equivalence with homology or synapomorphy and taxon. Nelson in 1979 stated it as “efficiency and informativeness” of cladograms. I think it is roughly synonymous with the older, pre-evolutionary, notion of “affinity”.

      But I do not see Goethe doing this; his “homologs” (he of course does not use the term) are formal relations, not taxonomic ones, and he is almost entirely concerned with special homology. Oken I have not been able to find sufficient in English to tell, but the overall preoccupation of the “ideal” morphologists was, well, morphology, not taxonomy.

      I could well believe that Mueller did this, at least verbally, in his Für Darwin; I suspect it may predate Darwin in the linguistic literature (e.g., in Max Mueller).

      • John Harshman John Harshman

        Robin Craw seems to be saying that the cladistic parameter is a three-taxon statement. I suppose one might consider that a sort of logical atom of phylogenetics. Why it’s essential to any discussion of systematics is however beyond me.

        • John S. Wilkins John S. Wilkins

          I think it is, in the sense that if you are going to talk about the logic of a field you have to establish the axioms of that logic. It is interesting that despite 300 years of logical discussion by philosophers and mathematicians, the three taxon point is almost never discussed outside of systematics, not even by Locke or Russell, to name the two most significant workers on “relations”. Pierce is an exception.

          I suspect that when an idea is so deeply embedded in a discipline, its practitioners often fail to even recognise how significant it is.

  17. John Harshman John Harshman

    I find myself unable so far to have any sort of conversation with Robin Craw. Assuming that he’s trying to have a conversation with me, that is. Perhaps he isn’t.

    So. Anyone want to discuss a) whether systematics should be called science or art; b) whether species are special, i.e. not just taxa like any others; c) what Colin Patterson meant by “homology” and “synapomorphy”; or d) whether non-evolutionary definitions of those two terms (in biology, that is) are a good thing?

    And to answer Robin, I think a natural group is either a clade or a species, sometimes both. Systematics is a science because it works by rigorous testing of hypotheses, using methods and data to produce replicable results, and relies on logical argument to get from results to conclusions. It most definitely isn’t a matter of personal taste. And I think that’s as true for alpha taxonomy as for phylogenetics. Why would anyone think it isn’t a science?

    • John S. Wilkins John S. Wilkins

      I agree that species aren’t clades, but they are taxa like clades (and nothing above a species is a taxon unless it is a clade).

      A taxon is not necessarily monophyletic, but that is because monophyly is predicated upon the basic taxon: species, and species are not always monophyletic. Respeciation, hybrid speciation, paraphyly (once a species has arisen within a species) and so on can make a species aphyletic.

      Hence, to do cladistic taxonomy or systematics, one has to assume the existence of species (arguably; I am wondering if you can simply use specimens without them being surrogates for species, qua Ward Wheeler and Fred Pleijel) and they are the foundation for all natural groups.

      Me, I think that species are not phylogenetically derived objects. They are simply phenomena. “natural group” applies to a disjunct: either a [phenomenal and possibly theoretically identifiable] species, or any monophyletic group of species.

      The science/art distinction is not helpful, in my view. “Art” is so ill-defined as to be meaningless. Scientists have “art” in the sense they have tacit knowledge and techniques and standards that are not ye expressed, and expression of which wouldn’t be helpful anyway. All human activities do this. Classifying is an art in the sense that the training up of a classifier is something done by example (on training sets) over a long period. But there is nothing important captured in that case that is not also true of artificial neural net classifiers being trained on training sets; it is just that the specification would be intractable, and won’t help the birds fly (Feynman reference).

      As to Colin’s original meanings; I have to go dig the literature for that. Some other time.

      • John Harshman John Harshman

        As usual, I don’t agree. Some species are clades, and some are not. All taxa except species are supposed to be clades, if we’re doing it right. Nor do I think that species are necessary for cladistic taxonomy, though they’re certainly handy. Hybridization of course creates exceptions to our nice neat clades, and tokogenetic relationships are a form of hybridization. Perhaps we do agree, and you were just being a bit loose with the wording.

        Been a while since I read Patterson, but perhaps we can make a distinction between hypotheses of homology and homology itself, or hypotheses of synapomorphy and synapomorphy itself. What we have are hypotheses, and they begin merely as statements of similarity. When tested by congruence with other hypotheses, they may become confirmed to a greater or lesser degree, but the underlying definition, the asymptote of confirmation, is the condition in the common ancestor.

  18. Robin Craw Robin Craw

    Intentionality and transformation are two essential criteria of both art and science which also rely on experimentation, the work of peers and “case studies”. One can traverse this path from biology, chemistry and physics to art and craft by way of sense perception and the dual roles of “experience” and “experimentation”. Look at the similar situations of the science laboratory and the art studio as sites of sensory and scopic practices.
    Science and art – alternating attitudes of a society of mind functioning on either side of imagination and rationality- a logic of the complement in which the boundaries and limits run through individual lives rather than between persons and institutions. Identity as either scientist or artist becomes fractionated and fractured into a hybrid form.
    As art has become more textual, science has swung towards the pictorial. Biology and chemistry, in particular, have become obsessed with representation and the sign as well as becoming inceasingly pictographic.
    The endless hybridizations, repetitions, transformations, and translations immanent and inherent in any artistic or scientific practice breach the boundaries and closures upon which concepts and constructs such as “art”
    and “science” depend.

    • John Harshman John Harshman

      If anyone can explain it, I’ll give him sixpence. I don’t believe there’s an atom of meaning in it.

  19. Robin Craw Robin Craw

    Only 6d? Plenty of people have read it (it’s a series of snippets from an essay I wrote as a supplement for an art exhibition) and understood it. Do you follow contemporary art at all JH?
    As to affinity with Darwin affinity becomes genealogical relationship.
    Species are just the smallest definable taxa, they are not some privileged real unit while other taxa (e.g. genera,families) are artificial. The question is what is a “natural group”.
    Geologists seek natural classifications too, e.g. in descriptive stratigraphy where they describe “groups”, “formations” and “members”; and they have a nomenclatural code (not unlike the biological ones) and designate “type sections”.

  20. John Harshman John Harshman

    This is the sound of me giving up in sheer frustration. And I like to think I have above-average patience.

  21. Robin Craw Robin Craw

    Why? I’ve raised “serious” scientific issues to do with natuaral groups in both
    biological and geological “sciences” . Can’t you hack the pace man.

Comments are closed.