Falsifiability

[Almo]

http://cla.calpoly.edu/~fotoole/321.1/popper.html

Someone posted this excellent link.

A friend of mine had this to say:

I gotta agree with the dude on what he says---inductive reasoning is fundamentally flawed. Indeed, science is, in some sense, fundamentally flawed: we are always wrong (but "right enough" to be useful). I prefer science to other modes of thought precisely because it requires peer review by educated individuals (part of the reason I would listen with respect to a Jesuit minister, but really can't stand Protestant ministers). I don't think science is Truth, but it is close enough to be useful.

But, this doesn't ease my concerns about macro-evolution: I still see it as almost a psuedo-science (though with considerably more credibility). I think part of the problem is the rah-rah folks that support it (usually as an alternative to religion) just like other folks rah-rah for religion. But the main problem is the falsifiability---I just don't see how it can be falsified. It just doesn't seem testable. It's a *very* clever possible explanation--in fact, absolutely beautiful, but I don't see how there can be an empirical observation that contradicts it: folks will just use the old "god of the gap" argument. It's the Teflon Science: no criticism ever seems to stick to it.


I think he makes a good point about evolution. It does make testable predictions, but what experiment could truly falsify it?

What do you guys think?

[Matabiri]

This page (http://www.talkorigins.org/faqs/comdesc/section4.html) (amongst many others) offers ways in which various predictions made by evolution can be falsified.

[Tricky]

http://cla.calpoly.edu/~fotoole/321.1/popper.html

Someone posted this excellent link.

A friend of mine had this to say:



I think he makes a good point about evolution. It does make testable predictions, but what experiment could truly falsify it?

What do you guys think?
Well, evolution is supported by so many different things. You can't falsify them all at once. An example of falsifying macroevolution might be, suppose that before the days of DNA identification, you predicted that bat DNA would be similar to other mammal DNA. Then, after experimentation made DNA identification possible, you discovered it was actually closer to bird DNA. That would falsify the part that bats evolved from mammals rather than birds.

[KingMerv00]

http://cla.calpoly.edu/~fotoole/321.1/popper.html

Someone posted this excellent link.

I think he makes a good point about evolution. It does make testable predictions, but what experiment could truly falsify it?

What do you guys think?

1) The discovery of precambrian bunnies.

2) The discovery of a bird/mammal intermediate.

3) A shuffled fossil record.

4) ERV tests results come out different. (See link posted by Matabiri)

5) A vestigial organ that doesn't belong. (A frog with vestigial nipples or uterus for example)

6) And the rest.

[BillHoyt]

http://cla.calpoly.edu/~fotoole/321.1/popper.html

Someone posted this excellent link.

A friend of mine had this to say:



I think he makes a good point about evolution. It does make testable predictions, but what experiment could truly falsify it?

What do you guys think?
Macroevolution is change at or above the species level. Therefore, we'd need to observe a situation in which a new species ought to have arisen, but didn't. We've never yet observed such a situation; unfortunately we don't yet know exactly how to set up such a situation. On the other hand, though, we've already seen many new species arise.

[drkitten]

I think he makes a good point about evolution. It does make testable predictions, but what experiment could truly falsify it?


At some point, retroactive, "could-have-been" falsification is all that's really available, simply because the tests have largely all been done. Evolution is almost getting to that point.

As an example of what I'm talking about, suppose that I set an envelope in front of you and claim
that there's a Canadian hundred dollar bill inside it. This is obviously falsifiable, and there are many ways to falsify it. The easiest is probably just to open it and look inside.

So you do. And, lo and behold, you see what looks like a Canadian hundred dollar bill.

Is my claim still falsifiable? Well, I suppose it might still be falsified if the bill were counterfeit. So let's go to the bank and check it out. Guess what -- it's genuine. The bank says you're $100 richer.

Is my claim still falsifiable? (Maybe the bank teller made a mistake, but when three other banks agree with her, what then?) At some point, you'll simply have to accept that the bank -- and I -- am correct because the alternative hypotheses are so outlandish and far-fetched as to not be worth taking seriously.

Evolution is almost getting to that point. Historically, there are a thousand things that could have falsified evolution, of which the recent revolution in molecular genetics is probably the most obvious. If we had found that frogs were more closely related to rabbits than rats were, that would have seriously messed things up. Finding that the genetic code wasn't as universal as we thought it was would have done wierd stuff. Evidence of some actual macroevolutionary barrier would spoke the guns.

But for well over a hundred years, every possible falsification that anyone can think how to test has been tested -- and the theory of evolution, in broad terms, has passed every test. At what point do you admit that there really was a hundred dollar bill inside the envelope?

[Garrette]

You're nominated, drkitten. I've tried to find a simple and clear answer to this type of argument. I may copy it and quote it directly at some point if you don't mind.

[drkitten]

You're nominated, drkitten. I've tried to find a simple and clear answer to this type of argument. I may copy it and quote it directly at some point if you don't mind.

Of course I don't mind. And thanks!

[El Greco]

At what point do you admit that there really was a hundred dollar bill inside the envelope?

When you buy 100 dollar's worth of beer with it ?

Naysayers won't even bother to open the envelope, much less go to three banks.

If science is "fundamentally flawed", then what isn't ? What's the alternative, and how exactly is that falsifiable (as opposed to evolution for example) ?

[Luke T.]

You're nominated, drkitten.

Gagh!

I was just about to push the nominate button when I saw this. You beat me to it.

[Almo]

Thanks for taking the time to address this, guys. I will think about this for a while before responding. It seems a complex issue to me, and I want to understand it properly.

[pgwenthold]

As DrKitten implies, the concept of evolution right now is so broadbasedly supported that to claim it isn't science because it can't be falsified makes about as much sence as saying that gravity isn't scientific because it can't be falsified.

Now, there are many aspects of modern evolutionary theory that can, are, and will be falsified. For example, timelines for development of certain traits will change all the time as new information is brought to light. In principle, this is because the previously accepted timeline has been falsified by the emergence of new data. Therefore, aspects of evolutionary theory are tested and revised in the light of better evidence, just as is done in all areas of biology, physics, and chemistry.

However, to speak of the "falsifiability of evolutionary theory" is as meaningless as the falsifiability of atomic theory.

[Dr Adequate]

Evolution is almost getting to that point. Historically, there are a thousand things that could have falsified evolution, of which the recent revolution in molecular genetics is probably the most obvious. If we had found that frogs were more closely related to rabbits than rats were, that would have seriously messed things up
Great minds think alike:
Many things might have proved Intelligent Design, but they're all counterfactual.

For example, geology might have revealed a picture of the world entirely consistent with (for example) Genesis and Ussher's chronology, rather than revealing all those pesky intermediate forms. Morphology might have revealed that each animal had a Bauplan uniquely suited to its way of life; and might have turned up no examples of "Stupid Design", such as the recurrent laryngeal nerve. Genetic studies might have revealed no "tree of life" whatsoever above the species level. Biogeography might have shown no trace of the Sarawak law. And then I would be an IDer and a Young Earth Creationist to boot.

