Talk:Cladistics
From Academic Kids
| Missing image Cscr-featured.png Featured article star | Cladistics is a featured article, which means it has been identified as one of the best articles produced by the Wikipedia community. If you see a way this page can be updated or improved without compromising previous work, feel free to contribute. |
Unsorted Discussion
I am amazed at the rate of progress of this article. The Wikipedia model is truly the most exciting, interesting and fulfilling model for an encyclopedia that I have ever come across! Congratulations! user:exigentsky
I don't think polyphyly, monophyly, and the like are really cladistics terms, are they? After all they refer first and foremost to the evolution of the group in question; for instance the claim that the arthropods are polyphyletic is a hypothesis about their origins, which certainly cladistics have been used to help evaluate, but also more traditional lines of reasoning. From what I have seen, to, the trees constructed usually come out reasonably close to one another.
Go ahead and fix it to your taste. I don't have anything against the concept of cladistics, but the current practice violates a principle that it took a lot of painful experience to absorb. Namely, if the practitioners of an art can't/won't speak in clear, simple language, that's usually a symptom of massive confusion on their part. I do think the subject deserves an entry. DJK
I've seen some cladistics and as far as I can tell it uses no more jargon then any other field. A lot of the terms, as stated, are used by other methodologists as well. But more to the point, except for perhaps clade each of the words expresses a concept which doesn't really have a compact synonym in normal English. So they're somewhat unavoidable, just like jargon like endoplasmic reticulum is unavoidable in protistology.
Ah, one more thing I noticed. Cladistics is not a classification system, as stated on talk:Linnaean Taxonomy. It's a methodology for elucidating evolutionary relationships. Taxonomy ties in because cladists intend it to reflect the phylogeny of organisms. However this intention is shared by pretty much the majority of biologists. If you look at some of the more recent Linnaean schemes for, say, protists and flowers, you'd find everything up in the air for precisely these reasons. So objections to the practice on these grounds have kind of been superceded.
On these grounds I think I'm going to take a stab at rewriting the article. Please don't be insulted by this! I'll try and leave all your points except the above, but move them to a criticisms paragraph down at the bottom, since that seems to be the standard way of presenting a neutral point of view. Personally I think cladistics is neat, at least when it works, but takes forever to get used to. :)
Hmmm. I think we can all agree that the ASCII tree is rather attractive.
I strongly beg to differ with Josh Grosse. Turtles are no more closely related to snakes than are mammals. And if it's given that birds are dinosaurs but removed from them in taxonomy, then that's paraphyletic. jaknouse 02:26 1 Jun 2003 (UTC)
The reptiles are very definitely paraphyletic, for exactly the same reason as the dinosaurs: they include an ancestral form, together with some but not all of its descendants. Remember that the ancestral amniotes are traditionally considered reptiles, so it doesn't matter what the relative positions of mammals, turtles, and snakes are (just for the record, though, turtles are now considered somewhat closer to snakes and birds than to mammals). The problem with giving dinosaurs as an example of a paraphyletic group is that they aren't always taken to exclude the birds. Josh
- Yeah, that's right. Amniota comprises Synapsida, which gave rise to mammals, and Sauropsida, which includes anapsids (turtles) and diapsids (snakes and birds). The controversy over birds being dinosaurs has died down in recent years, but I think it is still too controversial to be a good example of a paraphyletic group. A better example would be class Osteichthyes (bony fishes) being paraphyletic, since one group of fish gave rise to Tetrapods. SCCarlson
The fishes are a good example of a paraphyletic group, but the Osteichthyes are sometimes taken to include the tetrapods. The same sorts of extensions are done with the Sacropterygii, Reptilia, and Dinosauria, while the Amphibia may be restricted to a holophyletic group. Among the major vertebrate groups, the only formal taxon I can think of which is unambiguously paraphyletic is the Agnatha, which relatively few systems still use.
