Specifically, what is the evidence for common descent?(Not quite) famously, Darwin mused about the similarities of taxonomic hierarchies in linguistics and biology and asserted that the hierarchies must ultimately point to common descent. (Chapter XIV, On the Origin of Species) That’s common descent as distinguished from microevolution.
The linguistic equivalent is the single origin of all languages (eminently unproven and deemed unprovable) as distinguished from a language family (with demonstrable relevant organic shared features).
Darwinists are welcome to present their evidence. From Rumraket, we have the observation that all organisms can reproduce, “Nesting hierarchies are evidence of common descent if you know that the entities sorted into hierarchies can reproduce themselves. And that particular fact is true of all living organisms.” Good start.
From Joe Felsenstein we have the doubt that the border between micro- and macroevolution can be determined, “OK, so for you the boundary between Macro/Micro is somewhere above the species level. How far above? Could all sparrows be the same “kind”? All birds?” Not very promising.
From Alan Fox, “Darwin predicted heritable traits. Later discoveries confirmed his prediction.” Questions: Which heritable traits specifically? Was there a principled improvement over Mendel? And how does this lend credence to common descent?
Thanks to all contributors.
The difference I had in mind is that predictions might be taken as future-oriented — what we will expect to find. Whereas abductive inferences can be past-oriented — what we would have expected to find, if the posit were true.
Moreover, the abductive inference brings out the modality of scientific reasoning: we are entitled to assert that P, where P is some unobserved or even unobservable process or entity, just in case, if P were the case, then would have expected to observe O, and we have observed O.
There are two principal ways that this inference can be challenged: by introducing further observations that can’t be accommodated by P, or by introducing alternatives to P.
The ID/creationist strategy has been to interpret the scope of P too narrowly, and then say, “but P didn’t predict O” for any O that’s unexpected. But this is just not right. Some unexpected observation can be unexpected in terms of being different from previous observations (so that straightforward induction fails) or from conflicting with an overly crude picture of what P consists of or entails.
What the creationist/ID person really needs here is to show that either there are some observations that cannot be made intelligible under any reasonable construal of evolutionary theory, or that design theory can explain observations sufficiently better than evolutionary theory does.
A scientific theory is successfully replaced when its successor can not only explain the anomalies that were problematic for the previous theory, but also explain everything that the previous theory did explain.
Design theory fails in both respects — there aren’t yet any observations that are fatal objections to evolutionary theory as correctly understood, and the features of life that evolutionary theory does explain very well — such as nested hierarchies across multiple lines of evidence — aren’t explained by design theory at all.
Given some of the statements hereon along the lines of “this is an entailment of Intelligent Design” and “well that’s an entailment of God”, I think some folks don’t really know what that word means…
Interesting how you can simultaneously not understand something and yet know it’s wrong.
Kantian Naturalist,
What do you think evolutionary theory explains?
Kantian Naturalist,
What do you think evolutionary theory explains?
Since we’ve been talking about common descent explaining (predicting, abducting, whatever) the nested hierarchy of genetic (and other) data, it would seem that you could have arrived at this from context. In fact, we’ve discussed this many times before, so it seems that memory could have helped you answer the question too.
I’ll note some additional observations that the Theory of Evolution explains:
1) the ToE explains why we see the fossil organization we see. For example, if the bible was accurate in it’s depiction of Noah’s flood, the fossil record would show a mix of all organisms all throughout the various strata, but that’s not what we find. We find very distinct organization of plant and animal groups ascending through the ages.
2) More interestingly, the ToE explains why we find no more recent groups of organisms related to extinct organism lines. In other words, we never find “jumps” between organisms with given traits that went extinct and some group of organisms appearing in the fossil record later with similar traits. Why? If there really were a designer, for example, we’d expect to see such “jumps” all over the place, since a designer would have no reason not reuse traits he or she liked. We this in human design all the time for aesthetic reasons (hey…the 50s are in! Let’s make roadsters again!), but we don’t see that. Evolution explains why.
3) The ToE explains why we never find any “Cambrian rabbits” and why we really don’t expect to. For that matter, it also explains why we do not find any “crocoducks” and why we really don’t expect to find those either.
4) The ToE explains why we see some diseases…like heart disease, cancer, and diabetes…show up repeatedly in some families.
5) The ToE explains why we are starting to see “Superbugs” that are resistant to antibiotics we’ve been using.
