Talk:Phylogenetic nomenclature

All those tags
Phew… that's way exaggerated. Let's pick the first example: click on the link to the article on the PhyloCode, and you will find the PhyloCode website, which states in no uncertain terms that the PhyloCode is intended to regulate phylogenetic nomenclature. For some other things, like the growing number of biologists dissatisfied with traditional nomenclature, I could cite 100 papers if I took the time. David Marjanović 18:39, 2 June 2007 (UTC)
 * One or two would be just fine. But this is an encyclopedia. The sources need to be available in this article, not through links. That a grwoing number of biologists are dissatisfied should ideally have at least 2 or 3 refs, to give an example of a 'number' of biologists who hold this view. With no ref, statements like that can seem POV and unencyclopeadic, more like a web site about PhyloCode. Dinoguy2 03:07, 3 June 2007 (UTC)

My latest edit
As you have probably already seen, I left two incomplete sections. I will try to complete them over the next… say, weeks. I only tackled the most urgent problems. To wit: David Marjanović 21:49, 22 August 2007 (UTC)
 * Sorry, Stemonitis! I managed to overlook your edit. It is very important to remember that, contrary to your edit, names, not taxa, are defined. Clades exist, and we discover them. Names don't exist in nature; they are conventions that we get to define. I should have fixed this glaring error in the first sentence much earlier.
 * I must look up if Goodrich is Edwin. I think he is.
 * I removed "Category: Phylogenetics", because phylogenetic nomenclature only uses whatever the results of phylogenetics are; it is not part of phylogenetics.

My even laterer edit
On the meaning of "some authors", I have a paper by Benton and probably someone else in mind. I'll look for it and cite it then. I'll also look up and cite the paper that describes the Phylogenetic Diversity Index.

On another note, the article looks unbalanced. I can't help it: I've never encountered anyone who demonstrably understands phylogenetic nomenclature but doesn't like it. Can someone else help…? David Marjanović 13:18, 26 August 2007 (UTC)
 * What do you mean by "demonstrably understands"? It seems pretty dodgy, not to mention unencyclopedic, to say the people opposed to total abandoment of taxonomy in favor of phylogenetics simply don't understand the latter.
 * That's why I didn't put that into the article, as you can guess. :-) David Marjanović 13:17, 15 September 2007 (UTC)
 * Juat typing "objections phylocode" into Google will give you a few hits from published literature.
 * I know – but all of them are, in my experience, misunderstandings. I have a paper in preparation on the latest few. David Marjanović 13:17, 15 September 2007 (UTC)
 * There is still too much subjective editorializing in th article: statements like "The problems is evident" (to who? The editor, obviously). The whole section "Perhaps most emportantly" (sic) consists of obviously biased interpretation based on papers by proponents of ranks.
 * I need to cite more. (Thanks for catching my typo!) David Marjanović 13:17, 15 September 2007 (UTC)
 * None of the papers cited (or currently, not cited) in this section object to the use of ranks, in fact they do the opposite,
 * An important point is the distinction between "use of ranks" and "obligatory use of ranks". That's what much of the article is about, and also what the references are about: they all argue strongly against the latter. For the fifth time: Ranks are allowed in phylogenetic nomenclature, but they are not part of the rules – they don't influence which names can be used, they don't influence which taxa get which names, and they don't influence the spelling of names. Here, read the thing again. David Marjanović 13:17, 15 September 2007 (UTC)
 * and the editor has used their positions to demonstrate why he or she finds the use of ranks problematic. This is unacceptable for an encyclopedia: i'm going to go ahead and revise some of this actually. Dinoguy2 00:37, 27 August 2007 (UTC)
 * In the process you introduced a few glaring mistakes, like deleting "obligatory" from one of the headlines, and rearranging a sentence so that it now seems to say that putting Aves into Reptilia requires Aves to be a genus, even though this is only the case if Reptilia is a class. (If Reptilia were in the process inflated to a domain or something, the problem wouldn't arise, at least not to this extent.) I'll try to fix them now. Thank you for several stylistic improvements and for changing "Advantages" to "Comparisons". David Marjanović 13:17, 15 September 2007 (UTC)


 * So, done.
 * I hope it is now clear that Goodrich did not coin the name Sauropsida. He didn't hide that fact: he cited Huxley.
 * I can't cite anyone for reasons why Goodrich's suggestion hasn't become mainstream because nobody seems to have written about the issue; as far as I know, only I have a paper on this in preparation.
 * I have some of the papers I cited as pdfs; if you want them, just tell me your e-mail address.
 * Also, if you can find someone who works on Mesozoic dinosaur phylogeny and doesn't use PN, please tell me. I follow that field and haven't found anyone in the last 10 years. Using ranks (as some Chinese workers on Mesozoic birds do) does not count, as I hope to have made clear. David Marjanović 15:52, 15 September 2007 (UTC)
 * I just added another reference and removed the "citation style" template because all citations now use the same style, as far as I can see. If you find a counterexample, tell me, or better yet, fix it. David Marjanović 22:20, 15 September 2007 (UTC)


 * Thanks to an anonymous editor for correcting "tenants" to "tenets"! That's something I, you know, "actually knew"… David Marjanović 23:11, 16 September 2007 (UTC)

PN vs PT vs Cladistic taxonomy, etc.
Is it possible to better sort out some of the overlap between these articles? This one obviously deals with phylo nomenclature, but phylogenetic taxonomy redirects to Cladistics (which, strictly speaking, is the method by which relationships are found using a computer program, not the classification system derived from that method. That article even contains a section of "Cladistic taxonomy". Doesn't CT=PT=PN, PN being the more correct name? Dinoguy2 00:37, 27 August 2007 (UTC)


 * Yes. Entirely correct.
 * (Except that a computer is not strictly necessary for cladistics; Hennig did his by hand. A computer only makes it possible to use reasonable numbers of taxa and reasonable numbers of characters and still find all most parsimonious trees in seconds to weeks rather than years.) David Marjanović 13:26, 15 September 2007 (UTC)


