23 August 2008

What is it that you hate about the PhyloCode?

Mention of the PhyloCode can incite some pretty strong emotions among biologists. Some are supportive, but there's a lot of vitriol out there. Consider, for example the title of Carpenter's (2003) paper: A critique of pure folly. Yes, a paper with that title actually got published in a scientific journal. Why the hate?

The thing is, it's sometimes hard to pin people down on exactly why they dislike it. Part of the problem is that the PhyloCode is surrounded by misconceptions. Much of what's been written about it is shockingly ignorant, considering that the code is freely available online and not horribly long (certainly shorter than any of the other codes). Consider Benton (2000), who seems to have not even read the then-current draft and proceeded to write an entire review based on what he imagined the PhyloCode to be like.

This is not to say that all critiques are uninformed. But of those that are, I find that they fall into two camps. They either raise points that have since been addressed in later drafts, or they are simple matters of taste.

As an example of addressed issues, the critiques of how the PhyloCode was going to handle species are now defunct, since it won't cover species. I myself raised dozens of smaller issues to the authors of the code, who then either emended the draft of the code or convinced me that no change was needed.

And as an example of difference of opinion, some people don't see the utility of the "crown clade convention" adopted by recent drafts. Some people just don't like phylogenetic nomenclature, period. The rank-based systems (which, incidentally, should not be called Linnaean systems, per de Queiroz [2005]) do well enough for many people's purposes, and they don't see a reason to change. Of those who like phylogenetic nomenclature, some don't see the advantages of a centralized approach, and would rather let it grow freely (e.g., Sereno [pers. comm.]).

That's all fine; it's good for people to point out flaws if they are later addressed, and it's understandable that not all people would agree on matters of opinion. But I see so much misinformation out there that I think it has to be the source of much of the dislike out there. One popular article is titled What if we decided to rename every living thing on Earth?, when the PhyloCode has always advocated using existing names when possible. Pickett's (2005) report on the first ISPN meeting is titled The new and improved PhyloCode, now with types, ranks, and even polyphyly, when the PhyloCode has never had types (only specifiers, which are similar but operationally different), has always allowed ranks, and has never allowed polyphyly.

So let me put a question out there: If you dislike the PhyloCode, what is it about it that you dislike? And are you sure it's something actually in the code?

17 comments:

  1. As an example of addressed issues, the critiques of how the PhyloCode was going to handle species are now defunct, since it won't cover species.

    Sorry, but this is a dodge. Because the current status quo for species names under any code requires the inclusion of a rank-based taxon in the form of the genus as part of the name, there's no way the PhyloCode can possibly achieve its goals without at least tackling species names.

    ReplyDelete
  2. Phylocode is crap. Long Live Linneaus!

    That reminds I have read a phylocode draft since 2004. I'm sure its still crap though ;p

    ReplyDelete
  3. ...there's no way the PhyloCode can possibly achieve its goals without at least tackling species names.

    See, this is what I'm talking about. Nobody reads the damn thing! An entire article (Art. 21) was added a couple of years ago addressing exactly that issue.

    Also, note that they only way that the PhyloCode uses species is as proxies for their type specimens (Note 11.1.1), so the current draft doesn't really need to bother with what, exactly, a species is. Then it just relies in the rank-based codes to tell what those specimens are. In it s current form, it's just a clade code (Art 1.1).

    Some people don't like the fact that the PhyloCode is no longer going to cover species, and I think that's a valid complaint. I also think it's not realistic for the PhyloCode to do a better job than the current codes any time in the near future, though, so I support dropping species from the code.

    ReplyDelete
  4. Phylocode is crap. Long Live Linneaus!

    Read the de Queiroz paper mentioned above. Phylogenetic nomenclature as laid out by the PhyloCode is actually more consistent with how Linnaeus himself used ranks (i.e., without standardized suffixes) than the rank-based codes are.

    That reminds I have read a phylocode draft since 2004. I'm sure its still crap though ;p

    Thanks for completely validating me! Once again, someone hates it but hasn't read it and refuses to state why they hate it.

    ReplyDelete
  5. See, this is what I'm talking about. Nobody reads the damn thing! An entire article (Art. 21) was added a couple of years ago addressing exactly that issue.

    Sorry, I've read the article in question, and it's still a dodge. "A prenomen is the first part of a species binomen. A prenomen has no necessary tie to any categorical rank under this code." In other words, the "genus name" serves no purpose in the context of the PhyloCode, but we're going to try to make people feel better by allowing a pseudo-genus name sop.

    The species that is referred to as Vultur gryphus or Vultur gryphus Linnaeus under the ICZN may be referred to under this code as Vultur gryphus or Vultur gryphus Linnaeus or gryphus Linnaeus 1758. Any of these forms of the species name may be associated with additional clade names to indicate hierarchical relationships (see Rec. 21A); for example, Aves/Vultur/gryphus Linnaeus 1758 or Aves/Vultur gryphus Linnaeus.