Just to note a few things…
The empirically observed matching branching nested hierarchical patterns found in the fossil and genetic records. The mechanisms by which species have evolve over generations to produce those empirically observed patterns. The reason we occasionally see atavisms in extant animals.
There’s more but that’s a good start. What does ID-Creationism explain?
So the former is a thing, sometimes, in some way, but the latter, even though it’s assumed to be a thing too, shall never be discussed, despite the fact that you are participating in a thread where the OP is about it. Got it.
The same had been found prior to any evolutionary “prediction”: Yup, these things look more similar to each other, let’s classify them together, and more disparate things let’s classify further apart. Darwin’s contribution was to project the assumption of common descent on the classification. The tree by itself is no proof or evidence of anything, it just makes you assume things that you then should test independently.
From behind your link again
gibbon CCTACAGCCC AGCCAAACGA CACTAA
orangutan CCTACCGCCT AGCCATTTCA CACTAA
human CCCCTTATTT TCTTGTCCGG TGACCG
chimpanzee TTCCTCATTT TCTTACTCAG TGACCG
gorilla TTCCTTATTC TTTCGCCTAG TGATTA
So, sometimes some letters line up. You do some statistics and you see it’s not chance. Therefore common descent. Got it.
Either chance similarities (which nobody ever suggested, so this option should be tested only for fun or “just to be sure”), common descent (which I see you now tacitly admit is never specifically tested!) or constitutive similarities by virtue of all the species being part of the same harmonious nature, the same overall environment where everything is made of the same building blocks in different ways (if you are happy enough with an untested assumption, why not this one?). In my view, “survival” is not for the fittest alone, but for the biosphere as a whole.
Basically, you are saying that given your assumptions you expect to see things as they are. Guess what, so do I. Yet we have different assumptions. For example, given “survival of the fittest” I would indeed expect only the fittest to survive – just one species, the fittest, until their food runs out. I would not expect a bunch of weak species hang around for long just to serve as food for the fitter. Instead, if nature itself were to survive in a longer term, I would expect survival be a smaller part of a bigger picture, where objective morality plays a role.
What I’m saying is that none of your expectations, be it about evolution or anything else, makes much sense to me. Maybe mine don’t make sense to you either, so we are even.
Erik:
Holy moly.
Wrong Eric. For the umpteenth time, there is only one explanation based in observable fact for the ability to construct similqar trees from independent data. That is if they share a genealogical relationship.
No Eric, the point is that you’re being a hypocrite if you refuse to consider why-it-is that you accept a common genealogical relationship for very closely related species, but reject it for those more distantly related.
You accept the exact same methodology for closely related species which you reject if they get more distantly related. Why do you do that?
Sorry, but I don’t accept it for “closely related” (more properly, “closely similar”) species either. My criterion for the time being, because nobody has suggested anything better, is ability to interbreed. If the species cannot interbreed, then they are not related. Apes and humans are not related a la Darwin. They are just similar.
Okay, so your view is that all species on the planet were independently created. There is no such thing as speciation in your view.
Okay, that is your view. Just so you understand, evolution does not require that two species be able to interbreed to be related. So amazingly, in order to convince you that macroevolution is true, we should have to demonstrate that macroevolution is false. Macroevolution is evolution above the species level. The formation of a new species. But Erik, if a new species formed, it would only be a new species if it could not interbreed with it’s ancestor. But if it can’t interbreed with it, you would claim they’re not related.
That’s brilliant Erik. Absolutely brilliant! You’ve deliberately made it impossible to demonstrate macroevolution to your satisfaction. Because if we could, you have a priori defined them to be unrelated. In your own words, if they can’t interbreed, they’re not related.
Erik? Have fun with that!
It’s not up to me to require anything of evolution. If evolution has something positive going for itself, evidence or such, lay it out. If not, then don’t. I simply accept the situation, whichever way it is.
So, any evidence of formation of new species a la Darwin? Reminder: “We can draw a tree” and Very Long Time have already been discussed, thoroughly.
I keep bringing up (lack of) interbreeding because it suggests that there’s something in the genes that keeps species as they are, instead of enabling evolution. If evolution is possible, then genes should enable it, not prevent it. Inability to interbreed blocks an obvious mechanism of evolution. If crocodiles and ducks could interbreed, crocoducks would have a chance at natural selection and everything would be well for evolutionary variety of species.