 * While I am at it, let me say that the Cladistics article is full of problems. It's so offputting that I don't dare lengthen its vast discussion page. The smallest is that it says "synapomorphy" when "autapomorphy" is meant – Hennig found it great fun to invent new technical terms, so he introduced "autapomorphy" (auto- = self) for any innovation of one taxon and "synapomorphy" (syn- = together) for the shared innovations of sister-groups. I'll give that article a lengthy treatment, perhaps next weekend, or the one after. David Marjanović 16:09, 15 September 2007 (UTC)
 * Now that i'm looking, all the cladistic/phylo articles are an absolute wreck. Check out the "See slao" sections--most of those are redundant with each other and could easily be merged to make one or two articles. Well, not "easily", that's one big project... Dinoguy2 09:31, 16 September 2007 (UTC)
 * Click the link in this article to "Linnaean nomenclature"! That's a redirect to an even more absolute wreck. Poor Linnaeus. He has really deserved better than that. David Marjanović 23:15, 16 September 2007 (UTC)

Note to self
The third paper on this page is online and freely accessible in its entirety, and even in HTML, so I should cite it at every occasion, and then some. If anyone else wants to do that to save some of my limited time, feel free… David Marjanović 23:08, 16 September 2007 (UTC)

Species problem
A more in-depth discussion of species under PN could be useful. Especially in taxa with good dispersal capabilities, evolution does not require species (if not defined under a phylogenetic species concept - which has nothing to do with PN however) to be monophyletic in any meaningful way, especially to molecular systematics.

Any allopatric and reproductively isolated population derived from a still-extant progenitor population that continues to evolve as a single entity presents a problem for PN. There exist various approaches to deal with this problem, but it is a major point of contention in PN. Discussing this in detail would underscore that PN (in the PhyloCode sense) is a brand-spanking-new topic (albeit with a long and dignified ancestry). At present the article seems too set and done, suggesting that there are two monolithic camps and that's that. But this is not correct; many taxonomists as of 2007/2008 seem to prefer to keep out of the debate and simply use a sort of "expanded" Linnean system allowing unranked and basal taxa and don't seem too unhappy with that (the ICZN permits that, as it only mandates a small set of ranks and leaves everything else to decision by consensus). Dysmorodrepanis (talk) 21:10, 3 February 2008 (UTC)


 * The PhyloCode will not regulate species names (at least not in its first edition). To name species, you have to use the rank-based codes, and you can use whatever species concept you like. The PhyloCode will only regulate clade names; under most species concepts species are not necessarily clades. David Marjanović (talk) 15:44, 7 June 2008 (UTC)

Editorializing
There is some really inappropriate commentary going on in this article. A few examples just gleaned from things added/removed in the most recent edits:

Perhaps most importantly, ranks encourage the misperception...
 * Who says this is most important? Who says the ranks encourage misperception? The cite is from a study that treated them as if they were meaningful. The editor is apparently disagreeing with peer-reviewed, published usage. If this isn't OR I don't know what is.

Note that this is probably not a particularly good example:
 * Again, says who? Obviously this is the opinion of the editor.

Explaining this stuff is one thing, but it should be done in a completely citable way. Any argument that has not been made in print, no matter how sound, should not be included. If these points have been argued in print, they should be phrased as "such and such argued, in the year x, that that this example (previously used by so and so) was faulty." I've half a mind to be bold and go in to decimate this thing myself, it's possibly my least favorite article after Prehistoric Park ;) Dinoguy2 (talk) 21:57, 3 February 2008 (UTC)

Edit 7 June 2008
'''It looks like someone tried to have a discussion with me in the article itself. Please let's have it here. That's what the discussion page is for.'''

I have changed too much to fit into an edit summary:

David Marjanović (talk) 20:35, 7 June 2008 (UTC)
 * I have deleted the second paragraph of the introduction, because it lacks any context and is wrong anyway (as explained in the section "Extensions", PN can be used to define paraphyletic and even polyphyletic taxa – only the PhyloCode will forbid that).
 * I have changed "traditional" to "rank-based". This is more meaningful and avoids POV implications like ancient vs modern.
 * In the section "Lack of obligatory ranks" I deleted the paragraph after the quote, the one that began with "Note that this is probably not a particularly good example" because it is… well… wrong. Because they have different ranks, Homininae and Hominidae are never treated as synonyms by the ICZN, even if they have the exact same contents under a given classification, and don't compete for priority.
 * In the next paragraph, someone added a hidden comment "did any taxonomist really think this way since Darwin?" on the issue of ranks misleading people into believing ranks are somehow real at some level. Astonishingly enough, this happens all the time, as documented in the references later in that sentence. The most astonishing case is this paper by Benton (the APP website is currently down, so the Google cache is here), which says in the same sentence – in the abstract! – that ranks are not real and that ranks are "valuable surrogates" for actually quantifying biodiversity.
 * I have deleted the Nixon & Carpenter 2000 ref because it's not only way out of date, but also because it is chock full of misunderstandings that have been corrected again and again in more recent literature (much of which is cited). I hope this does not come across as POV. The latest papers on PN are de Queiroz (2007) and Laurin (2008), both of which argue strongly for it; there does not seem to be a paper that argues against it and is similarly recent. If one ever comes out, please add it. Benton (2007)… well, see above.
 * I have deleted the sentence "Taxa of the same rank can only be considered equivalent if they are placed in the same taxon one rank higher, and only if the taxonomic treatment is phylogenetically sound." That's because it's still wrong. "Equivalent" in what sense? In terms of contained number of species? No. In terms of included morphological or genetic diversity? No. In terms of age? No – sister-groups do fulfill that criterion, but being placed in the same taxon of the next higher rank is not the same. In terms of ecological dominance? No…
 * The same holds for the first half of the following paragraph. There isn't even such a thing as "phylogenetic taxonomy" – did you mean the principle of naming only monophyletic taxa except for species?
 * The second half does have a valid point: mandatory suffixes indicate nesting (although this works even better (Google cache) without mandatory ranks…), but this belongs in the PhyloCode article, which will set rules for PN, not in the PN article itself. So I deleted that, too.
 * I have reworked the introductory paragraph to the "History" section and removed the section-stub tag.
 * I have added lots of references.


