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.
Common descent implies vertical inheritance. A binary tree is exactly what we would expect to see if CD is true. If that wasn’t enough, you can examine different independent lines of evidence. Morphology, embryology, etc… They all happen to support the same tree of life. Not just some nested hierarchy, the SAME one. What are the odds?
You won’t get much stronger evidence in support of any other theory out there.
Try to apply your ridiculous hyperskepticism to gravity: oh. micro-gravity works. Apples fall from trees and stuff, but do you have evidence that tiny micro-gravitational events can accumulate over millions of years to produce large planets? Macro-gravitational accretion has never been directly observed! God could have simply created those planets so that they look like they were formed by accretion!
There is no line. There can’t be. It’s a gradual process. Evolution 101
Interesting idea. I would suggest that the rules of the game require all details of how each data set was created to be placed in escrow (simply posting it PGP encrypted with the private key to be revealed at the end of the game would work fine).
A somewhat unrelated question for the knowledgeable please. Are (some?) paternity tests based on Joe’s bootstrapping?
Erik,
Nope. Not the same. Which is why I invite you to screw the language analogies. Human language has a particular constraint: the range of sounds capable of being made by the human voice. But equally, given there is no task for language beyond communication (if one includes one’s interior monologue), a well-formed linguistic string can be anything the human can get its chops round. It is not under the same constraints as a genetic sequence, which cannot be just anything. Yet, as far as mapping is concerned, the amino acid translation assignments are not specially constrained – there is no chemical constraint that says AUG must be methionine. On the face of it, these assignments have stayed constant for 4 billion years, since the genes that generate these assignments are highly conserved – highly similar. The apparent reason is that there is little lability in descent, not that there is a constraint at origin.
You can’t compare universal sequences of hundreds of amino acids with the sound ‘ug’. You are comparing a system in which a phylogenetic signal has little chance of retention even over a few tens of thousands of years with one in which it has persisted for 4 billion years, and saying they are both equally uninformative. That’s why analogies have the power to mislead.
And why isn’t it good for examining phylogenetic data?
The line is conventionally defined at the level of species, with speciation being intermediate. Microevolution is change of allele frequencies in populations. Macroevolution is anything other than that. There is some confusion because there is another definition of macroevolution commonly understood by the public, i.e. big changes in morphology. But that could simply be, and probably almost always is, just accumulated microevolution.
As for the evidence of macroevolution, Erik has ignored every attempt to provide him with it.
Not that I’m aware. And I don’t see how you would apply it.
The fact that we can re-engineer the code to add non-natural amino acids confirms that there is no strict constraint at origin.
It’s common design. Obviously.
Yes, like the evidence that we have for biologic universal common descent. Real linguists would recognize universal common descent of language if it just had the sort of evidence of derivation that exists in biology, and even your whole ridiculous complaint about Darwin’s overuse of analogy (true, but hardly important) depends on exactly that.
It can show correlation of cause and effect. Of course you need more than just statistics to indicate what the cause is (who suggested otherwise?), but once you know what the causes are you know what the probabilities are and can test those using statistics. That’s the point of the genetic evidence, it shows the correlations expected if evolution processes occurred–like the trees.
No, you’re absurdly ignoring the fact that statistics can test causes. You’ve also failed completely and utterly to show that any other cause would be at all likely to produce the correlations in the data. Of course you’ve ignored all of the specific examples given.
See, this is just it, no real expert is dumb enough to think that the two are different processes at the level of inheritance (that is, the same sort of data show derivation with or without speciation). That’s your belief in pseudoscience and ID/creationism (deny it if you like, you’re using the same dumb “arguments”).
So? Mostly we don’t even care if it’s universal common descent, any more than Darwin did when he suggested it could be one or more forms of life that were created, then evolved. We just follow the evidence that indicates universal common descent.
No, you clearly understand almost nothing about the matter.