A nonsensical statement if I ever read one. You’re taking on the part of the unconvinced, and you’re asking for evidence. Yet when it’s pointed out to you that you’re obfuscating, you flail and say nonsense.
Yes yes, we get it Erik, your ability to plug your ears and act out denial is evident to us all. Well done.
Yes and your responses were nonsensical when they came up. You’ve completely avoided addressing the point I just made, for example, that you’ve defined your way out of ever being able to concede that two species are related by descent.
Erik. Just think for a moment. The idea is that new species form so that they are no longer able to interbreed. And the question is, did the many species we observe now emerge by such a process? Did they emerge from common descent and become unable to interbreed over time? That’s what we’re saying they did.
But YOU want us to prove that they did, by showing that they are related THROUGH BREEDING. In which case you’d claim they aren’t two species, but one.
That doesn’t strike you as exhasperatingly idiotic, Erik? I guess it doesn’t. And since it doesn’t, we’re done. You’ve made yourself appear like an idiot in order to avoid conceding there is evidence for common descent. That’s good enough for me. Thank you for playing Erik.
Except that I didn’t define my way out of anything. You are free to offer your own definitions so we can compare. It would be the first time you would be defining something, by the way. You should do it more often.
This is so dumb, it is SUPERNATURALLY dumb. Literally, MIRACULOUSLY dumb. It is so dumb, I have become a believer again. I am convinced a divine omnipotent person had to intervene and suspend the natural order for so UNBELIEVABLY stupid a statement to come to exist.
Surely, Erik, you must be a God. PRAISE THE LORD WHO SHALL HENCEFORTH BE KNOWN AS ERIK!
Erik, we’re done. You may declare yourself the victor now in whatever way you feel a need to.
I know. And the evidence for this is….? Or is it indeed just say-so?
I know the claims of the theory of evolution. This thread is about the evidence for the claims. And how the claims make sense, if they do.
What does that even mean? Why would being part of the same harmonious nature create a nested hierarchy in mitochondrial DNA sequences? What same overall environment do you mean? What are the building blocks you refer to here? What different ways?
I can see why you might want to remain ignorant of genetics, because it lets you be satisfied with meaningless word salad like that, and dismiss the genetic data as “just a bunch of similarities” and the hierarchy of life as “just similar things being put together”.
By the way, you may think the OP is about universal common descent, but it never says so. The Origin isn’t about universal common descent. Darwin never presents an argument or even a claim for universal common descent. Not even his language analogy is about universal common descent of languages.
For the exact same reasons as any and all similarities do. The task for the evolutionist is to show that the tree derives from and points to something more than just similarities.
Any and all things you call similarities, some of which you call homologies.
I can see why you might want to remain stuck in unexplained concepts, occasionally adding to them to keep matters confused. What I am attempting is to get you to explain your presuppositions, but well, you win. For you “we can draw a tree” proves evolution (while evolution “predicts” and “explains” the tree – and you are not at all circular there, definitely not!). At the same time, the obvious fact that you can draw a tree on anything seems to require no attention. Nothing to do when one refuses to examine one’s own presuppositions.
Yes it is and yes he does. Let onlookers judge the full quote once again. Emphasis mine.
Erik,
Go on then, give us an example. You can draw it out and scan it and add it as a picture to a comment. Pick a few things, and “draw a tree”.
I already did. One example was cups and vases. Another was this image https://s-media-cache-ak0.pinimg.com/originals/49/d5/66/49d5669cc6ce8a2ec5a461ae32d52f21.jpg
Now, you can justifiably tell that those things don’t evolve and the justification has nothing to do with the tree. If it were only for the tree, you would have to conclude that they evolve, a cup into a vase or vice versa. Similarly, the claim that species evolve into other species must have its confirmation elsewhere than in the tree.
Organisms reproduce themselves, and they mutate when they do. Cups and vases don’t you dolt. If cups and vases reproduced themselves, and suffered mutations during reproduction, then yes, you could in fact infer common descent if there were objectively nesting patterns in their attributes.
Why is this so hard for you to fathom?
Those aren’t nesting hierarchies. It’s just a chronology of processor designs. The fact that the picture is titled with the word “evolution” and that somebody elected to draw a line and put individual processors on the branches does not in any way imply it constitutes an example of a true branching process of evolution.