 * The article still reads to me, from the point of view of a serious but non-professional botanist, too much like an argument for PN and not a description/discussion of PN. There may not be papers arguing against PN in favour of rank-based approaches, but you also need to look at what is actually being done ("do what I do, not what I say"). So e.g. if you look at the field of angiosperm classification, dominated by the APG, you find that both clade-based and rank-based names are considered important. Circumscribing Linnean Families and Orders based on phylogenetic evidence is a major activity. Alongside this goes free use of informal clade names. (It's too early to tell, but I suspect that in most fields of biology, this kind of mix-and-match between formal Linnean ranks and informal clades is the likely outcome, rather than PN purity.) At least in botany, I can cite plenty of authors who by their actions obviously consider rank-based nomenclature worthwhile; they don't seem to bother to write papers which say so, they just do it. Peter coxhead (talk) 00:30, 21 December 2010 (UTC)


 * Returning to the point, why is PN an "alternative" to rank-based nomenclature as the first sentence says? This seems to set up an opposition from the start. Perhaps it's really an "addition to" or a "complement to". Peter coxhead (talk) 00:35, 21 December 2010 (UTC)

Philosophy
According to the Philosophy section, "phylogenetic nomenclature follows the nomenclature in other sciences in trying to define its terms as precisely as possible. . ." On the contrary, scientists require that terms be defined just well enough for the purposes of research, application, and communication within their fields, seldom as precisely as possible. Economists, for example, use an admittedly imprecise concept of inflation, defining more exact measures only as the need arises. In physics, the concept of energy is imprecise and has been repeatedly extended with an eye to preserving the conservation law. Even the PN edifice rests on an ill-defined notion of species; if its supporters sought maximum precision, they would either formulate a definition or else adopt one that has been proposed. As David Marjanović notes above, the Phylocode allows you to use whatever species concept you like. Peter M. Brown (talk) 17:09, 3 August 2011 (UTC)

More than physics, the approach of Whitehead and Russell's Principia Mathematica is similar to that of phylogenetic nomenclature. As the latter defines taxa in terms of the undefined notion of species, so the former defines mathematical entities in terms of the undefined notion of an atomic proposition. Peter M. Brown (talk) 18:30, 7 August 2011 (UTC)


 * After I began this article, the PhyloCode gave up on species, at least for now. It will only regulate clade names. Clade names will not be defined with species as specifiers, but with specimens as specifiers. (It will be allowed to use species names as shorthand for their type specimens.) So, PN does not rest on any notion of species; it rests on the notion of "specimen in a collection", which is straightforward enough. David Marjanović (talk) 10:42, 22 September 2012 (UTC)


 * I was unaware of the shift. The Phylocode article should be updated, not only by deleting the claim that species may be a specifier, but also by noting that an "ancestor" for PN purposes must be an organism or a breeding pair, not a species. Peter M. Brown (talk) 23:01, 24 September 2012 (UTC)
 * Letting individuals found clades means that the structure is no longer hierarchical, doesn't it. Clades can now overlap. Peter M. Brown (talk) 01:20, 25 September 2012 (UTC)

History
There are several things in this section that might deserve modification: —Preceding unsigned comment added by 160.111.254.11 (talk) 22:09, 11 June 2008 (UTC)


 * I'm not sure that the statement "in accordance with the only code of biological nomenclature that existed at the time" with reference to Haeckel 1866 is correct in that the international codes of botanical and zoological nomenclature didn't really exist at the time.


 * The statement "today, the term "cladistics" is only used for the method of phylogeny reconstruction" is incorrect, as the term is also used for the school of taxonomy in which only monophyletic groups (clades) are recognized as formal taxa.


 * It seems a bit artificial to equate the origin of phylogenetic nomenclature with the first journal article to present a phylogenetic definition (Gauthier, 1986). Such definitions had been "published" eariler in both a thesis (de Queiroz, 1985) and a dissertation (Gauthier, 1984), and reference to the possibility of formulating such definitions in theory had also been made (Ghiselin, Rowe).  The formulation of phylogenetic definitions originated in discussions that Jacques and I had in 1983 while we were working on several different projects that later included phylogenetic definitions, including my 1985 Master's thesis, his 1984 Ph.D. dissertation, the 1986 paper (Saurischian monophyly ...) derived from Jacques' dissertation, and two chapters in the book Phylogenetic Relationships of the Lizard Families (Gauthier, Estes, and de Queiroz, 1988; Estes, de Queiroz, and Gauthier, 1988).  Some of this history is described in the Preface to the PhyloCode.


 * The statement that no entomologists use phylogenetic nomenclature is not entirely correct (see Caterino, 1998 and Bethoux, 2007).


 * The last sentence implies that Gauthier was involved in the initial drafting of the PhyloCode. In reality, it was just Cantino and de Queiroz, though Donoghue and Gauthier urged us to do it.

Kevin de Queiroz 160.111.254.11 (talk) 22:51, 11 June 2008 (UTC)

Should the Clade types diagram be here, in Definitions?
While it's relevant, the diagram is almost completely disconnected from the text. Its conventions are not explained and the terms that it is supposed to clarify are not used except that "branch-based" occurs once in passing, deep in the History section. Without an explanation of the term, the latter occurrence is unhelpful. The sentence involved can easily be revised to eliminate it.

The diagram's caption mentions "the apomorphy mentioned in the apomorphy-based definition" but there is no apomorphy-based definition in the article.

The diagram is used to excellent effect in the Phylogenetic definitions of a clade section of the Cladistics article. I suggest that it be eliminated from Phylogenetic nomenclature and that a hatnote be added to Definitions:


 * For types of phylogenetic definitions, see Phylogenetic definitions of a clade.

Things would be different if someone were prepared to enhance Phylogenetic nomenclature to make real use of the diagram. As this is already done in Cladistics, though, that would be unnecessary effort. As things stand, the diagram is clutter. Peter M. Brown (talk) 00:10, 22 August 2011 (UTC)

More on definitions
Nomenclature issues do not belong in the cladistics article. Cladistics = the science of phylogenetics, not nomenclature. Clades are not defined, they exist independently of our brains; names are defined. David Marjanović (talk) 10:51, 22 September 2012 (UTC)


 * Sadly, most of the articles around this area need considerable improvement. Feel free! Peter coxhead (talk) 10:47, 23 August 2011 (UTC)