- I must beg to differ with both Jaknouse and Josh. Both are partially correct and both are partially mistaken. It is true that _if_ birds are descended from dinosaurs, AND _if_ Aves are excluded from Dinosauria (AND hence excluded also from Reptillia), THEN both Reptillia and Dinosauria must (by definition) be paraphyletic. I happen to believe this to be the case, but I am no longer really in the dinosaur business. I believe that the cladistic purists now use the group "sauropsida" to indicate the clade that includes turtles, lizards, snakes, crocodillians, dinosaurs, birds and any other descendants of their common ancestor. HOWEVER, mammalia is definitely the living sister group to the sauropsida. The position of turtles is poorly understood, but it is generally believed to be either basal to the squamata (snakes and lizards) or somewhere around the archosauria (with a converent anapsid condition of the skull). In either case, they are probably not closer to the mammalia than to other sauropsids. The Mammalia and the Sauropsida seem to be the only living Amniotes if the latter are defined as a node-group. I agree whole-heartedly with Stephen about using the fish as an example of a paraphyletic group. Another choice to consider would be the invertebrata. --Cladist Jan 12 2005
I'm planning to add a brief section in the textual criticism article about the use of cladistic analysis techniques borrowed from biology. I just thought I'd run it past people here first for coherence:
"Cladistics is a technique borrowed from biology, where it used to determine the evolutionary relationships between different species. The text of a number of different manuscripts is entered into a computer, which records all the differences between them. The manuscripts are then grouped according to their shared characteristics. The difference between cladistics and more traditional forms of statistical analysis is that, rather than simply arranging the manuscripts into rough groupings according to their overall similarity, cladistics assumes that they are part of a branching family tree and uses that assumption to derive relationships between them. This makes it more like an automated approach to stemmatics.
The major theoretical problem with applying cladistics to textual criticism is that cladistics assumes that, once a branching has occured in the family tree, the two branches cannot rejoin; so all similarities can be taken as evidence of common ancestry. While this assumption is applicable to the evolution of living creatures, it is not always true of manuscript traditions, since a scribe can work from two different manuscripts at once, producing a new copy with characteristics of both.
Nonetheless, software developed for use in biology has been applied with some success to textual criticism; for example it is being used by the Canterbury Tales Project (http://www.cta.dmu.ac.uk/projects/ctp/about.html) to determine the relationship between the 84 surviving manuscripts and four early printed editions of the Canterbury Tales."
Obviously it needs some wikifying, but I wanted to check I hadn't horribly misrepresented what cladistics is before posting it to the textual criticism article. Any thoughts? Harry R 12:20, 6 Jul 2004 (UTC)
- OK, I've taken your silence as approval (or indifference?) and posted it to textual criticism. If anyone sees a problem, could you take it up there please? thanks. Harry R 09:35, 8 Jul 2004 (UTC)
Harry: Interesting idea. Only criticism: "stemmatics" is spelled "systematics." --Cladist
All: I'm amazed at the depth of this article on a relatively obscure field. The authors are to be commended. However, there is some confusion in the Cladistic Methods section. It is certainly true that it has "taken some time for cladistics to settle in." There is much debate about the use of Cladistitcs. However, the sentence continues, "...and there is some questioning over in just what sort of circumstances cladistics is applicable." While there is debate about the use of numerical methoids like Parsimony and Maximum Likeelihood, I don't think there is much formal debate about the use of phylogenetic systematics (that is, the adherence to phylogenetically relevant, clade-based taxonomies). Phenetics does not have many supporters in the systematics world. In my experience, even those systematists who do not care for numerical methods would never purposefully erect a new paraphyletic taxon. Thus, it is not _cladistics_ that may be regarded as inappropriate under certain circumstances, but the methods of _numerical_taxonomy_, including parsimony. In fact, Henig never mentioned parsimony. This is a fine distinction that is often glossed over in the literature, but I would love to see it clarified in this article. Perhaps we could change the above-quoted sentence to read: "Cladistics has taken some time to settle in, and there is still wide debate over how to apply Henig's ideas in the real world." --Cladist July 20 2004 (incidentally, I chose my handle before I saw this article. I didn't intend to sound so arrogant)