    Or, for that matter, Aves gryphus. If the "prenomen" is divorced from rank concepts, there is absolutely no requirement for favouring any specific one of the higher taxon names that species appears in to function as the prenomen. This has some pretty significant consequences for questions of homonymy - can you imagine how many Angiospermae officinalis there must be out there?

    Note that I am not here trying to defend the current codes over the PhyloCode - I am well aware of the limitations of the current codes in this respect. I'm saying that the PhyloCode, however much it might like to, cannot ignore species names.

    ReplyDelete
  6. Sorry, I've read the article in question

    Sorry I assumed otherwise. Now we're talking!

    In other words, the "genus name" serves no purpose in the context of the PhyloCode, but we're going to try to make people feel better by allowing a pseudo-genus name sop.

    It's not about making people feel better; it's about allowing species names to be formed as they would be formed under the codes that actually govern them. And note that you do have to use a genus name when coining a species, per the rank-based codes.

    This has some pretty significant consequences for questions of homonymy - can you imagine how many Angiospermae officinalis there must be out there?

    That form is intended for use when there is no genus-derived clade that includes the species. It's a way to clarify the phylogenetic position. Since "Angiospermae officinalis" clarifies nothing, there's no reason for anyone to use it. A more proper usage might be for species without a proper genus, like, say, Tetanurae hesperis (Waldman 1974) (="Megalosaurus" hesperis Waldman 1974).

    Perhaps something should be added to caution against the usage of ambiguous binomina? I think that would make a good recommendation.

    I'm saying that the PhyloCode, however much it might like to, cannot ignore species names.

    Kind of hard to make that claim when there's an entire article titled "Species Names"! Maybe it would be better to say that you don't like how they handle species names, in which case I'd like to know how you would have them handle species names. I'm certain there are better ways, but I'm not convinced that there are any more realistic ways, for the time being.

    ReplyDelete
  7. That form is intended for use when there is no genus-derived clade that includes the species. It's a way to clarify the phylogenetic position. Since "Angiospermae officinalis" clarifies nothing, there's no reason for anyone to use it. A more proper usage might be for species without a proper genus, like, say, Tetanurae hesperis (Waldman 1974) (="Megalosaurus" hesperis Waldman 1974).

    But as you well know, "intended" is entirely different from "actually happens". And what about names originally coined under the PhyloCode, as opposed to converted after the event? These have no "genus-derived" name by definition.

    Perhaps something should be added to caution against the usage of ambiguous binomina? I think that would make a good recommendation.

    But in effect, completely pointless because it wouldn't mean anything in the practical sense. Imagine two sister-species with the exact same species name - all the author needs to do is invent a meaningless prenomen for at least one and bang, you've got an unambiguous binomen for both. It's a system that's just begging to be abused.

    Which brings up another question that wasn't addressed - what becomes of names replaced under the current codes due to homonymy? Particularly secondary homonymy?

    Maybe it would be better to say that you don't like how they handle species names, in which case I'd like to know how you would have them handle species names.

    Maybe it would be. I would say that some form of uninomial system is the only way that makes any kind of sense under the principles of the PhyloCode.

    ReplyDelete
  8. But as you well know, "intended" is entirely different from "actually happens".

    Yes, but I seriously doubt we'll see any papers referring to "Angiospermae officinalis". (Still, it is worthy of a recommendation, I think.)

    And what about names originally coined under the PhyloCode, as opposed to converted after the event?

    Well, since species can't be coined under the PhyloCode, this is a non-issue. Remember, the PhyloCode may reference species (as a proxy for type specimens), but it only governs clades.

    Imagine two sister-species with the exact same species name - all the author needs to do is invent a meaningless prenomen for at least one and bang, you've got an unambiguous binomen for both.

    Prenomina have to be genera under some rank-based code, so it's no more "meaningless" than the current situation. And the current codes have the same sort of problem (e.g., there are two Hadrosaurus minor).

    Which brings up another question that wasn't addressed - what becomes of names replaced under the current codes due to homonymy? Particularly secondary homonymy?

    Homonymy of species names is handled by the rank-based codes. And in any case ambiguity can always be resolved by a citation in tandem with a sufficiently exclusive prenomen.

    I would say that some form of uninomial system is the only way that makes any kind of sense under the principles of the PhyloCode.

    I am in agreement here. But think what a huge task this is and how much opposition the PhyloCode already faces. It's simply not realistic, at least at this point in time. (Look how long it's taking just to implement the clades-only code!) Should the PhyloCode catch on, though, it's certainly something that could be addressed down the road.

    Clades first; species may follow, if need be.

    ReplyDelete
  9. David Marjanović01 October, 2008 08:30

    Haaah, my ancient rant is still up and apparently useful. :-)

    Im my experience so far, nobody who understands phylogenetic nomenclature (count them, zero) dislikes it. It's very similar for the PhyloCode.