 * On re-reading the article, I'm struck again by how confused and hence confusing it is. Consider the use of words like "definition" and "meaning" in the Definitions section. "Under the rank-based codes ..., names themselves do not have definitions" – on the contrary, names most certainly do have definitions under the various Codes. The definition and thus meaning of the name "Asparagaceae", for example, is quite clear: it is the taxon at the rank of family which includes the genus Asparagus. Since priority applies at the family level under the ICBN (now the ICN), the definition of the name Asparagaceae is a strict one. (The circumscription of the Asparagaceae is another matter, and is determined by research. But the same is true of PhyloCode clade names: research may alter the phylogenetic tree and thus the members of a clade.) Linnaean and PhyloCode approaches use different definitions of names: Linnaean approaches tie a name to a rank and a type; PhyloCode approaches tie a name to a portion of a phylogenetic tree. Each approach has advantages and disadvantages, although I don't think that you would learn that from the article.
 * As I've pointed out before, there are also serious WP:NPOV issues. No advantages of ranks are given; the lengthy quotation from Ereshefsky in the Lack of Ranks subsection is disproportionate.
 * The relevance of these points to Peter M. Brown's comments above are that I think that substantial re-writing is needed; the problems cannot easily be fixed by small incremental edits, although I will try some. Peter coxhead (talk) 02:55, 24 August 2011 (UTC)


 * Good to read the term “Linnaean”!  A different approach to the nomenclature of nomenclature.  Calling the approach “Linnaean” or “traditional” relies on the history of its use while calling it “rank-based” picks out,  as the defining characteristic, an a feature of its practice that can be precisely specified.  This is closely analogous to the differerence between Linnaean and phylogenetic nomeclature.


 * Perhaps it is imprecise to say that, from a Linnaean standpoint, names do not have definitions. There is a valid point here, though, which needs to be preserved in any revising.  Your definition of "Asparagaceae" contains the Linnaean rank “family” as a defining term.  “Everyone accepts that Linnaean ranks are subjective”—that’s a quote from Benton, surely the most prominent defender of the Linnaean approach.  What adherents of phylogenetic nomenclature are trying to do is to eliminate subjectivity.  Whether or not that’s a good idea, it is surely a coherent one.  Peter M. Brown (talk) 15:34, 24 August 2011 (UTC)


 * In rank-based nomenclature, names are indeed defined in terms of a type and a rank: Asparagaceae = the family that contains Asparagus. This is a definition – even though an incomplete one, because the ranks aren't defined.
 * "Linn(a)ean nomenclature" is a term that many people now dislike. That's because Linnaeus didn't care all that much about the ranks and didn't know the concept of type – to him, everything was simply obvious and self-evident. Types, as well as the idea that (some) ranks should have standardized endings, only date back to the Strickland Code (1842/3).
 * David Marjanović (talk) 10:51, 22 September 2012 (UTC)
 * Asparagaceae is definable in terms of rank and type. Though I don't have the source, I doubt that the family was defined that way when the name was introduced in 1789. As far as "Linnaean" goes: while it is sometimes confusing when adjectival forms of proper names are applied to views not held by the persons named, the practice is quite established. Marx famously declared, "I am not a Marxist". Peter M. Brown (talk) 15:12, 22 September 2012 (UTC)
 * This isn't about how Asparagaceae was defined in 1789, it's about how it's defined now under the ICN: the taxon of family rank that contains Asparagus. As far as "Linnaean" goes, both Kevin de Queiroz (coauthor of the PhyloCode) and Alain Dubois (defender of the ICZN, though he wants to have his own modifications introduced) have argued against it, the former emphasizing ranks, the latter types. Should I track down the citations and add them to the Linnaean nomenclature article? David Marjanović (talk) 10:51, 23 September 2012 (UTC)


 * You are presupposing that Asparagaceae (or perhaps "Asparagaceae") is defined under the ICN. The code, however, claims merely to regulate nomenclature, providing definitions only of words like "cultivar" (in Appendix VII), words that do not name taxa. Even if you can produce a reliable source for your claim that names are defined in terms of a type and a rank (a claim that is also made in the lead), you have a problem with neutrality; Monsch (2003), for example, argues that names do not have definitions. The ICN entries, on this view, only fix the reference of names, while a definition—if one were possible—would provide the meaning. The meaning/reference distinction is much discussed in recent philosophy. Peter M. Brown (talk) 16:23, 24 September 2012 (UTC)


 * The use of "Linnaean" to characterize a tradition much modified from Linnaeus is well explained at Linnaean nomenclature. Peter M. Brown (talk) 17:14, 23 September 2012 (UTC)

Defense of ranks - undue weight?
The long sections attempting to defend rank-based nomenclature are getting a little out of hand. (see this recent edit If it's not complete original synthesis to begin with, maybe a separate article is necessary. Most of the defense seems to be editorial (bringing in old philosophical comments) or by a small handful of researchers (especially reliant on Benton). Even the edit summary is puzzling: of course rank-based taxonomy is unscientific! Does anyone claim it's a science rather than an art? How exactly would one falsify the "hypothesis" that Arthropoda is a phylum? Without defining what phylum means, ranked taxonomy can never be considered anywhere close to a science.

At any rate, this article seems to be less about what Phylogenetic nomenclature is and how it's used, than how it differs from ranked taxonomy and why "ranked taxonomy is just as good, dammit!" MMartyniuk (talk) 12:45, 26 August 2011 (UTC)


 * Taking up MMartyniuk’s points in order:


 * WP:UNDUE prohibits editors from giving the views of small minorities as much attention as widely-held views. Views of significant minorities are OK.  It is doubtful whether users of rank-based nomenclature are a minority at all.  If so, they are certainly a significant one.  This policy is not relevant.


 * I don’t see my edit, which MMartyniuk cites, as an attempt to defend rank-based nomenclature, but rather as an effort to introduce some balance. That aside, his comment implies that there are other long sections with that orientation.  I do not see any.


 * The text I replaced claims that phylogenetic nomenclature accords better than rank-based nomenclature with standard practice in other sciences, citing a paper by Michel Laurin. Following     WP:RSUW, I have attributed this view to Laurin rather than leaving it as an outright statement.  In the interest of balance, I have also reported other views.


 * I doubt that a separate article is necessary, but this matter bears discussion. See my last bullet point, below.


 * Kuhn's insights are "old philosophical comments"? In science, much of what was published 50 years ago is perhaps old and outdated, but not in philosophy.  Much contemporary philosophy of science takes Kuhn's views very seriously.


 * MMartyniuk and David Marjanović, above, have provided no good reason for being so scornful of Benton. David Marjanović provides a quote, out of context, supported by a dead link.  As nearly all of Benton's papers can be retrieved from here, perhaps the link can be updated.