So many things might have confirmed Intelligent Design. But what would confirm it now? It's like being asked "What evidence would it take to convince you that the Earth is flat"? Well, if it had been, then pictures from space would have shown this ... but they didn't. It's too late now.

[Almo]

As DrKitten implies, the concept of evolution right now is so broadbasedly supported that to claim it isn't science because it can't be falsified makes about as much sence as saying that gravity isn't scientific because it can't be falsified.

Falsifiability isn't how difficult it is to prove false; if it were, you'd be right. The aritcle in the OP talks about falsifiability as how difficult it is to design an experiment with a possible outcome that would clearly disprove the theory. In this sense, gravity is very easily falsifiable. Measure the descent of an object in space. Gravity predicts its path VERY precisely. If the object went up instead of down, that would be it. The argument is that such simple experiments don't exist for evolution. I'm not asserting that this is the case, but that's the argument.

The example of Marxism in the article shows this. It's often easy to reinterpret observations to fit Marxism, if you're devoted to that means of defense. But if the object falls up instead of down, that easily falsifies gravity.

[pgwenthold]

Falsifiability isn't how difficult it is to prove false; if it were, you'd be right. The aritcle in the OP talks about falsifiability as how difficult it is to design an experiment with a possible outcome that would clearly disprove the theory. In this sense, gravity is very easily falsifiable. Measure the descent of an object in space. Gravity predicts its path VERY precisely. If the object went up instead of down, that would be it. The argument is that such simple experiments don't exist for evolution. I'm not asserting that this is the case, but that's the argument.


If that's the case, then it is blatently wrong. There is a simple experiment that could be run. Breed animals. If the DNA of the offspring were always identical to the identity of the parents, with no mutation occuring at all, then evolution would not be possible.

At some museum I saw a computer exercise where you were supposed to "kill" bugs flying around the computer screen. Some bugs were bigger and slower than others, and, as expected, after a given amount of time, you tended to have killed more of the bigger, slower bugs, and the faster, smaller bugs survived. This is in fact an evolution experiment. If natural selection did not occur, you would have had the same likelyhood of killing small or large bugs.

There are all kinds of trivial experiments that can be done to test evolution, and you don't need to go to outer space to do it.

[BillHoyt]

If that's the case, then it is blatently wrong. There is a simple experiment that could be run. Breed animals. If the DNA of the offspring were always identical to the identity of the parents, with no mutation occuring at all, then evolution would not be possible.

At some museum I saw a computer exercise where you were supposed to "kill" bugs flying around the computer screen. Some bugs were bigger and slower than others, and, as expected, after a given amount of time, you tended to have killed more of the bigger, slower bugs, and the faster, smaller bugs survived. This is in fact an evolution experiment. If natural selection did not occur, you would have had the same likelyhood of killing small or large bugs.

There are all kinds of trivial experiments that can be done to test evolution, and you don't need to go to outer space to do it.

You're discussing microevolution, not macroevolution. Macroevolution's falsifiability is the thing being questioned.

[drkitten]

The aritcle in the OP talks about falsifiability as how difficult it is to design an experiment with a possible outcome that would clearly disprove the theory. In this sense, gravity is very easily falsifiable. Measure the descent of an object in space. Gravity predicts its path VERY precisely. If the object went up instead of down, that would be it. The argument is that such simple experiments don't exist for evolution. I'm not asserting that this is the case, but that's the argument.


Well, the question is whether a replication of an experiment that's already been done counts as another experiment or not. In a sense, there are zillions of trivial little experiments that could be done to falisify evolution -- but they're all stupid.

Continuing your gravity example -- yes, if I drop a rock and it falls down, I have technically tested gravity. If I then pick up a different rock and drop it (and it falls down as well), then I've technically tested gravity again. I can do this for as many different rocks as there are on the beach, and when I've finished, I can start all over again with the first rock, because it might not fall down this time.

I could do that -- but, god, what a hell of a life.

Similarly, I could test to see if the DNA of this frog was more closely related to that rabbit than the DNA of this rat is. Assuming it isn't (as I expect), I could then try with that frog, until I've harassed and tortured every little froggie in the pond -- and then I move on to another pond. "God, what a hell of a life."

At some level, we've concluded to our satisfaction that all rocks behave pretty much the same. We've also concluded to our satisfaction that frog DNA is all pretty much the same, as is rabbit DNA, as is rat DNA. So I don't need to test every individual frog, rat, rabbit, hamster, trout. magpie, and orange tree. One frog, one rabbit, one rat -- and I'm done.

In fact, finding evidence that frog DNA differs substantially within a species -- after all, humans and chimps are different species, and they're 99% similar -- would in and of itself be strong evidence against the theory of evolution. But we've done that test, too.... and so far, frog DNA is all much of a muchness. So we get back to the "well, if this had turned out differently, evolution would have been falsified, but that didn't happen" issue.

[pgwenthold]

You're discussing microevolution, not macroevolution. Macroevolution's falsifiability is the thing being questioned.

Falsify microevolution and you falsify macroevolution.

It really is that simple.

[BillHoyt]

Falsify microevolution and you falsify macroevolution.

It really is that simple.
It would be if micro were falsified. If it isn't, though, one still needs to falsify macro separately. They are separate and separable issues. This is basic logic, by the way. The negative of if p then q is ...?

[pgwenthold]

It would be if micro were falsified. If it isn't, though, one still needs to falsify macro separately. They are separate and separable issues. This is basic logic, by the way. The negative of if p then q is ...?

This is not basic logic, because macroevolution is not separable from microevolution. You cannot have macroevolution without microevolution. Therefore, if the question is what experiment can you do to falsify macroevolution, then the answer is that any experiment that will falsify microevolution will do.

Macroevolution only differs from microevolution in degree, not in kind (no pun intended). AFAIK, biologists only consider macroevolution to be an accumulation of a lot of microevolution. It occurs by the same mechanisms as microevolution, but is just an accumulation over a larger number of steps.

[CurtC]

The wording that I've used when discussing micro vs. macro, is this: macroevolution is nothing more than the accumulation of a bunch of microevolution.

[Almo]

In fact, finding evidence that frog DNA differs substantially within a species -- after all, humans and chimps are different species, and they're 99% similar -- would in and of itself be strong evidence against the theory of evolution.

Here's a large piece of the puzzle, in my opinion. As the article in the OP pointed out, Newtonian mechanics makes extremely precise predictions. If the rock falls any direction but straight down, gravity is out the window. Yes, there are other effects, like non-inertial reference frames and stuff, but the basic experiment is very simple, and any deviations are small. We can even predict how far off the rock fall thing might be given other effects. I assert that given the precision of gravity's predicitons it would be easy to determine if an experiment had in fact showed a problem, as is the case with General Relativity messing with it.