- Hi, thanks for the comments. Feel free to Be bold and improve the article in ways that you see fit. For large textual/structural changes (particularly for deletions), it's always good to discuss them here in Talk, but go for your life! We really need somebody to help work on the various articles on phylogenetics, systematics and the like. My knowledge is more in the population genetics area and my grasp of systematics is much less firm. There is much duplication and some of the various articles could be merged, or at least structured and interlinked in better ways. --Lexor|Talk 07:58, 20 Jul 2004 (UTC)
Thank you. I have begun working on it (check out the first section). Next I want to talk a bit more about synapomorphies in the Intro section. Eventually, I'd like to start tying all the systemtics articles together, but I don't think my advisor would count that toward my dissertation requirements... --Cladist July 21 2004
Cladistics can refer to numerical taxonomy or to clades-only approaches to classification. Note the latter is not the same thing as phylogenetic classification, which may allow paraphyletic groups, and is rejected by at least some notable biologists. In particular I think it is unpopular among those specializing in basal groups, like protists or extinct vertebrates, where it has led to numerous difficulties. I've notably changed the article to reflect this. Josh 10:03, 10 Oct 2004 (UTC)
Josh: I beg to differ. In common parlance (as if common people speak of cladistics vary often) you are correct, however, I believe you are missing the history of the terms. Numerical Taxonomy was a rising field before Cladistics became popular. Numerical Taxonomy referred to a suite of techniques in which mathematical "rigour" was applied to the science of taxonomy. Cladistics was a separate development in which the use of synapomorphies was held to be philosophically superior to the use of symplesiomorphies (not to be confused with symplesiosaurs, which were the first aquatic organisms believed to perform musical concerts). However, Henig did not recommend any particular numerical approach, and did not foresee the mathematical problems that are caused by excess homoplasy in the dataset. Parsimony Analysis, the first blend of numerical Taxonomy with Cladistics, was the solution proposed. Thus, we can use Cladistics in the original sense of "clades-only approaches to classification," or in the modern sense of parsimony-based cladistic numerical taxonomy, but we would be wrong to use 'Cladistics' as synonymous with 'Numerical Taxonomy'. Parsimony-based Cladistic methods are at best a sub-set of Numerical Taxonomy. Again, in common parlance students of "phylogenetic classification" might decide to use paraphyletic groups, but strictly speaking the two are mutually contradictory. "Phylogenetic Classification" is the use of phylogenies to produce classifications, and hence does not brook paraphyly. If it did allow paraphyly, there would be no limit to the arbitrary taxa that taxonomists could erect. Do you perhaps have "Phylogenetic Cassification" confused with "Evolutionary Classification?" In the latter, many evolutionary attributes (including similarity of adaptations) were collected to construct groups, and hence paraphyly was occasionally allowed. However, these are contentious issues. I would like feedback before changing anything in the article. --Cladist November 9, 2004
The classifications I'm referring to don't allow groups based on convergence, which would be polyphyletic, but do allow classifications based partly on plesiomorphies, which are paraphyletic. I don't think I have the terms confused, or if I do, so do enough biologists that I'd say their meanings aren't really set in stone. See the following note.
Part of Henig's genius was in recognizing the distinction between paraphyly and polyphyly (which had hitherto been treated as the same phenomenon), and showing that the exclusion of descendants from a group was as problematic as excluding the common ancestor.
I've removed this, because it seems wrong on several counts. First, it doesn't make any sense for him to have discovered that paraphyly is as bad as polyphyly if nobody noticed the difference before. Second, I'm pretty sure paraphyly was originally identified with monophyly, and a few biologists still use the term that way.
Third, and most importantly, not everyone agrees paraphyletic groups are bad. In fact, they're necessary to provide a comprehensive classification of ancestral organisms, as argued by people like Cavalier-Smith and more or less admitted in the PhyloCode guidelines. To put it simply, what family did the Hominidae evolve from? Either there is no answer, or the answer is a paraphyletic family. It's hard to say the former is a definite improvement, and Hennig certaintly didn't show it was. Josh
Josh: I have restored the quote that you excised, with the correction that it was monophyly, and not polyphly, that was confused with paraphyly. I think you will find that this makes more sense. As for your objection that some biologists accept paraphyletic groups, this is true; however, they are not cladists. Many biologists reject cladistics outright because it is inconvenient.