 * No one is claiming that it is a scientific hypothesis that Arthropoda is a phylum. Equally, someone's claim that barnacles are arthropods cannot be dimissed as unscientific just because she has a rank-based concept of Arthropoda.


 * I concur that the article is mostly concerned, not with phylogenetic nomenclature itself, but rather with the ways in which it differs from rank-based nomenclature. As rank-based nomenclature held sway from Linnaeus to Hennig, though, it is difficult to view phylogenetic nomenclature except against this background.  I do not think that either editors or readers can reasonably be expected to do so.

Peter M. Brown (talk) 21:13, 26 August 2011 (UTC)


 * The article should primarily (1) explain what phylogenetic nomenclature is (2) explain its relationship to 'traditional' Linnaean nomenclature (3) set out, in as neutral way as possible, the debates between proponents of the two systems. As far as I can see, although the article is still in my view rather poor, Peter M. Brown is improving it. Peter coxhead (talk) 08:04, 27 August 2011 (UTC)

Edit 22 September 2012
Several citations have yet to be added! Please give me (or others) time to track them down before you cover everything in "citation needed" tags!

I have tried to portray phylogenetic nomenclature on its own, instead of in contrast with rank-based nomenclature. I think this immediately cuts down on the controversy.

I have:
 * Removed references to "phylogenetic taxonomy" (that hardly even means anything) and to cladistics. This article is about nomenclature, not about phylogenetics. Cladistics and phylogenetic nomenclature are orthogonal to each other; you can do one without the other, and many people have.
 * Explained the definitions by type and rank that taxon names have in rank-based nomenclature. It's curious that that was missing. Was that my fault? :-]
 * Reimported (and expanded) the explanation of the different types of phylogenetic definitions from the Cladistics article, where they do not belong. The headline they had there was "Phylogenetic definitions of a clade", which is wrong: names are defined – clades exist outside our skulls!
 * Removed most of the philosophy section. Much of it, including the de Queiroz (1994) reference, didn't even recognize the definitions in terms of a type and a rank used by rank-based nomenclature; the Benton (2000) reference is simply outdated, referring to a very early draft of the PhyloCode in particular; the Kuhn quote was completely and utterly beside all points that I can see; and so on.
 * Replaced the chemistry example from Laurin (2008) with the chronostratigraphy example from the same paper. The latter can be quoted and is self-explaining; the former could only be quoted in a potentially inflammatory way (Laurin gets quite sarcastic there).
 * Modified the history section according to Kevin de Queiroz's remarks above. That the PhyloCode will be implemented is now just about certain; there is even hope that this will happen soon (within the next few months), but I cannot cite the source for that.
 * Molecular phylogenetics is not a related article. It's... just... not.
 * Fixed the format of the references (...or so I hope!) by not using the templates.

There probably should be a controversy section. Most of the controversy, though, is a mess of controversy about phylogenetic nomenclature in general, the PhyloCode in particular, the usefulness or lack thereof of ranks (mandatory or not), and sometimes other issues like the convention of only naming clades.

David Marjanović (talk) 20:31, 22 September 2012 (UTC)


 * The Benton paper is a rich and multifaceted argument. Some of it, indeed, relates to abandoned features of the Phylocode. Much of it does not.


 * What on earth is wrong with citation templates? Keeping citation format consistent is easier when they are used. When I need to add a new reference, will I now have to study your versions in detail, trying to keep the spacing, order, punctuation consistent with what is there? Will Rjwilmsi be able to use AWB to fix things when I mess up? Peter M. Brown (talk) 23:01, 22 September 2012 (UTC)


 * All of the Benton (2000, 2007) papers has been refuted in the literature that's already cited. Feel free to argue about it here, though I won't have much time next week.
 * I have no idea about the templates, actually. In 2008, I tried to cite chapters in edited books, there was no way to do that with a template, and someone said in the source text of the article that nobody should use the templates and they're "bad code". So I tried doing without them. – Frankly, a consistent citation format looks nice, I definitely try to be consistent with what's already there when I cite something, but it's just cosmetics. The important part is that the information is there.
 * BTW, the template doesn't tell you how to deal with multiple authors, and it doesn't tell you whether to abbreviate journal names.
 * David Marjanović (talk) 10:46, 23 September 2012 (UTC)


 * Benton's 2000 paper is involved in current controversy. Recent papers on different sides cite it approvingly &#8203;&#8203;&#8203; and disapprovingly &#8203;&#8203;. Declaring it "refuted" violates neutrality. On the other hand, you are so totally unsympathetic that you cannot reasonably be expected to present Benton's view in a balanced way. (Please don't take that as a criticism.) If no one else comes forward before I get around to it, I shall have to do the job myself.


 * , though poorly named, is available for citing papers in edited books. For multiple authors, I generally use   and    (,    etc.). Or you can use   with multiple entries:  .  Another option is.


 * You are to be commended for undertaking the revision of these articles. A number of editors have griped about their poor organization but you are the first actually to do the work of improving things.


 * Peter M. Brown (talk) 16:40, 23 September 2012 (UTC)

@David Marjanović: it's probably impossible to write this without sounding patronizing, but you need to re-read WP:NPOV. The purpose of the article is to explain, as neutrally as possible, what the topic is about and its relationship with related topics. As Peter has noted above, opposing views cannot be dismissed as "refuted". It's not Wikipedia's job to support or refute such views, merely to report accurately what reliable sources say. Peter coxhead (talk) 09:34, 25 September 2012 (UTC)
 * I've made a few small edits where the wording seems to me not to be neutral. More difficult to deal with the the imbalance in the discussion of ranks. The article must, of course, explain fully why PhyloCode adherents believe that the need for ranks in the traditional approach is a major weakness, but it must not do so in an unbalanced way; there needs to be roughly equal space devoted to explaining why adherents of the traditional system continue to use and defend ranks. Peter coxhead (talk) 09:50, 25 September 2012 (UTC)

Proposal
The claim in Phylogenetic nomenclature that the definition of a Linnaean taxon consists of a type and a rank is incorrect and unsourced and belittles the considerable effort that biologists invest in introducing taxa. It must therefore be replaced. I would prefer to restrict the word "definition" to formulations providing necessary and sufficient conditions for the application of the target term but I think that it would be more constructive here to use a wider meaning, one including the "Systematic Paleontology" sections of articles introducing taxa in the Linnaean tradition, specifically including, where present, the diagnosis. This can be confusing since, as David Marjanović has noted, the diagnosis may not accurately characterize the type. The claim that a taxon is defined by a type and a diagnosis can be sourced, however, and Marjanović has not provided a source for his claim that the definition consists of a type and a rank despite a [citation needed] tag.