You really seem to have a hard time fathoming how it is that trees are actually constructed. It’s not just that some idiot sits down, draws a line and then arbitrarity puts individual entities on the branches.
Organisms don’t mutate randomly while reproducing. They reproduce within their own species and remain essentially unchanged indefinitely, as attested in every species called “living fossils”. You dolt. We’ve been over this already. We were supposed to be done.
And of course you can tell me how it’s really done. Try explaining it and I will compare it to the book on cladistics that I’ve read.
OK, I’m calling Poe. No one can be that ignorant of basic biology.
Now you’re just blatantly denying reality.
This is where your refusal to think has taken you. Concrete real world facts are just brainlessly denied. Now, for reasons not entirely clear, Erik has decided to declare that mutations don’t take place during reproduction.
That’s just fantastic Erik. You are really good at this denial. Just say no! Bad evolution, bad!. No! No! It doesn’t happen. No mutations! No!
May I suggest you put it in all caps next time to really emphasize how thoroughly you deny empirical realities? We wouldn’t want to think someone could have doubts about your sincerity here.
As would always be the case, whether macroevolution was true or not.
If species A suddenly started reproducing with species B, they would by definition constitute the same species. For fucks sake Erik. Hahaha oh my god why do I even have to explain this to a grown man. Holy fucking shit you take the cake buddy.
Since they unassailably mutate, this simply isn’t true.
This term doesn’t mean what you think it means. None of the so-called living fossils are “essentially unchanged”.
I agree, we’re supposed to be done. But you keep digging deeper and I just can’t help myself trying to try to pull you back up.
Erik probably interpreted “Organisms don’t mutate randomly while reproducing” as crocs randomly morphing into ducks as they copulate. And that’s an inference based on all the other amusing stuff he’s said about what he believes evolution is.
LOL. Okay, let’s look at distance matrix methods using DNA as an example.
1. First, we start by constructing multiple alignments of sequences.
2. Then we construct a table listing all pairwise differences (the distance matrix).
3. Then we construct the tree from their pairwise distances.
Let’s take a simple, made-up example to get the idea across:
Gorilla: ACGTCGTA
Human: ACGTTCCT
Chimp: ACGTTTCG
From this alignment the idea is we count the genetic “distances” between each pair of sequences. The distances in this sense is simply the number of mutations separating them. I’ve highlighted the differences in bold here to make it easy to see.
So we make a table like (the distance matrix) this:
……|..Go..|..Hu..|..Ch..|
Go..|…-….|…4…|…4..|
Hu..|……..|…-…|…2…|
Ch..|……..|…….|…-…|
I trust you can see that I have counted correctly. Four mutations separate Human (Hu) from Gorilla (Go). So we put a 4 in the tabe. Four mutations separate Gorilla and Chimp, so we put a 4 in the tabe there too. And two mutations separate Human and Chimp.
Now we basically forget all about the DNA sequences and work exclusively from the table we made and use this to draw the tree, where the length of the lines on each branch corresponds to the distances on the distance matrix. Four “steps” should separate human from gorilla. I attached a nice little picture of the tree for you that I drew.
There you go. There’s more (such as finding optimal branch lengths, and getting rooting information, but that’s enough to get you started). But that’s really the basics. Count differences, put them in a table, draw trees with branch lengths corresponding to the “distance” in number of differences.
If you’ve read a book on cladistics, why are you still so incredibly ignorant that you think some asshole putting microprocessors on a line constitutes an evolutionary tree?
That wasn’t an answer. Any and all similarities don’t create a nested hierarchy. Some clouds are similar to other clouds in all sorts of ways, yet there is no nested hierarchy of clouds. Please try again to answer the question. How would a nested hierarchy of mitochondrial DNA sequences arise other than through common descent?
Again, way too vague to be an answer. I think you can’t answer, since you have no idea what mitochondrial DNA sequences are.
Your strawman again. “We can draw a tree” isn’t what I’m saying at all. Yes, you can draw a tree on anything. That’s not at issue and it’s irrelevant to the subject. But in the case of phylogenetic trees, the data all conspire to force one particular tree. It’s the consilience among different bits of data that’s the evidence for common descent, not “We can draw a tree”.
Now, what presuppositions am I failing to examine?