But in the quote above, we have "...differs substantially within a species." The theory of evolution is not a mathematical theory that makes precise predictions. How much is "substantial?" How much of a difference would be necessary to indicate a problem with micro-evolution?

[BillHoyt]

pg, curtc:

I think you're both missing the formal logical fallacy here. It is called denying the antecedent. If (falsify micro) then (falisfy macro) This much is true. If micro were to fail, macro would also fall. But, if not (falisify micro) then not (falisify macro) is not valid. That is fallacious reasoning.

Seeing population changes take hold intraspecially is good. Not seeing problems with evolutionary theory intraspecially is also good. But it does not mean macro theory is okay. We've passed Go, but have not yet collected $200.

[CurtC]

I think we're answering two different questions. Pg was answering the question "what is an example of something that would falsify the theory of evolution?"

An example would be something that falsifies microevolution. It's not an exhaustive list, just one example. I think you're answering the question of what would be an exhaustive list.

[Lynx2174]

pg, curtc:

I think you're both missing the formal logical fallacy here. It is called denying the antecedent. If (falsify micro) then (falisfy macro) This much is true. If micro were to fail, macro would also fall. But, if not (falisify micro) then not (falisify macro) is not valid. That is fallacious reasoning.

Seeing population changes take hold intraspecially is good. Not seeing problems with evolutionary theory intraspecially is also good. But it does not mean macro theory is okay. We've passed Go, but have not yet collected $200.

but there's no difference in the theories. in fact, there is only one theory. if microevolution is true, then it follows almost obiously that macroevolution is possible. (though not necessarilly correct or how it happened; you could then go and falsify parts of natural history, and show that things did not evolve through macroevolution.)

but as long as microevolution can be taken as a valid explanation, then macroevolution is also a valid explanation. there isn't any seperate mechanism differentiating the two. you wouldn't be able to falsify it seperately unless you found some kind of effect preventing the micro changes from forming new species. and since IIRC the formation of new species has already been observed, that is right out.

[Kosh3]

Being falsified, strictly, by one point of evidence, can be rather dubious however - what I believe is called naive falsificationism - of the sort Popper seemed to indicate that he endorsed.

For as Lakatos (and Kuhn and Feyerabend) points out, if one example of facts counter to the theory was all that was required for the abandonment of the theory, then Special Relativity would have been falsified in its year, and Newtonian physics would have been falsified by the motion of the moon*. Obviously however they were not abandoned then and there by their inventors, and went on to be very productive theories.

*actually the motion of the moon was later found to be not inconsistent with Newtonian physics, although Newton couldn't work it out, and Kaufman's gamma ray experiments were found to contain experiment errors (but that does not alter the fact that it was thought to count as falsifying evidence).

[PixyMisa]

pg, curtc:

I think you're both missing the formal logical fallacy here. It is called denying the antecedent. If (falsify micro) then (falisfy macro) This much is true. If micro were to fail, macro would also fall. But, if not (falisify micro) then not (falisify macro) is not valid. That is fallacious reasoning.

No, it's not.

Because macroevolution is microevolution. They're the same thing. It's like saying that just because we've shown that you can have one litre of water doesn't mean you can have two litres of water. Even worse, it's not two, it's some unspecfied and arbitrarily chosen number.

Inductively, if you have microevolution, it is impossible to avoid macroevolution.

[BillHoyt]

No, it's not.

Because macroevolution is microevolution. They're the same thing. It's like saying that just because we've shown that you can have one litre of water doesn't mean you can have two litres of water. Even worse, it's not two, it's some unspecfied and arbitrarily chosen number.

Inductively, if you have microevolution, it is impossible to avoid macroevolution.
Science is not induction, pixy, it is retroduction. You are falling into the denying the antecedent fallacy. You are also arguing circularly. Please look at my previous post once more.

[BillHoyt]

but there's no difference in the theories. in fact, there is only one theory. if microevolution is true, then it follows almost obiously that macroevolution is possible. (though not necessarilly correct or how it happened; you could then go and falsify parts of natural history, and show that things did not evolve through macroevolution.)

but as long as microevolution can be taken as a valid explanation, then macroevolution is also a valid explanation. there isn't any seperate mechanism differentiating the two. you wouldn't be able to falsify it seperately unless you found some kind of effect preventing the micro changes from forming new species. and since IIRC the formation of new species has already been observed, that is right out.
I'm not arguing against evolution or the neo-Darwinian synthesis. I am stating, as clearly as I can, that macroevolution does not necessarily follow from microevolution. It must be and can be falsified separately. I am arguing against one of the most ancient of formal logical fallacies: denying the antecedent.

And I will state it once more, hoping y'all catch it this time. if (falsify micro) then (falisify macro). Valid. If NOT (falsify micro) then NOT (falsify macro). INVALID.

[ReFLeX]

If the rock falls any direction but straight down, gravity is out the window. First of all, no. Your simple experiment does not disprove gravity, because supposing there were unknown forces at work and we really did see a path other than straight down, that wouldn't mean gravity wasn't working properly. Far from it.


Second of all, microevolution is a creationist mechanism to separate observation from evolutionary theory, I'm surprised a skeptic would even assert there is such thing.

BillHoyt, only you have proposed "If Micro" then "Macro". Science did not observe things in that order.

[BillHoyt]

First of all, no. Your simple experiment does not disprove gravity, because supposing there were unknown forces at work and we really did see a path other than straight down, that wouldn't mean gravity wasn't working properly. Far from it.


Second of all, microevolution is a creationist mechanism to separate observation from evolutionary theory, I'm surprised a skeptic would even assert there is such thing.

BillHoyt, only you have proposed "If Micro" then "Macro". Science did not observe things in that order.
I didn't say what order science observed them in, and, moreover, the order of observation is irrelevant. What is relevant is how we go about attempting to falsify them. Now also note I did not write "if micro," etc. I'm not sure why my point is not getting across.

[drkitten]

Here's a large piece of the puzzle, in my opinion. As the article in the OP pointed out, Newtonian mechanics makes extremely precise predictions. If the rock falls any direction but straight down, gravity is out the window.

Um, not quite. You have a somewhat idealized view of the theory of gravity. There's a reason that all of the problems in a typical high school physics text include weasel phrases like "neglecting air resistance" and such like. Newton's mechanics makes extremely precise predictions -- but they don't apply (with accuracy) to confetti, as you can easily see for yourself.

Part of "falsification" is defining not only what results you expect, but also defining the range of results that you expect. A theory of geology may be able to tell you that there is iron ore in this region, but it won't be able to tell you where every crystal lies. A theory of weather prediction may be able to tell you that it will rain tomorrow, but not where every drop will fall. When the National Weather Service says that tomorrow will get between 1 and 3 cm of rain, failure to specify that there
will be exactly 2.467205 cm is hardly a failure of the theory.