Will the following do as a replacement for the initial sentences of the section? I am well aware that many Wikipedians know far more about these matters than I, but I do think it important that the sentences be replaced.


 * In the traditional system (often called Linnaean nomenclature or rank-based nomenclature), species names are defined by specifying a type, generally a specimen that is included in the species, and a diagnosis, a statement intended to to supply characters that differentiate the species from others with which it is likely to be confused.(ref)(ref) In articles dated 1905(ref) and 1906,(ref) for example, H. F. Osborn defined Tyrannosaurus rex based on two specimens that he had studied; one was chosen as the type and a diagnosis was constructed from characters like hollow limb bones that the two had in common. Since no names had been previously established for the genus and family containing the type specimen, they automatically acquired the names "Tyrannosaurus" and "Tyrannosauridae". Osborn provided no diagnoses for these higher level taxa, as he did not recognize any dinosaurs other than the T. rex fossils as members of the genus or family. In other cases, diagnoses are written at the time a taxon above the level of species is named; when the genus Paracharnia was formally split off from Charnia, for example, new diagnoses were provided for both genera.(ref)


 * The ranks are not defined. . . . [Continuing with existing text]

References











Peter M. Brown (talk) 14:20, 28 September 2012 (UTC)


 * Peter, I'm not sure that this totally captures the difference between defining a taxon and defining its name. You also only write about "species names" and don't make clear how rank comes into it. But I'm not quite sure how to get it right, so maybe you need to add something and then we can think about it. Peter coxhead (talk) 20:45, 28 September 2012 (UTC)


 * I really need help here, and I'm sure that other Wikipedia readers do as well. What important difference is there between defining a nonlinguistic entity and defining the term that refers to it? The rules of soccer define what it is for a ball to be in play; couldn't one equivalently say that the rules define what "in play" means in a soccer context? The order Primates is so defined as to include lemurs but not colugos; the term "Primates" is so defined as to be applicable to lemurs but not to colugos. Where is the difference? Peter M. Brown (talk) 15:56, 29 September 2012 (UTC)


 * I apologise for being cautious here, but I've had my fingers burnt elsewhere trying to write about biological nomenclature (see e.g. Talk:Synonym_(taxonomy), but this isn't the only place). The name "Primates" (like all early names, and especially Linnaeus's) is tricky to discuss because the conventions and rules of zoological nomenclature have changed since the name was created. So if you don't mind let me take a slightly different example.


 * Suppose you had written "the family 'Cercopithecidae' is so defined as to include Old World monkeys but not New World monkeys; the term 'Cercopithecidae' is so defined as to be applicable to New World monkeys but not to Old World monkeys. Where is the difference?". Well, the term "Cercopithecidae" is so defined as to be applicable to any family which contains the genus Cercopithecus. Subject to priority, the term "Cercopithecidae" could be applied to New World monkeys if it was decided that they belonged in the same family as Old World monkeys, perhaps because the understanding of phylogeny changed or because it was decided to reduce the Infraorder Simiiformes to the rank of a family. All the term "Cercopithecidae" is defined to mean is "whatever set of genera are included in the same family as the genus Cercopithecus". One particular circumscription of the family Cercopithecidae, the one now found in all standard works, equates to the Old World monkeys. But the term and the precise circumscription are independent, so that unless qualified, either explicitly or by context, the term "the family Cercopithecidae" doesn't have a precise meaning if "meaning" is taken to include the composition of the family. This doesn't, of course, mean that any old set of genera could legitimately be lumped together and called the family Cercopithecidae so long as Cercopithecus was included, because of the requirement for a taxon to make biological sense in terms of characters and/or phylogeny so that the members have to have some similarity to Cercopithecus.


 * This is a very odd way of assigning names; it's not the way we normally do it. It's something like defining the USA as whichever country includes the city of New York (whereas what we actually do is define the USA geographically and then ask what cities it includes). The biological nomenclature analogy is that if the USA–Canada border ever got redrawn to put New York in the former Canada, then (assuming priority) that country would automatically and by definition become "the USA" and the rest of the former USA would automatically and by definition have another name. All this is very difficult to explain well, and this article certainly isn't the place! Peter coxhead (talk) 17:12, 29 September 2012 (UTC)


 * I understand very little of the discussion in Talk:Synonym (taxonomy) and am not sure I want to. One statement that I do understand and agree with is yours: "Personally I think it's better to avoid using the verb 'define' when applied to a taxon", though not because "it can be ambiguous between naming and circumscribing" but because a definition is generally understood to capture a meaning and neither naming or circumscribing, in my view, does this. I am a lapsed philosopher (I do have a Ph.D.) and read a lot about meaning prior to abandoning the field in 1988. Somehow, though, I do want to replace David Marjanović's claim that the definition of a Linnaean taxon consists of a type and a rank. Perhaps I do not need to use the word "definition" in the replacement. Peter M. Brown (talk) 20:52, 29 September 2012 (UTC)

I'm sure you're right about "define" (I also think that my wording "it can be ambiguous between naming and circumscribing" is poor). Let's get back to the article. Why is the first paragraph of Phylogenetic_nomenclature there at all? Why start a section which should be about phylogenetic nomenclature (PN) by discussing a different kind of nomenclature? I now think I would remove this material from there altogether; this section should clarify what PN is and then later sections discuss the differences between it and standard biological nomenclature. The main differences can, I think, be discussed without getting too bogged down in precisely explaining the latter. [Off-topic: as to the meaning of "meaning", I hold that Wittgenstein gave the right answers in the Investigations...] Peter coxhead (talk) 21:13, 29 September 2012 (UTC)


 * Eliminating the paragraph is fine, but we do have to draft those later sections. Right now, Phylogenetic_nomenclature, the following section, still contains the objectionable claim that taxon names are defined by a type and a rank.


 * Your discussion at 17:12 UTC is very helpful and clears up some long-standing confusions for me. I could wish that it were on Wikipedia where I might have stumbled across it.