I apologize. He actually is speaking about universal common descent of languages, though he acknowledges that operationally it can never be determined. But isn’t that a reasonable inference from H. sapiens once having been a small and local population? If that population had language, which it must have, must all languages not be descended from it? Still, why should “all languages” be mapped to “all life” rather than to, say, “all animals” or “all vertebrates”?
At any rate, he wasn’t talking about universal common descent of life in the origin. And it isn’t so much as hinted at in the OP. Finally, it’s irrelevant to relationships among primates, or birds, or anything other than the totality of life.
Get this: The point of your exercise was to demonstrate how I have a wrong idea about building trees, whereas you have the right idea. You didn’t demonstrate this. You only made it blatantly clear that I have had the right idea all along: More similar things belong together in the tree, things with more differences belong further apart.
And all this was not the main point. You totally missed the main point. The main point is this: How can you tell that the tree represents DESCENT, not just differences? Because, you see, you only counted differences. That’s all you did, nothing else. So what justifies your belief that the tree represents common descent?
Rumraket’s exercise showed that just counting the differences (that’s differences as in plain differences, not even homologies or anything, no assumptions needed about the differences, simply count the differences) is sufficient to produce a hierarchy. So right back at you: How do you draw a tree based on similarities that does NOT produce a nested hierarchy?
The presupposition that the tree you are drawing really represents phylogeny. Where did that assumption come from? Based on what?
It was a poor example, and of a method that’s very seldom used. In fact almost all methods of phylogenetic analysis do not reduce to “similar things belong together”. If you were really interested, you could look up least-squares fit, parsimony, or maximum likelihood phylogenetic methods. None of them rely on clustering of the most similar species. They rely on finding trees that best accord with the pattern in the data.
You don’t. A tree is by definition a nested hierarchy. Again, for perhaps the 20th time, the ability to draw a tree is not evidence of nested hierarchy in the data or of phylogeny. It’s the strong preference of the data for one tree versus all others that reveals a strong nested hierarchical pattern in the data and thus provides evidence of phylogeny. Rumraket’s example would not demonstrate any such strong preference.
(21) It’s not the ability to draw a tree that counts as evidence. It’s the strength of the hierarchical pattern in the data. So far you have been able to suggest no alternative explanation, just vague hand-waving. Until you come up with an explanation for nested hierarchy that’s as good as or better than phylogeny, we have to provisionally accept phylogeny. That’s how science works.
How do you know the Algonquian languages are related by descent?
I’m happy to see you finally get how a tree is actually made, which raises the question why the fuck you’d bring up that chronology of microprocessors as if it constitutes a phylogenetic tree.
We’ve covered this already for fucks sake. A single tree can’t by itself imply descent. You need multiple (that means at least two) independent (meaning one is not the cause of the other, for example morphology is not the cause of metabolism or vice versa*) data sets to construct trees from, and the entitites in question (the ones who’s attributes you use to draw the trees) must of course already be capable of reproducing and as such, predict the emergence of independent lines of data that yield the same tree.
That is what implies descent. Why does it do that? Because there’d be no reason to expect independent attributes to yield the same tree if there’s no mechanism that constrains them to the same tree pattern.
* Or the particular sequence of metabolic gene A, is not the cause of the sequence of metabolic gene B or vice versa. For example, the sequence of your cytochrome C gene is not causally linked to the sequence of your GAPDH gene. So if we construct a tree using cytochrome C sequences from several different species, there’d be no reason to expect (unless they share common descent) that a tree constructed from their GAPDH gene, to yield the same tree.
Is it sinking in yet?
You asked me how to make trees, I told you. Now you know.
See above.
In all honesty, I took a course in computational molecular evolution about 3 years ago. Never used it and barely remember it. Had to go back over some crappy notes I took and I actually had some on maximum likelihood and least-squares, but don’t remember anything about it even with the notes. The only one I could sort of glean from the notes alone was the distance matrix 😛
Actually, Rumraket’s analysis was a sort of example of a least squares fit, though because the data were perfect no fancy math was necessary. Further, since there were only three taxa it required an assumption of a perfect molecular clock to root. If we add a 4th taxon as outgroup it would be easy to construct an example to show that he wasn’t just clustering by similarity. For example, we could find distances represented by this tree, where D is the designated outgroup, A and B are most similar, but B and C are closest relatives.
Got it already a long time ago. Multiple trees on different levels of data do not change the principle: The ancestors are unobserved, so evolution is unobserved. All you have is the assumption projected on the tree(s).