Sorry, it’s the causes that require that macroevolution must cause the patterns seen. The causes of macroevolution are limited that way. No other sort of cause (like intelligence) has the limits that known evolutionary processes do in biology, so have no cause to produce the same sort of patterns of derivation.
Glen Davidson
Thanks John. Is that’s because bootstrapping is used to evaluate nodes in a phylogenetic tree, where you start with many sequences, while paternity tests only involve two sequences?
I know. That’s why it’s unacceptable. Because in reality, there are lines all over the place, both fine and coarse.
It is okay to examine phylogenetic data by means of statistics, but you have to have established by some other means that it is phylogenetic data. Human and chimp may consistently land close in the tree, but statistics does not tell whether the tree represents a family or something else. Twins and lookalikes are both equally similar, but lookalikes do not belong to the family, while twins do.
So yours is an a-priory rejection of the theory. There can’t be a gradual pathway, therefore evolution is bunk. Absurd
Worth drawing the distinction between an ‘origin’ and the furthest point we can see by lumping all extant Life into a big bucket and seeing where it coalesces. They are different points in history.
I think that there was indeed a single cell – one of millions at the time – worthy of the retrospective designation LUCA – Last Universal Common Ancestor. If ‘vertical’ LUCA was contaminated by a bit of HGT, you’d need to go back a bit to the common ancestor of the donor and recipient.
None of this says there was definitively one original replicator at the Very Beginning, ancestral to LUCA and her contemporaries. But ‘she’ is the only one to have left modern descendants.
Bootstrapping is just resampling with replacement to generate some idea of the shape of the underlying distribution. You could do it with a single sequence. The question is what that resampling would tell you in any particular case. I don’t see what it would tell you in a paternity test.
Erik,
If you were presented the data blind, you would not be able to tell if there was such an objection or not. As I’ve said before, if I give you 2 sequences from up to 2 species, but don’t tell you whether it was 2-from-1 or 2-from-2, you would presumably tentatively conclude common descent from the data given. After all, they might be Spotted and Common Sandpipers, and we don’t have a problem with them. Then I pull back the curtain, showing that the possessor of one was much hairier than the other, and you recoil, gagging. “Aaaargh! Need … more …. info … !”.
What causes extensive sequence identity?
There’s a subtle point where you are wrong and that subtle point happens to be exactly analogous in linguistics and in the evolution theory.
You say, “a well-formed linguistic string can be anything the human can get its chops round”. In reality, a well-formed linguistic utterance is grammatical. That’s what it means to be well-formed in language. Over time, it can change theoretically into anything, but it must remain grammatical at all times. And due to required grammaticality, the changes are subject to laws, they are not (entirely) random.
Then you say, “It is not under the same constraints as a genetic sequence, which cannot be just anything.” Right, the genetic sequence must fit with whatever else is in place. But again, over time it supposedly can theoretically change into anything, right? According to some laws of nature, I assume.
At least for me, the analogy has been very informative and an excellent guideline.
Very good. Does genetic evidence show the causes too, not just correlations? Because what I am reading just above:
If Felsenstein has been programming phylogenetic trees all his life, trees intended to outline the true genealogy of all nature, it should be directly applicable to parenthood tests. If not, why not?
Thank you. Much better than Joe Felsenstein’s answer earlier.
What could that “something else” be? You need at least one alternative hypothesis.
You are talking about the subject as if you knew something about it, but in ways that show you don’t. If the best way to get from London to New York is by jumbo jet, does it follow that the best way to get from the living room to the kitchen is also by jumbo jet?
It reveals correlations between causes and effects.
That’s why genomes are studied.
Glen Davidson
Erik,
One can change every syllable, gradually, and word order, and it still remain ‘well-formed’. But all phylogenetic signal would still be lost, in a few tens of thousands of years. The constraint of grammar is not equivalent to the constraint of an amino acid assignment. More like you can’t change ‘ah’ to ‘uh’ because to change it in one word forces the same change in every word (and it’s lethal!). That would be an equivalent constraint, and does not happen. Your analogy is poor. The basic analogy between evolution and language is a good one, but over-extension of it, and consequent misunderstanding of points of equivalence, a waste of intellect.