[Paul C. Anagnostopoulos]

I'm not arguing against evolution or the neo-Darwinian synthesis. I am stating, as clearly as I can, that macroevolution does not necessarily follow from microevolution. It must be and can be falsified separately. I am arguing against one of the most ancient of formal logical fallacies: denying the antecedent. This depends entirely on the definition of macroevolution, doesn't it?
If we define macroevolution to be "change at or above the species level," then it depends on what that means. What does it mean? It means a change that some taxonomist decided resulted in a species bifurcation. This is a somewhat arbitrary decision, because species are separated for various reasons, not just inability to interbreed. In fact, many extinct species were defined as a species for reasons other than interbreeding ability.

In order for this to have arisen by something other than accumulated microevolution, you would have to argue that every arbitrary taxonomic species was set down on Earth as is. And you would have to rework this argument every time taxnomic changes are made, which is happening quite frequently these days with molecular discoveries. That makes no sense.

Macroevolution is defined in terms of an arbitrary classification scheme established by humans. It follows from the definition of that classification scheme.

~~ Paul

[BillHoyt]

In order for this to have arisen by something other than accumulated microevolution, you would have to argue that every arbitrary taxonomic species was set down on Earth as is.
Gould seems to argue very well against this, and yet doesn't take on the fundamentalist stance you pose in this false dichotomy.

[Almo]

First of all, no. Your simple experiment does not disprove gravity, because supposing there were unknown forces at work and we really did see a path other than straight down, that wouldn't mean gravity wasn't working properly.


True, it does not disprove gravity. But it very easily points out what to look for as alternatives. There's no impetus to try to reinterpret the results and the theory of gravity. You just look for the other external forces that are acting.

When the adherents to a theory reinterpret any result to match their idea, that's what's unfalsifiable. In the rock-drop experiment, the effect of gravity is clearly delineated from the other effects of Newtonian mechanics. Those effects are well-understood, and make easily testable predictions.

In the case of the frog, you find the frog and the rat have 2% differing DNA. What does this mean? Maybe 3% would be a lot, maybe 1% would be a lot. Evolution is not a theory that provides numbers; we get an idea of what is a large difference in DNA composition empirically. But when the rock goes up, it is dreadfully obvious that some other effect is at work. Even if it's falling by 1.00001% of gravity's prediction, you know that other forces are at work, and don't try to bend the observation to fit gravity. You look for the other forces.

[DrMatt]

Macroevolution is change at or above the species level. Therefore, we'd need to observe a situation in which a new species ought to have arisen, but didn't. We've never yet observed such a situation; unfortunately we don't yet know exactly how to set up such a situation. On the other hand, though, we've already seen many new species arise.

The concept of "species" is human-derived and is not a fundamental fact of nature, but even so, for every even halfway useful biological definition of species, speciation events have been observed. Look up speciation.

[Paul C. Anagnostopoulos]

Gould seems to argue very well against this, and yet doesn't take on the fundamentalist stance you pose in this false dichotomy.
How did I present a false dichotomy? Since macroevolution is essentially defined taxonomically, if there is even one taxonomical distinction that was not put down as is, it is an example of macroevolution.

Perhaps we need a better definition of macroevolution, because I cannot see how it doesn't simply follow from taxonomy.

~~ Paul

[BillHoyt]

The concept of "species" is human-derived and is not a fundamental fact of nature, but even so, for every even halfway useful biological definition of species, speciation events have been observed. Look up speciation.
Dr Matt,

I'm not arguing against evolution in any way, shape, or form. The discussion concerns falsifiability and what it takes. The contention I'm making is that NOT falsifying microevolution does not mean we accept macroevolution.

Let's say we've exhausted our null hypotheses about change within a species. We set up experiements that demonstrate, time and again, that show populations respond to selection pressure. We, therefore find we can't falsify microevolution. That does not mean we can't falsify macroevolution. (Please understand, I am setting aside, for the moment, the fact we've observed it. I'm trying to get across what falsification means and that failure to falsify at one level does not mean we automatically accept what we think follows. This follows from the fallacious nature of denying the antecedent.)

[Melendwyr]

"Inductive reasoning is flawed"?! What the... you might as well say that deductive reasoning is flawed, since starting out with incorrect assumptions does not necessarily lead to correct conclusions.

What a maroon.

[BillHoyt]

How did I present a false dichotomy? Since macroevolution is essentially defined taxonomically, if there is even one taxonomical distinction that was not put down as is, it is an example of macroevolution.

Perhaps we need a better definition of macroevolution, because I cannot see how it doesn't simply follow from taxonomy.

~~ Paul
Paul,

We're talking about falsification here. Set taxonomy aside. That preceded Darwin by a few centuries. We're now in the 20th century, and in the neo-Darwinian synthesis. We have satisfied ourselves that the theoretical framework of the neo-Darwinian synthesis holds up to all our tests within species populations. I'm only saying that failure to falsify at the micro level does not automatically mean we've failed to falsify at the next level. That's where Gould comes in. You stated that the accumulation of micro changes causes macro evolution. Gould disagrees; he thinks there's an even bigger mechanism at work.

Does that clarify it, or am I still yabbering Russian?

[Paul C. Anagnostopoulos]

Gould seems to argue very well against this, and yet doesn't take on the fundamentalist stance you pose in this false dichotomy.
Aha. But what is Gould arguing? Isn't he arguing for big changes "all at once"? How fast is that? Even if one generation had enough mutations to make it a new species (according to any definition of species), that is still microevolution, even if you also want to call it macroevolution.

Are you trying to say that we need to demonstrate single-generation speciation?

~~ Paul

[BillHoyt]

"Inductive reasoning is flawed"?! What the... you might as well say that deductive reasoning is flawed, since starting out with incorrect assumptions does not necessarily lead to correct conclusions.

What a maroon.
Both induction and deduction are flawed. This is why science adopted this synthetic technique called retroduction. Induction's flaw is that n instances is no guarantee that the n+1th instance will follow suit. Deduction's flaw is that you can never use it to generate base premises. It tells you nothing really new.

[Paul C. Anagnostopoulos]

Does that clarify it, or am I still yabbering Russian? I need a definition of macroevolution that makes it something other than a "synonym" for taxonomy. Otherwise it follows without need for evidence.

As I said above, maybe you mean single-generation speciation. Two-generation speciation? Really quick speciation?

What does Gould say? I agree that if he is proposing a fundamentally different mechanism, evidence is required.

~~ Paul

[Melendwyr]

Both induction and deduction are flawed. Clearly this involves some exciting new definition for the word 'flawed' that I wasn't previously aware of.

[PixyMisa]

I'm only saying that failure to falsify at the micro level does not automatically mean we've failed to falsify at the next level.