Darwin had a different explanation besides evolution. He rejected the different explanation, but for no good reason. And he posited evolution also for no good reason. (I’m talking about “good reason” insofar as metaphysics is concerned, and metaphysics is relevant here because that’s the level where Darwin made his point. On the empirical level his reasoning was vaguely inductive only, insufficient for his conclusion. That’s why he brought the analogy with languages – empirical biological data was insufficient to his point. I’m sure we both agree to not go into metaphysics, because you don’t do metaphysics, even though Darwin’s point was essentially that and my objection to it is crucially that.)
The task here is to find out if you have a better reason.
Let’s take your pic and proceed by baby steps so that I don’t miss anything. If you bear with me, it will be good. Given your expertise, you think that spouting some technical terms should convince people, but this is not how it works with me. To me your tactics removes any chance to verify if your conclusions are warranted. And this has been going on far too long.
Questions concerning your pic:
Letters A, B, C, D are taxa, right?
Numbers 1, 2, 3, 5 are characters, right?
Whatever happened to number 4? Did its line go extinct and this is why it’s omitted? If included, where would it be placed? Because if it would be placed somewhere on the line C, I don’t see how you could call C and B most closely related while A and B are most similar.
I have more questions, but let’s see how you answer these first.
Because, first, prior to having drawn any trees or isoglosses on linguistic data, we know by philological evidence that languages change over time, current languages descend from earlier languages and they can borrow stuff from each other. One language can supplant another, with or without traces left after the fact. All this is known prior to drawing any trees. The problem with the theory of evolution is that we know nothing like this about species.
As to Algonquian languages specifically, you are likely asking how linguists proceed to verify the relationship of a group of languages they are encountering for the first time. The first thing to do is to collect the relevant data, i.e. to describe the phonology, morphology, vocabulary, syntax and semantics to a sufficient extent.
Syntax does not have many elements to it and is easily borrowed in language contact, so similarities in syntax, particularly broad typological features, are considered insufficient to prove organic relationship, particularly when they occur with languages that are in geographical contact.
The same apples for vocabulary, except here a distinction is made with core vocabulary versus the rest. General linguists like to find out universals and there is a wordlist called Swadesh that contains pronouns, words for body parts, basic numerals, and key verbs and nouns that are considered unlikely candidates for borrowing. This provides an easy method to compare languages with, but in my view it’s too easy. I don’t think there are fixed language universals (there are more abstract categorical ones). In my view, the core vocabulary must be constructed separately for each part of the world by areal comparison in order to see what concepts in fact are key in that area. Language contact in different areas, given different geography and culture, has different nature. Additionally, when comparing words across languages, you must observe the meanings of the word. Non-specialists tend to forget this and sloppy or overenthusiastic linguists tend to overlook this and end up cherry-picking and jumping to conclusions. Additionally, words across languages that have different surface representation have to be connected by plausible sound laws in order for the languages to be organically connected. Those sound laws yield a reconstruction of sound history that provides a relative timing of phases of development, an extremely valuable tool to determine either organic connection or the direction and timing of borrowing. Words across languages that are superficially the same are usually very recent borrowings and imply no organic connection, or, if all the words are superficially the same, we are simply studying a single language, not different ones.
Universally the most reliable level is morphology. If morphological material and typology (along with the relevant meanings, duh) are shared, then those languages are related. End of story, pretty much. If other levels are shared, it simply confirms what the morphological connection already implies. If other levels are not shared much, while morphology is, there will be much scrutiny of possible phases of internal development and/or borrowings. Exception would be with languages that basically have no morphology, such as Chinese and Vietnamese. There you have to rely on phonology and syntax. (Wikipedia says linguistic morphology is to do with words. I was taught differently: It’s to do with morphemes, conjugation, declension, and derivation, and I’m talking the way I was taught, to keep things simple. The terms for words are vocabulary, lexicon, not morphology.)
Phonology is the “prime matter” of linguistics. All other levels of language have material existence by virtue of consisting of sounds. However, there are some tricky points. First, sounds can change, so no specific shared sound shows any necessary connection between languages, just like no specific shared word shows any necessary connection. Shared phonology (sounds as a systemic entirety) with plausible explanations as to the differences may show a connection. Second, sounds have no meaning. Morphemes (meaningful sequences of sounds) do. At the same time, there may be units of meaning that are not materially represented by anything (i.e. the surface-level representation is ø). The existence of such units of meaning is detected by systematic grammatical analysis. When the analysis does not yield those categories, then it’s usually not justified to say that those categories are materially there.