This is why I urge people to drop analogies for anything more than illustrative purposes. Entire threads become about whether X in one is equivalent to Y in the other. It does not matter.
Your questions make zero sense in the context of what you quote.
Yes. Macroevolution is evolution above the species level. Microevolution is evolution at or below the species level.
Here’s 29 Evidences for macroevolution, for the tenth time.
Still none of this commentary makes ANY sense in the context of the piece of my post you quoted.
Erik, are you even fucking aware what you are trying to argue that this point?
Yes you are. Directly and literally, that IS what you postulated. You postulated that the evidence for common descent can be equally accounted for by some insane fucking ID/creation plan.
Here are your own words: “I’m fully aware that the expectation of a pattern follows given a particular hypothesis. Such as the hypothesis of a creation plan. Or ID theory.”
In the context of what you are responding to with those two sentences, they only make sense as a claim that the pattern can be accounted for by a “creation plan” or “ID theory”.
I have now explained why that is utterly ludicrous. And you seem to have gotten the point, because you’re now trying to say you didn’t mean to postulate anything like that at all.
Fine, I accept your concession.
What was Darwin “up against” and how is any of your crap a “report”?
No, you’re responding to us, and you’re somehow hellbent on denying the principle of inferences about historical events from observed processes and statistical hypothesis testing.
What is meant by a “tangible causal pattern”? You’re not making any sense, you’re just blathering to avoid direct concession.
Apparently you don’t, because the very argument you erected to try to deny the inference of common descent, attacks it by attacking inferences, thereby undermining all historical methods and the principle of inferential reasoning itself. Yet you simultaneously seem to be just fine with making inferences, just not for common descent. Which means you’re cherry picking, or have some sort of double-standard.
The particular case of inferring from nesting patterns of shared and derived characteristics that they were produced by a process of descent with modification. As applied to both the case of closeley and distantly related species.
This complete irrelevancy has absolutely nothing to do with what I’m talking about. I’m talking about inferring the causes of the existence of certain patterns and objects for which we were not around to see how they originated. I’m showing that you, in fact, also accept inferences about past events, without having directly observed them happening, by reasoning from the present.
I readily concede I cannot show it to you now that you’ve made it clear you don’t want to see it.
For everyone else, here: 29+ Evidences for Macroevolution
The Scientific Case for Common Descent.
Please demonstrate that in the same way you want macroevolution to be demonstrated.
Were you there?
And immediately next I said, “Is this how you intended to come across?” Obviously, I meant that you’re coming across like an IDist or creationist. And you replied, “Yes, that’s how I intended to come across.”
Fine.
To infer the causes, we must know that the things we are talking about are subject to that sort of causality. This is the very point of contention. The existence of the pattern by itself says nothing.
A human and an ape may very much look like related. Well, so do cup and vase. Like cousins. If you say this comparison is inapplicable, then why and how? The “pattern” is the same, isn’t it? What’s the difference? Obviously, something more is needed than the “pattern”.
The standard is not “Were you there?” The standard to infer that humans invented writing is the general knowledge that humans observably learn to write at every generation.
Similarly, to infer macroevolution we should know somehow that species do that sort of thing, that they morph into each other all the time no problem. But in reality, we observe stability of species rather than morphing. There’s morphing only up to a limited point and it tends to reverse itself as if variability revolved around an epicenter. There are quite clear features of normalcy, e.g. breeding is an intra-species thing, not across species. These are observable barriers to macroevolution. There must be something demonstrable to explain these barriers away, to justify the contention that it’s more natural to fail to observe those barriers than to observe them.
Oh, really? This could make you famous Erik. Go head, what are those?