    Admittedly, species names were a real problem, but then came Article 21 (less than two years ago, though).

    Perhaps something should be added to caution against the usage of ambiguous binomina? I think that would make a good recommendation.

    Good idea.

    Which brings up another question that wasn't addressed - what becomes of names replaced under the current codes due to homonymy?

    Of course it was addressed. The PhyloCode will not govern species names. Synonymy, homonymy, and all the rest of governing is left to the rank-based codes.

    there are two Hadrosaurus minor

    Which ones are those?

    ReplyDelete
  10. (If only the Dinosauricon were still up....)

    Hmm, I might be mistaken.

    The original is Hadrosaurus minor Marsh 1870. There appears to also be a Hadrosaurus minor Colbert 1948, but it might just be Colbert referring new material to H. minor Marsh 1870. I can't tell without seeing the original.

    I had thought there was another species named minor which had been named under some other genus and later transferred to Hadrosaurus, but now I cannot find it. I might be thinking of Trachodon minor Hay 1902, but this is just H. minor Marsh 1870 transferred to Trachodon.

    ReplyDelete
  11. Upon reading this and the newly updated phylocode, I take back what I said at another place. This is really satisfying, apart from issues with species names not being covered.

    To give an example of how this makes me feel, I am currently enamoured with the fungus gnats. Now, under the rank based classification, I'm not really sure what to call this clade. Classically, the fungus gnats were Mycetophilidae, ranked family level in the Linnaean system. Someone decided that the clade of the black winged fungus gnats (now family Sciaridae) needed to be split off due to apomorphies to show of its importance in a family status level. Subsequently, all the other larger clades in the classical Mycetophilidae are being split off because the first split makes the family designation paraphyletic. Now if I want to talk about fungus gnats overall what do I use? Sciaroidea? Mycetophilidae sensu lato? The way these names get changed around with rank changes, and seen as greater or lesser importance due to this, is confusing to say the least. Phylogenetically, the clade still is real, it still includes all the classical Mycetophilidae, and I'm sure that few people would disagree this is not a monophyletic grouping. But the name and rank changes make it very difficult and confusing to talk about them. Fixing the clade names in place (unless we learn a clade isn't phylogenetically sound) would be so helpful.

    A thought I had previously that bothered me, was that this would encourage "splitters", but now that I consider it, splitting is only a problem under a /rank based/ system due to the changing of names and rank, but in a phylogenetic nomenclatoral system, its not changing names but rather inserting new names. The governing clade stays the same with new names inserted underneath.

    ~Kai,

    Sighs a major sigh of relief.

    ReplyDelete
  12. This whole mess sounds so similar to the barcoding fiasco/arguments. Instead of "Barcoding is the cancer killing traditional taxonomy vs barcoding is the savior" its "phylocode is the cancer killing traditional taxonomy vs. phylocode is the new wave". Except, I find people who talk about phylocode much less prone to evangelism than the barcoding crowd. Then again, the former is a movement towards a satisfying reworking of taxonomy and the latter is a taxonomic tool that is sometimes heralded as the be all end all in taxonomic work.

    meh, so its not as good of a comparison as I thought. the arguments still feel similar. The "way we've always done things" versus "the way of the future", often with no middle ground. Not commenting on the content of this blog, just on the back and forth in the peer review.

    ReplyDelete
  13. Kai, thanks, that's very encouraging! (Hopefully the species names thing will be addressed further some day, but right now it would be biting off more than the ISPN can chew.)

    Other Anonymous Person, what in science (and the culture around science, including nomenclature) doesn't proceed this way?

    I think there is a middle ground in this case, though, and it's spelled out in the PhyloCode's Preamble: 3. This code may be used concurrently with the rank-based codes. You don't have to choose one; you can use both. (I'll admit there is some potential for conflict, but hopefully it should be minor.)

    ReplyDelete
  14. It may proceed that way, but that doesn't mean it has to be nasty. are enough scientists really that petty?

    "other anonymous person"

    ReplyDelete
  15. Anonymous overload!

    Well, scientists are just big stinking primates, after all....

    ReplyDelete
  16. Instead of splitters, use of the PhyloCode could suffer from "super-namers" - those who insist on labelling every node on a phylogeny with a name, no matter how poorly supported - and a number of its detractors have complained against it this point. I suspect that it is true that the PhyloCode may lead to more names floating about. Personally, though, I don't see this as a problem, because you can always quite happily ignore those names you don't happen to find useful at the time.

    And as for whether rankists or PhyloCodists are more prone to evangelism, my impression is that it depends largely on which side you happen to support yourself ;-P

    ReplyDelete
  17. Chris, that's why I don't see it as an issue. One of my professors said that names are useful when you want to talk about something you find interesting. If a particular node isn't the subject of your research then it's not important to talk about it.

    Also, at least we are talking about rank versus clade based systems instead of still using phenetics. :)

    ~Kai,

    Knows well that primates plus ideas equals trouble. ;)

    ReplyDelete