Over time, language may cange by internal development and/or contact so much that the material data remains inconclusive. When the evidence does not allow to posit an organic language family, then it’s not posited. Where there’s no evidence, there are no claims of common descent.
Or you can draw trees on shared broad typological features (such as “has SVO word order” or “has voiced-voiceless opposition in plosives”) to connect all languages of the world and you may think it proves something. In this sense Darwin’s analogy applies, but professional linguists do not operate on this ridiculous standard.
You really don’t understand science. It’s all inference. Ancestors are unobserved, but so are atoms. So is fusion in the sun. So is the bond structure of a benzene ring. (Well, that last was observed fairly recently, sort of, if you accept scanning tunneling microscope images as “observed”. But we knew about it long before that.) Phylogenetics is no different.
What different explanation are you thinking of? Is there a reason you’re unwilling to trot it out and defend it?
No, that isn’t why. The analogy was to clarify why the empirical biological data supported his point.
Since you have no clue regarding Darwin’s reason or my reason, I don’t know how you’re going to address the task.
If you would just take the trouble to learn the least little thing about the subject you’re pontificating on, you wouldn’t consider the technical terms an obstacle.
Yes, the letters are taxa. No, the numbers aren’t characters. They’re the lengths of the tree branches. You should also imagine a distance matrix among A, B, C, and D in which the distance between two taxa exactly matches the sum of the lengths of the branches between them. For example, A is (2 + 1 + 5 =) 8 units from C. See Rumraket’s prior example, of which this is just an elaboration. There is no number 4. B and C are most closely related because that’s what the tree shows: they share an ancestral node not shared with any other taxon.
Of course we do. We know that species change over time. We know that current populations descend from earlier populations. We know that there is horizontal transfer and that one species can replace another in a locality, though neither of these seems relevant to relationships among species or within Algonquian.
So you didn’t observe the ancestors of Algonquian, so by your reasoning you can’t say the languages are related. Or are you exercising a double standard?
Once again you reduce phylogenetics to universal common descent or nothing. OK, so since there’s no good evidence that all languages are related, I can confidently affirm that no two languages are related. Now in fact there are all sorts of features of genomes that change at different rates. Some of them are only useful for shallow comparisons, some only for very deep comparisons, like your “core vocabulary”. The latter turn out to be conserved enough to test universal common descent of life. In this way the analogy to language breaks down, but it does so in a way that benefits phylogenetics. We have the data for universal common descent as well as descent within narrower groups. We can discuss either, if you like.
But you still haven’t thought of an alternative hypothesis that fits the data as well or better. The simple explanation is that you can’t.
Erik,
This is pretty obtuse. One has, on the one hand, ‘variety pairs’ such as the Common and Spotted Sandpipers. As you have discovered, they can interbreed. Therefore, on your criteria, they are ‘related’. On the other hand, there are many similar pairs that cannot interbreed, yet their differences in genetic sequence are pretty much of the order of those between the Sandpipers. There is a continuum of degrees of interfertility, not a dichotomy, likewise a continuum of genetic difference. The two are indeed correlated, since the final checkpoint is next-generation meiosis; perfectly viable hybrids are often infertile because the two parental genomes are simply too different – successful meiosis depends on a significant degree of sequence identity.
What is one to make of that data? How, indeed, did Sandpipers get to be different despite being – on your criteria – related? Why would one reject the perfectly logical possibility that the inability to interbreed in those cases that display it is simply due to an amount of additional divergence beyond the bimodal state found in those Sandpipers?
But for fucks sake, the question is if species which exist now and which can’t interbreed, once evolved to that point, from common ancestors. We already know they can’t interbreed now. So if they can’t presently interbreed, that cannot constitute a falsification of the very hypothesis that they became unable to do so through independent divergence from a common ancestor.
Rather, the only way to test that now (since we can’t travel back in time and directly observe historical events) is to try to figure out if the theory of common descent of some two or more species, make certain predictions that can be tested observationally. Is there something we should uniquely expect to find, if species A, B and C share common descent? Yes there is. Independent data sets constructed from heritable attributes should be able to be arranged into statistically significantly congruent phylogenies. This is uniquely a prediction of the theory that species A, B and C share common descent.