A curious observation – that I’m not trying to turn into a derail – is that I’m assuming this is the same Erik that insisted that ‘universal’ Flood stories were indicative of a single historic Flood – ie, Common Descent.
That seems inconsistent, though I’m aware I might appear open to the reverse accusation.
You are oversimplifying again. The change is legitimate only if the sequence remains grammatical at each and every point. You can’t change it just so and expect it to be grammatical. Grammar is regular, a set of rules, accordingly legitimate change is also regular, lawful. When it oversteps the boundaries, it’s not well-formed.
I guess you oversimplified here too. Under which conditions would the phylogenetic signal be lost? Does it happen often?
Erik,
Do all languages have the same grammar rules? Why don’t we have the same grammar as Latin or German, our progenitors? Are you saying that there is no gradual transition in linguistics due to the grammar constraint? That leaves the known lability of grammar itself as a bit of a difficulty.
You completely missed my point that this is not a good analogy for the universality of a codon assignment, which is the same in ALL proteins and, furthermore, the sequence underlying the assignment is conserved throughout life. These are important observations not shared with language.
Under conditions of change, where the similarity has eroded to the point of being unrecognisable. There is less data in a syllable than a protein domain, and so loss is bound to happen more rapidly. Loss of signal is much easier in the case of a set of 1- and 2-syllable words than in the hundreds-of-base DNA sequences that underlie the genetic code. Particularly given that there is no lethality attached to change.
Probably. Certainly Chaucer’s English is barely intelligible to us. That’s 700 years or so. Multiply that by ten. What do you think might happen on geologic timescales? Again with the misleading analogies.
And now recall what you were trying to prove with this point, if anything. Something about common descent, I hope.
Erik,
I was answering a question of yours, you dope! If you don’t consider loss or retention of phylogenetic signal relevant to a discussion of evidence on common descent, I’m not sure I can be of much more assistance.
From my point of view, you brought it up pretty much out of the blue and I asked questions to try to figure out if it’s relevant to anything.
Macroevolution and common descent are questions about causal links and a statistical measure is irrelevant to such questions. This is where we stand right now.
But supposing, for the sake of discussion, that phylogenetic signal proves or at least indicates evolutionary genealogy, then we can move on to next thing: noise. Or is the signal so loud and clear that there is no noise in the data? Whenever you measure the signal, all doubts concerning where to place the data on the tree are resolved? What is the data that underlies the signal? And how do you determine retention and loss of the signal?
Yes, there is noise. That’s what Allan was talking about when he referred to loss of signal. If, for example, Phe changes to Ile and then to Trp, you have lost whatever signal was present in the change from Phe to Ile. Any phylogenetic analysis must account for multiple hits and homoplasy.
Your question is poorly formed. The data are sequences, protein or DNA, and they are always placed at the tips of any tree. There are never any doubts possible concerning where to place them. You perhaps refer to doubts regarding the shape of the tree, i.e. the relationships among species. As with anything in science, all conclusions are provisional and doubt can never be completely eliminated. We can however eliminate them for all practical purposes as the strength of support for a hypothesis increases.
Once again an unclear question. The data are observed protein and DNA sequences. The signal would, I suppose, be the pattern of similarities in those sequences among taxa that point toward a particular tree.
I suppose the signal would be considered retained if the data fit a particular tree much better than they fit any other. Or, to put it another way, if different subsamples of the data point toward the same tree (bootstrapping, which has been mentioned here, is one way of testing that). The signal has been lost if no tree is strongly favored over another.
Erik,
Out of the blue? I’m pretty sure I was prompted into discussing the similarities and differences between language strings and genetic ones by your good self.
Erik,
One relationship’s noise is another’s signal.