Only if you define macroevolution as different to microevolution. Every evolutionary theorist I have read has pointed out that this is a false dichotomy. There is no "next level".

That's where Gould comes in. You stated that the accumulation of micro changes causes macro evolution. Gould disagrees; he thinks there's an even bigger mechanism at work.

No he didn't. (Past tense, alas.)

[William Parcher]

Aha. But what is Gould arguing? Isn't he arguing for big changes "all at once"? How fast is that? Even if one generation had enough mutations to make it a new species (according to any definition of species), that is still microevolution, even if you also want to call it macroevolution.

Are you trying to say that we need to demonstrate single-generation speciation?

That example sounds more like saltationism.

[BillHoyt]

We're talking at cross-purposes here. Let me try again. Micro is change within a species. Macro is change that creates a new species. The neo-Darwinian synthesis put together Darwin's original formualation, and our new knowledge of genetics to create a theoretical framework in which micro leads directly to macro. This does not necessarily follow. The mechanisms responsible for microevolution may be quite different from those for macroevolution. This was my point about Gould. He coined "phyletic gradualism" to describe the neo-Darwinian view of macroevolution. He thinks the mechanisms are quite different. His theory is "punctuated equilibrium."

Now, please understand I'm not saying inheritance isn't through the genes. Neither am I arguing that macroevolution didn't occur. Neither am I arguing for PE over PG. I am simply addressing the erroneous statement that not falsifying micro means we've not falsified macro. The mechanisms can be quite different.

Falsification has to occur bit by bit, piece by piece, prediction by prediction. Look back at the decades of research that were needed to not falsify all the consequences of Einstein's papers.

And I reiterate, and I'm not sure why this doesn't cinch my point: it is a formal fallacy to ever state that denial of the antecedant means denial of the consequent. Logic just doesn't work that way. If p -> q, and we conclude ~p, we have not demonstrated ~q. Please, guys. Look this up if you don't believe what you're reading here.

[sphenisc]

Only if you define macroevolution as different to microevolution. Every evolutionary theorist I have read has pointed out that this is a false dichotomy. There is no "next level".



No he didn't. (Past tense, alas.)

Yes, he did

Under 'Gould on Richard Goldschmidt'
http://www.stephenjaygould.org/people/richard_goldschmidt.html

[BillHoyt]

Yes, he did

Under 'Gould on Richard Goldschmidt'
http://www.stephenjaygould.org/people/richard_goldschmidt.html
To add to this understanding, let's look away from Gould and Goldschmidt, etc, for a moment at the standard neo-Darwinian synthesis. Under that framework there is at least one key mechanism that needs to accompany macroevolution: reproductive isolation. This is not a pre-requisite for intraspecies evolution, but is a minimum pre-req for speciation.

(Yes, I know this is a bit oversimplified, but it is illustrative.)

[Ed]

Evidence of some actual macroevolutionary barrier would spoke the guns.


In the name of all ordnance buffs everywhere let me say that it is "spike" the guns.

To spike a gun is to render it unusable. The term originally referred to muzzle-loading cannon (http://en.wikipedia.org/wiki/Cannon). Driving an iron hand-spike (part of the gun's equipment) into the touch-hole on the cannon's breech would enlarge the hole and leave the gun unusable until repaired, or bushed, an hour or more work.

Wiki

Proceed.

[ReFLeX]

In the name of all ordnance buffs everywhere let me say that it is "spike" the guns.



Proceed.
GuarAntee?

ETA: Oooh, that's not going to make sense up here on the second page.

[drkitten]

In the name of all ordnance buffs everywhere let me say that it is "spike" the guns.


My apologies. But have you ever noticed how close the 'o' and 'i' keys are on a standard QWERTY layout?

[ReFLeX]

I didn't say what order science observed them in, and, moreover, the order of observation is irrelevant. What is relevant is how we go about attempting to falsify them. Now also note I did not write "if micro," etc. I'm not sure why my point is not getting across.
No, the order is logically important. Because denying the antecedent looks exactly the same as modus tollens, the way you wrote it.

[BillHoyt]

No, the order is logically important. Because denying the antecedent looks exactly the same as modus tollens, the way you wrote it.
I'm not discussing observation. I'm discussing falsification of theoretical frameworks.

This is what I wrote:
And I will state it once more, hoping y'all catch it this time. if (falsify micro) then (falisify macro). Valid. If NOT (falsify micro) then NOT (falsify macro). INVALID.
The first if...then is modus tollens. The second if...then is denying the antecedent.

[William Parcher]

I'm not discussing observation. I'm discussing falsification of theoretical frameworks.

Right. But I agree with Paul (and others) that if one is going to try to falsify and try to understand what would be necessary for falsification... then one must first have functional definitions of what it is that is the target of potential falsification.

Falsification of anything is a definable endeavor in itself. But when one speaks about falsifying anything in particular, then definitions about the target(s) of falsification can have some bearing on how this could be accomplished.

[ReFLeX]

I'm not discussing observation. I'm discussing falsification of theoretical frameworks.

This is what I wrote:

And I will state it once more, hoping y'all catch it this time. if (falsify micro) then (falisify macro). Valid. If NOT (falsify micro) then NOT (falsify macro). INVALID.


The first if...then is modus tollens. The second if...then is denying the antecedent.

The first one can't be modus tollens because you haven't negated anything. It looks more like modus ponens. If it was M.T., then it would look more like the second one, thusly:
1. falsify micro > falsify macro
2. ~ falsify macro
Conclusion: ~ falsify micro

If premise 2 was ~falsify micro and we concluded ~ falsify macro, then that would be a fallacy, but that's the difference of the order. The way you've written it is ambiguous.

...whatEVer...:cool:

[BillHoyt]

The first one can't be modus tollens because you haven't negated anything. It looks more like modus ponens. If it was M.T., then it would look more like the second one, thusly:
1. falsify micro > falsify macro
2. ~ falsify macro
Conclusion: ~ falsify micro

If premise 2 was ~falsify micro and we concluded ~ falsify macro, then that would be a fallacy, but that's the difference of the order. The way you've written it is ambiguous.

...whatEVer...:cool:
My error; I meant to write modus ponens for the first one. Thanks for the correction.

On the second one, I have what you describe. The "antecedant" refers to the conditional, or first premise. What I wrote is corrrect. ("If NOT (falsify micro) then NOT (falsify macro). INVALID.") For p, I am using "falsify micro), for q, (falsify macro). Translating: if ~p then ~q. Here's a link for you. (http://www.fallacyfiles.org/denyante.html)

[drkitten]

Right. But I agree with Paul (and others) that if one is going to try to falsify and try to understand what would be necessary for falsification... then one must first have functional definitions of what it is that is the target of potential falsification.

Falsification of anything is a definable endeavor in itself. But when one speaks about falsifying anything in particular, then definitions about the target(s) of falsification can have some bearing on how this could be accomplished.