Rumraket,
The fly in Erik’s ointment, though, are those groups which can interbreed but are different (eg the Sandpipers). He’s pushing hard-line species-immutablism (and hence an Ark that’s more of a Tardis). But how did they get to be different?
To be honest, I don’t even see what interbreeding really proves if you’re going to go that route. They were separately created (last Thursday) but can interbreed because they were separately created to be very similar. But different.
So it’s all inference, evidence be damned. Now I see where you are coming from. This discussion has been going on for too long with the wrong guy.
The support was only inductive in a small way. At the same time there are obvious counterexamples that you have been ignoring. The conclusion of common descent does not follow.
No. The option is between: (A) there is sufficient evidence for common descent so that it can be reliably posited or (B) there is no sufficient evidence for it. A slightly different matter.
Does not follow.
The tree of a language family is fundamentally different from the phylogenetic tree in biology. In linguistics, we know from historical written records (1) that languages change and (2) in what way that happens. The tree merely illustrates this knowledge and is irrelevant as proof. In biology, the phylogenetic tree is both the illustration of the theory and proof of it at the same time. There’s no other evidence apart from the tree that species actually evolve as claimed.
Circular reasoning. If you set out to prove common descent of species, then “heritable attributes” are among the things you have to prove, not assume.
Looks like you are not interested in having solid definitional criteria for species. Too bad.
Common and Spotted Sandpipers look near-identical. And they interbreed. Compare humans in Africa, Asia, and Europe. More superficial differences than with those sandpipers, but are they different species or the same? Chihuahua and St Bernard, same question. Sandpipers shmandpipers.
Erik,
It is a perfectly reasonable criterion (albeit one of many possibilities) for species, but insufficient to be definitional for the purposes of relationship, which is what you want to use it for. You can’t simply define relationship out of existence, for precisely the logical reason I outlined. It is possible that presently non-interbreeding species have a common ancestor. If not, why not?
But they aren’t identical, which is why they are given separate names. How did they come to be different?
Eric, the attributes used are already known to be heritable. Such as genes and their sequences. It was proven, they are heritable. You’ve heard of this field called genetics and this molecule called DNA right? And genes yield certain phenotypes. Such as eye color, or height, or the distribution of bones in your arms, and so on. We understand how heredity works down to the genetic level.
Since they are de facto heritable, we can use them to test the prediction of descent from a common ancestor. If they (these things we already know are inherited from generation to generation) derived from a common ancestor, we should be able to produce statistically significantly congruent trees from independent genetic and/or physiological/morphological data.
Hehehe….good one.
I take it back. Some of us understand it. 😀
What do you mean “evidence be damned”? You don’t seem to understand that inference is from evidence, and that all science is such inference. For a linguist, you aren’t very good at understanding words. You have no idea where I’m coming from because you start from the assumption that what I say makes no sense and change my words in your head until they fit that assumption. Let me repeat: all science is inference from data. Even historical linguistics. All the same.
What are these obvious counterexamples? Are you talking about the cups and vases? If so, that’s back to your confusion between your ability to force a tree onto the data and the ability of the data to force a tree at you.
Again you resort to vagueness. Common descent of what? There is insufficient evidence for common descent of all languages, but there is sufficient evidence for common descent of Algonquian. The same sort of evidence that works for Algonquian is also available for lots and lots of groups of species, including the group of all life.
Of course it doesn’t follow. That was a demonstration of the absurdity of your claims about life, by analogy to languages. And there is no real difference between language and biology, at least not in the ways you claim. We know from historical records that species do change and in what way that happens: mutation, drift, selection. The differences between species are exactly the same sorts of genomic changes as happen within species and can be observed from generation to generation.
Speaking of history, there are no historical documents for Algonquian, yet you confidently assert that it’s a language family. You have no observations showing one Algonquian language changing into another. Nor do you have a continuous record showing Indo-European languages changing into other Indo-European languages. You have only a limited number of fossil snapshots, mostly clustered toward the tips of the tree. You are just unwilling to accept the same standard of evidence for biology as you do for languages. (In fact the biological evidence is better than the linguistic evidence in a great many ways.)
The tree isn’t proof, as I keep telling you. The data from which the tree is made are “proof” (bad word in science; linguistics too, but let that pass). You can’t explain the data, and you haven’t even tried.