Also, I have a problem with your comment about Hominidae and subjective classifications. Your arguments do apply to species, but not to other ranks of taxonomy. In cladistics, all clades are expected to have descendants, and these are automatically included in the clade. Thus, the hominidae will always, automatically include all of the descendants of humans, of chimps and of gorrillas (etc), even if they should become something very different. Linnaean taxa do not have this property, and this lack makes them problematic. Thus Linnaean ranks are artificial (because human priority determines which species will be included or excluded), and I don't think anybody who undestands the issue would seriously dispute this notion. SPECIES on the other hand, are a different matter. Species are notoriously poorly defined. Check the literature for the phrase "species problem." There are as many definitions of the word species as there are hairs on your head, and not all of them assume that a species is a real, natural unit. One real problem with some species definitions is that species are paraphyletic: that is, the descendant of a species is automatically excluded from the species. There have been species definitions that seek to avoid this issue, however.
In any event, the Homo sapiens problem you point out is a problem of species definition, not of cladistics. The best that the authors you allude to can say is that we should not require all taxa to be monophyletic, because we cannot require species to be monophyletic. This is a straw-man argument because cladistics is about higher taxa, not species.
Personally, I see it this way: Higher taxa/clades are sets whereas the species are elements of the sets. We expect a set to include all subsets, but we need not expect an element to include anything. The set of all primes includes all the primes we can count, plus the ones we haven't counted yet, but each prime is only a number, and does not include anything. If you object that species include individual organisms, I can respond that a prime number is equal to the sum of some other set of numbers - but this is a different kind of set. If you object that a set of primes is not analagous to higher taxa in that it includes only elements, but no sets, I can respond that the argument works just as well (or better!) with the set of all pairs of prime numbers, or the set of all pairs-of-pairs of prime numbers, ad infinitum. ----Cladist 05:49, Feb 14, 2005 (UTC)
I'm sorry, Cladist. You missed my point regarding the article. Hennig did not prove paraphyly is as bad as polyphyly. Many biologists don't think it is. Some of them accept cladistic methodology in full for uncovering and describing evolutionary trees, but think clades and taxa should be kept as distinct notions. I don't know if you count them as true cladists, but their opinions means that the the evil of paraphyly can't be described as a fact.
Also, I'm not convinced nobody noticed the difference between monophyly and paraphyly. It's true the latter have been called monophyletic, but in that case there is still a separate word for the former, holophyletic. When were they introduced, and was it a new distinction, or did it reflect something people already discussed in other terms? There is so much material that assumes everything besides clades is garbage and bends the truth, I would like to see a reference.
I'm entirely aware how cladistics works, and this includes both its strengths and weaknesses. First, one should realize that neither monophyletic nor paraphyletic groups exist in nature, since the evolutionary tree is essentially continuous. They're labels you apply to sections of it. Every cut you make gives a monophyletic section above and a paraphyletic section below. It's not inherently more valid to name either one.
The argument I gave does apply to taxa at all levels. The common ancestor of the Hominidae and Cyprinidae can be placed in a phylum, but it is impossible to give it a monophyletic family. Some take one way out of this, abandoning ranks. Some take another, abandoning the requirement of monophyly, although they may still use clades to describe relationships. Neither option is inherent in nature, and I think we need to keep this in mind.
As such, I think we shouldn't have the passage, or anything else that asserts that clade-based classification is inherently better. It isn't, just appropriate for different things. Josh
Josh:
You make some good points and some bad ones. I can see you have put a lot of thought into this. You are correct that I mis-read your argument re: Hominidae and Cyprinidae. It had nothing to do with the species problem. I especially appreciate your desire to see references. I recommend (a) Phylogenetic Systematics, by Willi Henig and (b) Inferring Phylogenies ch. 10 ("A Digression on History and Philosophy"), by Joseph Felsenstein. Would you be so kind as to tell me which of Cavalier-Smith's articles provided you with so much information?
I am going to ask you to agree to restore my paragraph, but precede it with a clause such as, "Proponents of Cladistics believe that..." or "For cladists,..."
The essential reason is this: If we qualify the paragraph as I requested, it is certainly true, whereas if we remove the paragraph, we remove essential information about cladistics and the history of Systematics.