I think part of the problem is the our understanding of the processes of micro- and macro-evolution. Paul points out, correctly, that according to our best current theoretical understanding, macro-evolution is simply the cumulative process of lots and lots of micro-evolution.

However, like any other scientific understanding, it might be wrong. Microevolution might be a separate and theoretically separable process from macroevolution, although that would contradict a hell of a lot of what we (believe we) know about biology. On the other hand, Einstein contradicted a hell of a lot of what we believed we knew about the physics of space and time.

There might, for example, be some theoretical barrier that a species cannot cross (which the creationists would instantly leap upon as a definition of baramin or kind, and I'd be happy if they did, because otherwise the word keeps shifting like a badly made gelatin dessert). In this case, we could legimately falsify macroevolution by demonstrating the barrier, while still leaving microevolution intact. Failure to falsify microevolution does not imply a failure to falsify macroevolution. (I should probably point out -- no, I have no evidence to support the idea of this barrier, and no, I do not actually believe that it does exist. I've also been wrong before.)

[Soapy Sam]

We're actually in the 21st century , Bill, but I think you really knew that.

If there is only one kind of evolution (ie the micro / macro distinction is spurious) then clearly the logical error does not arise. Disproving evolution if possible disproves the lot.
If we postulate six types of evolution, Bill's logical argument may be (though does not HAVE to be) right.

So do we actually have six types of evolution?
Not that I heard of.

Do we have two types?
Not in my opinion.

We may imagine a variation in embryonic developmental timing which produces wide variation in adult forms with no gene change at all. Is this macroevolution? Presumably.
Is it microevolution?
I would say "yes", because I see timing as part of the process. I can see that viewed simplistically- ie one gene=one characteristic, then the answer would be "no".

I think we have to choose the model which best reflects what actually seems to happen. That's why I feel that evolution is not simply change in allele frequency. It's change in frequency and function. Part of that function often may be feedback from the environment.

[drkitten]

So do we actually have six types of evolution?
Not that I heard of.

Do we have two types?
Not in my opinion.


But "your opinion," well-founded though it may be, does not rise to the level of logical proof.

Well-educated people have been wrong before. If we see something that seems to violate the simple model of a single, unified type of evolution -- well, then we were wrong.

[sphenisc]

We're actually in the 21st century , Bill, but I think you really knew that.

If there is only one kind of evolution (ie the micro / macro distinction is spurious) then clearly the logical error does not arise. Disproving evolution if possible disproves the lot.
If we postulate six types of evolution, Bill's logical argument may be (though does not HAVE to be) right.

So do we actually have six types of evolution?
Not that I heard of.


Okay, let me have a go.

Evolution by natural selection
Evolution by sexual selection
Evolution by artifical selection
Evolution by genetic drift (bottleneck effect)
Evolution by genetic drift (founder effect)
Evolution by migration

http://www.mansfield.ohio-state.edu/~sabedon/campbl23.htm

Is that what you meant?

[BillHoyt]

We're actually in the 21st century , Bill, but I think you really knew that.
I meant what I said, although it wasn't clear. I meant it as "fast forward to the twentieth century." Darwin / Mendel were 19th century. The neo-Darwinian synthesis began forming in the early 20th.

If there is only one kind of evolution (ie the micro / macro distinction is spurious) then clearly the logical error does not arise. Disproving evolution if possible disproves the lot.
If we postulate six types of evolution, Bill's logical argument may be (though does not HAVE to be) right.

So do we actually have six types of evolution?
Not that I heard of.

Do we have two types?
Not in my opinion.

We may imagine a variation in embryonic developmental timing which produces wide variation in adult forms with no gene change at all. Is this macroevolution? Presumably.
Is it microevolution?
I would say "yes", because I see timing as part of the process. I can see that viewed simplistically- ie one gene=one characteristic, then the answer would be "no".

I think we have to choose the model which best reflects what actually seems to happen. That's why I feel that evolution is not simply change in allele frequency. It's change in frequency and function. Part of that function often may be feedback from the environment.
In actual fact, the micro / macro distinction is not spurious. As I wrote before, we know that reproductive isolation is a key mechanistic difference between intra-special change and speciation.

As for the different types of evolution, actually the answer is yes, there are three different types. (Now I'm writing from a theoretical population genetics standpoint.)

Have I muddied up the waters sufficiently? :D

[Soapy Sam]

I meant what I said, although it wasn't clear. -BillHoyt

I know, but I couldn't resist it..

we know that reproductive isolation is a key mechanistic difference between intra-special change and speciation.

Again, I feel the distinction contributes nothing useful. Intra-specific variation is speciation, just early in the day. The difference is only the point in time we happen to sample. (You'll have gathered I don't rate "species" highly as a useful measure of evolution, or a useful measure of anything much.)
Imaginary scenario- In 1492 a plague wipes out all humans except Australian Aborigines. The Australian population shares a genepool, but has some internal variation, likely to be most marked between tribes with different geographical distributions.
One northern tribe recolonises New Guinea and their descendants recolonise the rest of the world, over 30,000 years or so, adapting to a couple of ice ages along the way and wondering where all the coal went. Meanwhile, back in Tasmania, the population remains isolated...
It seems very likely that after 30,000 years, we might have two distinct human species ,one in Tasmania, one everywhere else, whether through gene shift or attitude shift. Two populations which don't interbreed are exactly the same as two which cannot interbreed at the macro level. (From the "POV" of evolution). At the gene level that's not the case, they actually are different. (Maybe that's an example of Gould's different process? I don't know.)
What I'm getting at is that the process is continuous through time and space. Every gene complex is geographically isolated-inside a body- for a lifetime. Every community is isolated, because movement is limited, whether across a puddle, a field, a forest ,a continent -for a time- but not forever. Things move.
This is a geologist's biased view of course, where continents bounce around, rivers act like lawn sprayers and mountain ranges pogo. A biologist might see it differently.




Evolution by natural selection
Evolution by sexual selection
Evolution by artifical selection
Evolution by genetic drift (bottleneck effect)
Evolution by genetic drift (founder effect)
Evolution by migration -Sphenisc

We're back to definitions again. Those are not types of evolution. They are types of selective mechanism, any of which can result in evolution and all of which operate at the environment / organism interface to produce changes in the genome. But I can accept that any of them might operate by producing timing changes in the way the genome creates the organism, rather than changing the allele frequency in the organism.

[PixyMisa]

Yes, he did

Under 'Gould on Richard Goldschmidt'
http://www.stephenjaygould.org/people/richard_goldschmidt.html
No he didn't.

I have a copy of The Structure of Evolutionary Theory, though I haven't read it cover to cover - it's quite lengthy. If you can point to a page where Gould actually says what you say he said then I'll have a read and get back to you. But I've read almost everything Gould wrote for the popular press, and nowhere did I see any statement of the sort you're suggesting.