 * I have to doubt that we have "the right answers" about meaning. Wittgenstein's work was very important, but I think that later writers like Quine, Kripke, and Putnam (and probably some post-1988 folks I don't know about) have made important subsequent contributions. Peter M. Brown (talk) 22:17, 29 September 2012 (UTC)

Ranks
I've now attempted some re-writing of Phylogenetic_nomenclature; see what you (and others) think. I have a problem with the long Ereshefsky quote (apart from its length). It already has some correct qualifications/expansions in square brackets, but it would need more if it is to be correct. In the context of this article, it's particularly important to use terms which have technical meanings within the various codes very carefully. But Ereshefsky uses "taxon" wrongly. He writes, for example, "... for Wiley and Simpson, the name 'Hominidae' refers to two different taxa". But this is just wrong if "taxon" has the meaning in the ICZN. Principle 5 of the ICZN is clear: "To avoid ambiguity, the use of the same name for different taxa must not occur and is prohibited." Within the code, "Hominidae" always applies to the same taxon (namely the family containing Homo), but Wiley and Simpson give it a different circumscription, which is fine by the code because it explicitly does not attempt to regulate the content of taxa, only their names. What Ereshefsky's last sentence should say is "In brief, the Linnaean system causes Wiley and Simpson to assign different names to groups they agree have the same circumscription, and it causes them to give the same name to what they agree are groups with different circumscriptions." (Incidentally, the PhyloCode doesn't altogether avoid similar problems. So long as the exact phylogenetic tree is unknown, two authors could perfectly well give the same PhyloCode name to groups they consider to have different circumscriptions because they believe different trees are the correct one and hence that the descendants of the ancestor used to name the clade are different. Neither the rank-based codes nor the PhyloCode assign names based on precise known circumscriptions.) Now I can remove the Ereshefsky quote and substitute a correctly worded explanation of the perfectly valid point that Ereshefsky is making, but I'm not sure that this is appropriate; wouldn't it be OR to correct him in this way? Any ideas? Peter coxhead (talk) 09:55, 30 September 2012 (UTC)


 * By all means, remove the Ereshefsky quote and put his valid points in your own words; you write much more clearly than he does. Your problem is not OR but SYNTH: when two writers disagree, can you say so unless a reliable source has noted the fact? I think that it is impossible to write an extended Wikipedia article without some degree of SYNTH. How much is too much? Rather more, I think, than you're proposing here. Wiley and Simpson are in conflict (Ereshefsky says so); in particular, Wiley circumscibes Hominidae this way while Simpson circumscribes Hominidae that way. Few Wikipedians are finicky enough to object. Peter M. Brown (talk) 18:18, 30 September 2012 (UTC)


 * I've now done this. I've also removed the redundant discussion of ranks at the start of the Definitions section. I kept and re-wrote what I thought was relevant existing material but it wasn't sourced, so some more inline citations are needed. Peter coxhead (talk) 09:55, 1 October 2012 (UTC)


 * With reference to the Wiley/Simpson example, you write, "Thus under the zoological code. . .two families with the same name had different circumscriptions." This isn't clear, though, from your presentation of the example. Did Wiley include anything in his Hominidae that wasn't in Hominini? If not, the circumscriptions would be the same. Peter M. Brown (talk) 17:52, 1 October 2012 (UTC)


 * I assume so, but (sinfully) I didn't check on the original sources, just took it that what Ereshefsky wrote was correct in this respect. Given the date of Wiley, it's likely that he meant Hominidae as per Hominidae, but it needs to be checked and the text slightly tweaked. Peter coxhead (talk) 09:49, 2 October 2012 (UTC)
 * I have now managed to access the original Ereshefsky paper. Whoever originally set up the paragraph I removed as a blockquote was seriously at fault: it wasn't a quote from Ereshefsky but a partial paraphrase with added errors (e.g. Ereshefsky got the root "Homin" right). Ereshefsky does say that for Wiley Hominini was "part" of Hominidae, so I'll tweak the text to say this. The original Wiley source is a book which I don't have access to at present: Wiley, E. (1981), Phylogenetics: The Theory and Practice of Phylogenetic Systematics. New York: Wiley & Sons. Peter coxhead (talk) 11:09, 2 October 2012 (UTC)

Can't taxa on different phylogenetic nesting levels have the same rank?
Phylogenetic nomenclature claims that "Each level of nesting must have a different rank in the rank-based codes". Do the codes really say that? In the Taxonomy subsection of the Crocodilia article, the taxa Gavialoidea, Alligatoroidea, and Crocodyloidea are all listed as superfamilies within the order Crocodilia, but the Phylogeny subsection of the same article shows Gavialoidea on a different level, basal to the clade Brevirostres consisting of Alligatoroidea and Crocodyloidea. The source for the Taxonomy subsection is not provided; can we conclude that this source, whatever it is, conflicts with the source for the Phylogeny cladogram? Peter M. Brown (talk) 13:37, 13 October 2012 (UTC)


 * I think this is another example of the great difficulty in writing very precisely about this subject. The codes only prescribe that all ranks have a fixed order. If a rank-based classification is used to represent nested clades, each named level within a branch must have a different rank which decreases as the nesting moves upwards in the tree (and ranks must not be skipped). However, there's no requirement for the levels to match up across different branches. If you consider this cladogram:


 * there's no problem with the Superfamily Gavialoidea being at a different level from the other two Superfamilies so long as Brevirostres is not formally named; any number of unnamed clades can be present between ranks at the same level.
 * The problem comes when you try to produce a rank-based classification for the cladogram which includes the two extinct genera, since they must have the missing ranks supplied:


 * However, if these ranks are supplied, then the requirements of the rank-based codes are met, regardless of the fact that the Superfamilies lie at different depths from the root. Along any one branch the ranking is preserved. Peter coxhead (talk) 17:10, 15 October 2012 (UTC)


 * Peter M. Brown: your rewording of this is a real improvement! Peter coxhead (talk) 00:18, 16 October 2012 (UTC)


 * I hope that it is obvious how much I depended on your explanation. Incidently, "Peter M. Brown" is awkwardly formal; make it "Peter Brown", "Peter B.", or just "Peter". I specified the "M." when creating the account as a newbie, being used to the ambiguity associated with having a common name, but I now realize that there is no User:Peter Brown, so this was unnecessary. As of today, I have specified "Peter Brown" as my nickname. Peter Brown (talk) 15:12, 16 October 2012 (UTC)

Philosophy
The Philosophy section needs work. There are important philosophical differences between rank-based and phylogenetic nomenclature, but only one small aspect appears here. I also think that the long Laurin quote is unhelpful. I suggest that to the general reader, the obvious analogy to the way that boundaries are fixed in determining geological time periods is not phylogenetic nomenclature but circumscription (e.g. specifying the precise boundaries of a taxon by specifying what organisms are included in it), which neither system uses. An analogy with PN might be to define a geological period on the basis of all strata which were affected by ("descended from") a particular geological event.