Since you like analogies, think of a photocopier where you take a single white start page and then create sets of copies, writing down the actual copying ‘tree’ as you do so. If you spill random ink blots onto pages occasionally, gradually the white area will diminish in all lineages. One might think of the unblemished part of sheets as signal – it derives from the original white sheet – and the blots as noise. From the point of view of the original sheet, that’s true; relationship evidence is gradually being eroded. But the patterns of the blots can be used to reconstruct the more complex phylogeny, because a particular blot of a particular size will be shared by all its descendants and no others (barring coincidence). The signal for that relationship is the paper with blots, even if the blots are noise in another context. So, changes act as both noise and signal.
Allan Miller,
From your analogy I take it that you consider noise mutations accumulated by both parents germ lines prior to conception. When the child is conceived then those germ line mutations convert to signal. Do you think this is correct for all types of mutations?
You take it wrong. First, we aren’t talking about individuals here. We’re talking about species and phylogenetics. Second, we aren’t talking about mutation but fixation. Noise in phylogenetics would be any states in any pair of taxa that are the same for some reason other than inheritance from their common ancestor: convergence, reversal, horizontal transfer, incomplete lineage sorting, or just mistaken coding.
Deftly missing the point.
Why not go back to the post I linked to and explain your use of quotes. Surely you did that in order to emphasize the term should not be taken literally.
It should also be pretty non-controversial in closely ‘related’ species.
LoL.
You really don’t see it.
Species do not have ancestors, and species do not have descendants, at least not literally.
What barriers are those? I too would like you to describe them and provide your evidence for them.
What magic barriers makes it impossible for microevolutionary changes to accumulate over time into macroevolutionary ones?
Adapa,
If you read the whole paragraph he answers your question.
Perhaps you should have looked a little longer at the sentence just before the point you quoted.
A male squirrel and female squirrel copulate and a baby squirrel comes out, not a turtle or whatever. A squirrel and a turtle copulate, nothing comes out. Does not look like an observable barrier to macroevolution?
If evolution aims at the thing called reproductive success, then why is breeding an intra-species thing? If all species are all related as if cousins, why doesn’t breeding occur regardless of species? Why this arbitrary counter-productive restriction? What is the evidence or reasoning that this restriction is merely an apparent barrier to macroevolution and not really there?
Mung,
They do. Literally. You are just wrong.
I used scare quotes round ‘related’ in that context to emphasise I was avoiding assuming the conclusion – when Creationists talk of related species they may not mean actual genetic relationship, hence I acknowledged this by scare-quoting the term.
But if there is common ancestry, it is not common ‘ancestry’.
Erik,
What’s it counter productive for?
Anyway, it isn’t arbitrary. There are numerous mechanistic pre-zygotic and post-zygotic reasons for incompatibility (even between two given individuals within a species).
There is also ecology to consider. If 2 species have specialised to 2 niches, hybrids may be less fit because they fall between 2 stools.
Continued compatibility irrespective of degree of divergence is not a sensible corollary of common descent. Once speciation*** has occurred (ie, macroevolution), they have already crossed the incompatibility threshold, it should be obvious why they don’t subsequently re-hybridise at even greater taxonomic distances.
***Speciation, to note, being the stemming of gene flow between the two parts of a divided population, not ‘dog-into-cat’.
Erik,
*Forehead slap*
By God, he’s right! Macroevolution is impossible!
LMAO.
So a cat doesn’t give birth to a dog?
Is there any creationist canard too low for you, Erik?
Glen Davidson
Hahahahahaha
That’s it, I’m out.
Look, it’s not unthinkable that an obviously intelligent guy has spotted a fundamental flaw in a field that thousands of people have spent their entire lives progressing.
It does happen. From time to time. Is this one of those times?
I’m taking bets, anyone in?
Then where do squrtles come from?
Erik,
Perhaps go read this: https://whyevolutionistrue.wordpress.com/2014/07/16/there-are-no-ring-species/
While it’s a bit more complex then that excerpt suggests (go read the rest of it) that seems to neatly answer your question regarding how speciation can occur if mothers always give birth to the same species that they are.
This might also help: https://www.youtube.com/watch?v=OXypyrutq_M