On that web page, all we see is Gould discussing Goldschmidt's ideas (something I recall from Gould's other books), not Gould agreeing with those ideas.

[sphenisc]

No he didn't.

I have a copy of The Structure of Evolutionary Theory, though I haven't read it cover to cover - it's quite lengthy. If you can point to a page where Gould actually says what you say he said then I'll have a read and get back to you. But I've read almost everything Gould wrote for the popular press, and nowhere did I see any statement of the sort you're suggesting.

On that web page, all we see is Gould discussing Goldschmidt's ideas (something I recall from Gould's other books), not Gould agreeing with those ideas.

Gould on Richard Goldschmidt

""As a Darwinian, I wish to defend Goldschmidt's postulate that macroevolution is not simply microevolution extrapolated "

My bold

[drfrank]

OK, I'm surprised that no one's posted this before, which may have helped make some of the previous debate unnecessary:

Speciation and observed speciation events at TalkOrigins (http://www.talkorigins.org/faqs/faq-speciation.html)

Regardless of such observed speciation events, the evidence of evolution between species is as close to irrefutable as makes no odds. Take, for example, the speciation between apes and humans (well, in our last common ancestor that then became apes and humans), especially because it's the one Creationists hate the most.

a) Same vitamin-C pseudogene: ever wonder why you don't need to feed your dog fruit?
b) Incredibly high similarity in DNA sequence
c) 1 less pair of chromosomes than apes (23 compared to 24) - evidence of Robertsonian translocation (chromosome fusion event) in chromosome 2 (2 centromeres), and DNA material exactly where expected from this fusion.
d) observed Robertsonian translocation in humans (about 1 in 1000), and the fact that these individuals can still produce fertile offspring. Interbreeding amongst individuals with the same fused chromosome could then result in fertile individuals with 2 fused chromosomes, which is exactly the same difference between us and apes.
e) Hominid fossils charting many branches of evolutionary tree between apes and ourselves

Oh, and if that's not enough for you, check the evolution of the mammalian inner ear from the reptilian jawbones - there's some amazingly good intermediate fossils, plus tracking gene expression shows that our ear bones start from the same place at as reptilian jawbones when embryonic.

I'm undoubtedly leaving out an enormous amount of evidence too (as I'm not actually a biologist), but if anyone can explain why that isn't rock solid evidence of `macroevolution' I'd be pretty darn interested to know why.

Overall, denying the existence of `macroevolution' is just as ridiculous as believing you can summon UFOs on demand.

[sphenisc]

OK, I'm surprised that no one's posted this before, which may have helped make some of the previous debate unnecessary:

Speciation and observed speciation events at TalkOrigins (http://www.talkorigins.org/faqs/faq-speciation.html)

Regardless of such observed speciation events, the evidence of evolution between species is as close to irrefutable as makes no odds. Take, for example, the speciation between apes and humans (well, in our last common ancestor that then became apes and humans), especially because it's the one Creationists hate the most.

a) Same vitamin-C pseudogene: ever wonder why you don't need to feed your dog fruit?
b) Incredibly high similarity in DNA sequence
c) 1 less pair of chromosomes than apes (23 compared to 24) - evidence of Robertsonian translocation (chromosome fusion event) in chromosome 2 (2 centromeres), and DNA material exactly where expected from this fusion.
d) observed Robertsonian translocation in humans (about 1 in 1000), and the fact that these individuals can still produce fertile offspring. Interbreeding amongst individuals with the same fused chromosome could then result in fertile individuals with 2 fused chromosomes, which is exactly the same difference between us and apes.
e) Hominid fossils charting many branches of evolutionary tree between apes and ourselves

Oh, and if that's not enough for you, check the evolution of the mammalian inner ear from the reptilian jawbones - there's some amazingly good intermediate fossils, plus tracking gene expression shows that our ear bones start from the same place at as reptilian jawbones when embryonic.

I'm undoubtedly leaving out an enormous amount of evidence too (as I'm not actually a biologist), but if anyone can explain why that isn't rock solid evidence of `macroevolution' I'd be pretty darn interested to know why.

Overall, denying the existence of `macroevolution' is just as ridiculous as believing you can summon UFOs on demand.

I think you've missed the point, no-one is denying macroevolution. We're discussing how could macroevolution be falsified and in particular would a falsification of microevolution be sufficient to do so.

[drfrank]

I think you've missed the point, no-one is denying macroevolution. We're discussing how could macroevolution be falsified and in particular would a falsification of microevolution be sufficient to do so.

In the original post Almo quoted one of his friends as saying:

But, this doesn't ease my concerns about macro-evolution: I still see it as almost a psuedo-science (though with considerably more credibility). I think part of the problem is the rah-rah folks that support it (usually as an alternative to religion) just like other folks rah-rah for religion.

and he didn't seem to indicate that he disagreed with this part of the sentence, so I got the impression (perhaps falsely), that Almo was not convinced by the evidence. Plus, taking the tack that macroevolution is unfalsifiable and therefore unscientific is a popular Creationist ploy.

If I was mistaken about this, then I apologise.

I think DrKitten is doing a good job of explaining this, but I'll put my own spin on it anyway :)

Let's face it, any argument about `macro-evolution' is an argument about common ancestry, and generally involves that nebulous words `kinds'. It comes squarely from the classic "my grandfather wasn't a monkey" line.

In a strict sense, falsifying common ancestry would not falsify the possibility of macroevolution, but if there were no evidence of common ancestry there would be no evidence of macroevolution.

Common ancestry could be falsified by an enormous number of observations but, strangely enough, hasn't been.

At this juncture, I think talking about falsifying common ancestry or, indeed, `macroevolution' is like discussing what you'd do with your perpetual motion machine: pure mental masturbation.

[sphenisc]

Let's face it, any argument about `macro-evolution' is an argument about common ancestry, ..snip...

In a strict sense, falsifying common ancestry would not falsify the possibility of macroevolution.

What's the distinction you're drawing here? :confused:

[ReFLeX]

Here's a link for you. (http://www.fallacyfiles.org/denyante.html)Err.

[drfrank]

What's the distinction you're drawing here? :confused:
Well, to say that macroevolution has not occurred between species in the past does not necessarily mean that it's impossible in the future, but it would obviously destroy common descent.

However, when people talk about macroevolution they generally mean that which has already happened i.e. common descent.

My original statement that you quoted was identifying why people (Creationists and IDiots) argue about macroevolution, which is because they don't like common descent as it disagrees with Biblical `kinds'. That's the only real reason this discussion ever gets raised.

Edit: I agree - I probably didn't explain that well the first time, sphenisc. My apologies :)

[drkitten]

In the original post Almo quoted one of his friends as saying: [some stuff]

and he didn't seem to indicate that he disagreed with this part of the sentence, so I got the impression (perhaps falsely), that Almo was not convinced by the evidence. Plus, taking the tack that macroevolution is unfalsifiable and therefore unscientific is a popular Creationist ploy.