Now I have the following comments to make on your paragraph:
1. "Evolutionary taxonmist" is a phrase that is unfortunately similar to "Evolutionary Systematics." The latter was an early school of Taxonomy / Systematics proposed by Ernst Mayr and George Gaylord Simpson. You might in fact have been reading their works. In that case, you want to say "The School of Evolutionary Systematics, propounded by... etc." and state who said these things (Instead of "Other evolutionary...". Mayr and Simpson are not nobodies. Their work on Systematics and Population Ecology/Genetics are foundational to the rest of modern biology. However, their works on Evolutuionary Systematics are very much out of date. Active systematists mostly disregard the ideas you have cited, because it no longer matters what you call an organism, so long as you know its relationships to other organisms. Your sentence makes it sound as if there are many of these people, rather than a few hangers-on.
2. I looked up one of Cavalier-Smith's articles. He does use the term holophyeetic to mean monophyletic, and probably does this to distinguish monophyly from paraphyly. However, he also seems to be expressly concerned with whether or not a group is paraphyletic. Your paragraph would have it seem that modern writers do not care about this distinction. Cavalier-Smith may wish to keep the taxonomy as-is, but difference between paraphyly and monophyly is meaningful for him.
3. Your discussion of Hominidae is very unclear; at least it does not mean what you later described in this discussion page. I think this is where my confusion came from.
I suggest this alternative:
Other taxonomists argue that paraphyletic taxa provide information about significant changes in organisms' morphology, ecology, or life history; in short, that taxa and clades are both useful but separate notions. A few use the term monophyly in its older sense, where it includes paraphyly, and many use the alternate term holophyletic (which means monophyletic, sensu Henig) to emphasize the distinction between monophyly and paraphyly.
I have removed the Hominidae argument because it needs severe re-working, both in terms of content and clarity. But note that this is only a suggestion. For now I leave the decision to you.
--Cladist 12:44, Feb 20, 2005 (UTC)
Cavalier-Smith gives a quick defense of Mayr's system in A Revised Six-kingdom System of Life. He is not trying to keep the taxonomy as is, and in fact many of the paraphyletic taxa are ones he introduced. He doesn't distinguish them from monophyletic taxa because they are inferior, but because the distinction is still important to anyone studying evolutionary relationships.
For classifications of Protista, Cavalier-Smith is the only one who explicitly uses a system other than Hennig's, but I have seen a number of others that use paraphyletic groups without comment. I have seen various arguments against clades-only systems, and palaeontologists in particular seem to have issues with them, as one might expect since they deal largely with ancestral groups. Treatises on vertebrates that avoid paraphyletic groups seem not to exist. As such, while they appear to be in the minority, I am not convinced evolutionary systematists are only a few hangers-on.
I will try to find copies of the references. In the mean time, rather than restoring your quote, I've tried altering the entire section to make both positions less opaque. Please let me know if you have any complaints. Thanks, Josh
Josh:
Much better. I added a phrase that clarifies the reason why cladists regard non-synapomorphic characters as arbitrary. I still have problems with the next paragraph, but I can't think of a good replacement for now. Good work! -- --Cladist 18:10, Feb 21, 2005 (UTC)
Terms
Hey! Looks like we get a gold star -- nifty! Just a comment on semantics: I notced that someone used the term "developed" where "evolved" might have been more appropriate. The two are interchangeable in common parlance, but "develop" is unfortunately used in biological jargon to describe ontogeny (i.e. growth and differentiation of an organism), not evolution, and it is extremely important to distinguish the two. I corrected the example I found. Hope I don't sound like a word-nazi. --Cladist
Could someone please explain the term basal? The page was a redirect, but the term does not even occur in this article! Sebastian 07:04, 2005 Feb 7 (UTC)
References
Hi, I am working to encourage implementation of the goals of the Wikipedia:Verifiability policy. Part of that is to make sure articles cite their sources. This is particularly important for featured articles, since they are a prominent part of Wikipedia. The Fact and Reference Check Project has more information. Thank you, and please leave me a message (http://en.wikipedia.org/w/wiki.phtml?title=User_talk:Taxman&action=edit§ion=new) when you have added a few references to the article. - Taxman 18:50, Apr 21, 2005 (UTC)