This points to a feature of PN which its proponents seem to play down and is weakly represented in the article: just as we wouldn't know for certain which strata were affected by a particular event, we don't know (can never know?) the precise phylogeny of organisms, so that the believed circumscription of a PN taxon is just as subject to change as is the circumscription of an RN taxon. To the lay person and to biologists generally, the "meaning" of a named taxon is partly to do with what organisms it is believed to contain and thus which appear in lists in reliable sources. PN has the "advantage" that an omniscient being knows the precise circumscription of every name, whereas the inherent fuzziness of RN makes this impossible. But to non-omniscient beings, there doesn't seem much difference; indeed the believed circumscription in PN is perhaps more likely to change, since it depends only on phylogenetic tree evidence, whereas, at least traditionally, in RN both characters and phylogeny are taken into account. Peter coxhead (talk) 09:55, 1 October 2012 (UTC)


 * Responding to your second paragraph: good point, but it's not philosophy. I'm thinking of resurrecting Some criticisms of phylogenetic nomenclature from the old version of the Cladistics article in a new section of Phylogenetic nomenclature; your point, if it can be sourced, could perhaps go there. Nixon & Carpenter (2000), cited in that section, argue that PN is more likely to change than RN even without considering characters. (I am referring to their sand lizard discussion; while the paper is much cited, little attention seems to be paid to that argument.) Peter M. Brown (talk) 18:36, 1 October 2012 (UTC)


 * I agree that the section title is also problematic, although phrases like the "philosophy of classification" do seem to be used in the literature with a meaning something like the "underlying principles of classification". Peter coxhead (talk) 09:45, 2 October 2012 (UTC)


 * Using "philosophy" to refer to underlying principles seems unobjectionable, but I don't see your point as relevant to either. Stability of circumscription is seen by the PN crew as a virtue of their approach and can appropriately be considered in a section on philosophy. Stability of believed circumscription is something else; that the two are conflated is an objection, one that does not reflect an underlying principle but rather attacks the importance of such a principle. Peter M. Brown (talk) 15:05, 2 October 2012 (UTC)


 * The Laurin quote does need to be replaced. Specifying geological periods using type sections and specifying them by boundary stratotypes are equally objective, though the latter has proved more useful. Laurin does have a coherent POV, though his analogies are terrible (Mendeleev did not individuate elements by the number of protons, as the proton was not discovered until ten years after his death). I think that some of de Queiroz's work may be a better source for the view that terms should have a fixed circumscription, one that will not change under theory revision. For the opposing view, that the potential for such change is useful for scientific progress, I recommend Benton (2000). Peter M. Brown (talk) 16:01, 2 October 2012 (UTC)

Deleted quotation
Among other matters, presents analogies drawn from chemistry, stratigraphy, and geopolitics among his arguments for the superiority of phylogenetic over rank-based nomenclature. The stratigraphy analogy found its way into Phylogenetic nomenclature as a long quotation which I have now deleted on the grounds that the argument is not philosophical. Though this does not preclude the quotation from being restored to the article under a different heading, I urge that this not be done.

As I have made clear above, I regard these arguments as poor. This is, of course, merely my POV and provides no reason for excluding them. More to the point, they have been ignored in subsequent work by other PN advocates, writers like de Queiroz and Cantino, who I suspect agree with my evaluation. While other contentions in this paper have received attention in the literature, the chemistry/stratigraphy/geopolitics analogies have been cited only in one paper that Laurin himself wrote,, and in one that he coauthored,. In accordance with WP:UNDUE, a brief mention of these analogies would, at most, be appropriate. Since the arguments have gone nowhere, there is no justification for a 200-word quotation.

References







Peter M. Brown (talk) 15:36, 9 October 2012 (UTC)


 * I think you've made a very good case for removing this quote. The revised section seems to me a great improvement, presenting the issues very clearly and with appropriate support. Peter coxhead (talk) 08:32, 10 October 2012 (UTC)


 * And I thank you for the reference to the ICZN definition of "taxon", which I surely would not have come across by myself. Peter M. Brown (talk) 21:38, 10 October 2012 (UTC)

New diagram in Definitions
Maybe it's just me, but File:Clade names and phylogenetic hypotheses.svg seems confusing. In the right-hand figure, a line from the bottom goes up and apparently disappears behind a dotted triangle; it does not seem obvious how the line is related to the triangle that occludes it. The caption speaks of "the hypothesis that the relationships are as in the right tree"; are people going to understand what relationships the right tree shows? Peter M. Brown (talk) 22:44, 1 October 2012 (UTC)


 * Good point; it would be better if the dotted triangle finishes lower down. I'll fix it. Peter coxhead (talk) 09:41, 2 October 2012 (UTC)

Ranks limit names
The claim that the need for ranks limits names when only clades are named (which was removed) is basically true, although not clearly explained. I've tried to add something which is clearer without going into too much detail, which I don't think is relevant here. The key point is that if you name only clades then only so many "levels" or "depths" in a phylogenetic tree can be named otherwise you run out of ranks. If you allow paraphyletic taxa then it's easier, because such a taxon can span many levels in a tree. (Compare a classification which divides "cladistic reptiles" into "traditional reptiles" and birds, needing only two ranks, with say the diagram at Reptile, which has names at seven levels without even reaching birds.) Peter coxhead (talk) 10:46, 2 October 2012 (UTC)

Darwin and Hogg
The second sentence in the History section implies that  John Hogg's 1860 introduction of Protoctista as a kingdom somehow reflects Darwin's influence. Can any support for this be provided from reliable sources? Indeed, is there any reason to suppose that Hogg subscribed to anyone's evolutionary theory in 1860? He did hold that Protoctista is older than Animalia or Plantae, but the fact that some taxa are older than others had already been well established by fossil studies; Hogg's extension of this notion to the kingdom level does not necessarily presuppose evolution. Peter Brown (talk) 17:36, 19 January 2013 (UTC)