Well, there are two independent issues here. One is the question of whether macroevolution is falsifiable in theory, and the other question is the amount and kind of evidence available today to support the idea that macroevolution has occurred.

One of the common complaints about the theory of evolution -- and it's an invalid complaint, usually only raised by creationists and idiots, but apparently there are a lot of idiots out there -- is that it's an unfalsifiable "just so story" to which any kind of finding from the fossil record can be bent. And in the abstract, yes, those kinds of "just so stories" do in fact exist and have been proposed as genuine scientific "theories," which is the reason Popper came up with the falsifiability definition in the first place.

Popper himself suggested that both Marxist economics and Freudian psychology were unfalsifiable. The problem wasn't that there was no evidence to support them -- their proponents could produce realms of evidence. The problem was that there was no evidence -- not even a possible state of the world -- that could not be interpreted in support of these "theories."

Naively -- and I'm willfully erecting a straw man here -- interpreting the fossil record can appear to be something like that. If I find a new fossil hominid in the wrong layer of rock, then I simply update the story I tell about human evolution to explain it. If I find an Australopithecus buried in the Yukon, then that simply tells me that "obviously" Australopithecus migrated more widely than we had previously thought -- it doesn't tell me that our theories about the relationship between Australopithecus and Homo were wrong.

Now, let me burn the straw man. We can reinterpret our theories within rather broad limits, but there are some finds that would force us to re-examine our entire intellectual structure in major depth. We "know," for example, that humans and chimpanzees separated via speciation about six million years ago. We also "know" that mammals separated from reptiles about 170 million years ago. Finding a hominid fossil -- not just recognizably mammal, but recognizably human -- dated 200 million years ago would utterly and completely falsify this "knowledge"; we couldn't just say that "hominids originated earlier than we had previously thought," because what we think about evolution at the moment is that 200 mya, there not only weren't any hominids, but there weren't any proto-hominids, or even any protomammals. At this point according to our current theories, humans were just separating from turtles.


Common ancestry could be falsified by an enormous number of observations but, strangely enough, hasn't been.

At this juncture, I think talking about falsifying common ancestry or, indeed, `macroevolution' is like discussing what you'd do with your perpetual motion machine: pure mental masturbation.

Well, there's always the possibility of the Carboniferous hominid.... It might turn up (and monkeys []i]might[/i] fly out of my butt.....)

Part of the problem is that Almo is looking for a nice, clean cutoff. I think we can agree that a 200 mya hominid would cause chaos. How about 100 mya? How about 8 mya? Interpretation of evidence is always somewhat subjective -- I'd be willing to believe that we were off by two million years in our "story" of human/chimp speciation, but not by 20 million years. Other people may have different credibility threshholds. But what's "really" going on here is still falsifiation -- it's just a question of how radically we have to adjust our our old, falsified, theory to make way for the new one. A "minor" discrepancy in the evidence may be patchable with a minor tweak. A major discrepancy will probably produce a major overhaul.

At this point objection number two arises. It's hardly fair to talk about finding something that may or may not exists as a "falsification" -- we can't actually run experiments and expect to find fossils; they're serendipitous lucky finds, but not something on which we can rely. What are the experiments we can run to falsify them?

This is where the genetic experiments come in. Again, we've done enough gene studies that we know "about" how fast genes change, "about" how much difference will make two groups of organisms into reproductively isolated species, and so forth. And again, we can tweak these numbers if necessary -- but something that was way out of line with our expectations wouldn't be fixable by mere "tweaking."

The flip side of this discussion is that we've already run a hell of a lot of experiments, and that any proposed change to the theory of (macro)evolution would also have to explain the results of the historical experiments. This is, in broad form, known to science as the Correspondence Principle (http://www.skepticwiki.org/wiki/index.php/Correspondence_Principle) -- the idea that a new theory must also encompass explanations of the phenomena explained by the old one. In the words of the SkepticWiki: "[i]When an established theory is contradicted by a rival new theory, we are forced to ask why, if the rival theory is correct, was it ever possible to prove the established theory? Equivalently, why should a new contradictory theory be investigated, when the established theory has already been proven? The Correspondence Principle suggests that if a rival theory is correct, then the established theory must then have been a very good approximation. The Correspondence Principle can then be used as a kind of

[Almo]

This has been a very interesting discussion, and my pal has been reading it. He's a mathematician, and an incurable skeptic. Just a brief excerpt from his impressions:

You know, now I feel much better about the whole thing: the problem is almost certainly not with the theory, it's with my understanding of the theory. That list of predictions that are clearly falsifiable really helps---it's still the case that this may not be the mechanism, but I really feel much better about the scientific aspects of it (that is, it really *is* science and not just a clever explanation that fits the observed facts).

Big thanks to everyone for helping out on this one. I feel it's helped my understanding of the situation as well.

And drkitten... "as a kind of what?" :)

Edited to add teeny bits, and correct a typo

[Almo]

In the original post Almo quoted one of his friends as saying: Some Stuff...

If I was mistaken about this, then I apologise.


Not a problem. I really posted it to see how you guys would respond to his concerns. I had just read the Popper article, so I hadn't had enough time to digest the ideas to say much.

[brodski]

And drkitten... "as a kind of what?" :)


I think it was just meant a a teaser for the SkepticWiki article. :p

[drkitten]

And drkitten... "as a kind of what?" :)


Oh, damn damn damn damn damn. I corrected a typo, and it ate a third of my article.

Well, you can read the SkepticWiki article for yourself.

The basic idea is that whatever hypothetical theory replaces macroevolution will have to do at least as good a job of explaining what we've already found. Why, for example, do we have such a good match between genetic similarity and taxonomic? Why do we find that the amount of genetic similarity seems to correlate with the apparent age of fossils? Why do the chemical properties of the age of rocks correlate so well with the apparent age of fossils and the strata in which they are found? And why are there all these pesky intermediate forms -- feathered lizards, whales with legs, trunkless elephants, and ape-men with intermediate-sized brains?

Whatever theory eventually replaces our current theories of evolution is going to have to come to grips with the evidence we already have. It will not get to start with a clean slate and wish Ambulocetus out of existence....

[ReFLeX]

Irrelevant post. Oops.

[drfrank]

Not a problem. I really posted it to see how you guys would respond to his concerns. I had just read the Popper article, so I hadn't had enough time to digest the ideas to say much.

Well, glad we could all help :)

Sorry for being a little testy, but in the past when I've heard people say that they're `concerned' about macroevolution it usually translates as "Hello, I'm a hardcore Biblical literalist who will ignore any evidence you bring" ;)

As always, it's good to be wrong :D

[Almo]

Perfectly understandable. It's like those people who say "I'm not a racist, but..." They really get on my nerves.