In a recent UD post, Gil has been more specific than he often is, so I thought I would respond here:
The resolution of the debate about the creative powers of natural selection is dead simple and utterly trivial to figure out.
- Natural selection throws stuff out. Throwing stuff out has no creative power.
- Existing biological information, mixed and matched, can be filtered by natural selection, as in sexual reproduction, but nothing inherently new is created.
- Random errors can produce survivability quotients, but only in circumstances in which overall functional degradation supports survival in a pathological environment (e.g., bacterial antibiotic resistance), and only given massive probabilistic resources and a few trivial mutational events capable of producing the survival advantage.
- Random errors are inherently entropic, and the more complex a functionally-integrated system becomes, the more destructive random errors become. Anyone with any experience in even the most elementary engineering enterprise knows this.
To his first, I cite this:
To the second, I say: why not? Every mutation is something new, and that can be “mixed and matched” as well as “existing information”.
To the third, I say: this is simply not the case.
To the fourth, I say: this assertion assumes that the biological landscape is as rugged as the engineering landscape. It clearly is not. Engineered artefacts are usually highly vulnerable to slightly alterations – A stuck screw can render an entire motorcycle worthless, as Robert Pirsig noted. This is not the case with biological organisms, which countless slight variants are perfectly viable, as is evidenced by the fact that although all children (including even monozygotic twins) are unique, most are viable.
Yet, we are expected by Darwinists to believe that throwing a sufficient number of monkey wrenches into the complex machinery of living systems, over a long enough period of time, can turn a microbe into Mozart.
Is not unreasonable at all 🙂
So try to understand our position first before you criticize it.
Evolution is not a completely random process.
Understand our position.
Also, you can try to be less angry.
No, it’s called refuting a point, specifically, point 1.
Then we’re in agreement that nature has no goal.
I actually understand your position fairly well, and I believe Elizabeth can vouch for the fact that in previous discussions I have fairly restated and discussed the arguments she has made. By seizing on the plain text you avoid the import of Gil’s assertion. I have yet to see any indication that you understand the point he is trying to make.
My own response was simply an attempt to bring out what MAY be that reasoning, and may be no better an understanding than yours. However, to merely reject his statement on the basis of two simple contradictions shows a lack of charity.
Discussions of the nature of randomness and how it applies to natural selection are a bit premature. I do know, however, that Michaelangelo’s David is not the result of “random” sculture. Not even Henry Moores’ was produced in that manner.
What charitable reading of “throwing stuff away” can comport with biological variation and selection?
What part of “throwing stuff away” is consistent with gene duplication , etc?
Try as I may, the phrase seems more dishonest than merely wrong.
How’s this charitable reading?
Gil is arguing that unguided variation can’t create anything new that is also useful. It cannot feed selection with options that are simultaneously useful and more complex.
Since variation isn’t productive, selection is “Muntzing.”
Is that Gil’s argument.
As I read it, Gil is a software engineer, and he’s basically saying:
1) Evolution works just like software engineering.
2) Software generally cannot tolerate so much as a single accidental bit change.
3) Biology changes lots of bits all the time without crashing.
4) Therefore, the bit changes MUST be designed.
And so to Gil, the entire idea of random mutation is “transparent lunacy” when a single tiny but random change is guaranteed to render the organism nonviable. That’s how software engineering works, after all.
Gil tells an interesting story. He explains that when he was younger he was a devout believer in the Darwinian orthodoxy (yes, those are his terms). Then he read Denton’s “Theory in Crisis” and realized that his god was flawed, perhaps very seriously flawed. His faith was destroyed, until he turned to the Christian God, which gave him the perspective he needed to understand how false his previous faith truly was.
And ever sense, like a woman scorned, he has had nothing but the most hyperbolic rejection of biological concepts, which so flagrantly violate the principles of software engineering that only an idiot could believe otherwise.
And really, anyone who needs Absolute Certainty that much, is not going to be happy with how evolution works.
And perhaps Dodgen should check on what Michael Denton says these days. Denton is still a dissenter from the Modern Synthesis, in a teleological direction. But one of his central arguments was that molecular sequences fails to confirm common ancestry because it does not show frogs as intermediate between fish and mammals.
That was a deeply mistaken argument, and Denton has since backed away from it. You’d never know this if you read Evolution News and Views or Uncommon Descent, as they often cite Denton’s book admiringly, and without noting that one of its major arguments has been abandoned by Denton.
Could that be the very argument that originally persuaded Dodgen to abandon his original views?
I find Petrushka’s and Flint’s responses much more fair and interesting. They have made an effort to understand and engage the original argument – thank-you!
I’m going to throw in another possibility, that even if I can be shown to be wrong, hope you might find interesting.
That is the proposition that the process of sculpting David from a block of stone shows no creative power at all, when considered in a mathematical or information-specific sense.
For instance, the “final” product can be described by the information contained in the arrangement and ordering of the atoms involved in it. Since neither the number, or arrangement of these atoms changes from start to finish, how has the “information” contained increased? I’m not sure if it was Michaelangelo who said this, but a sculptor sees the final product before even beginning; the act of sculpting is merely the removal of that part of the sculpture that prevents those who are not the sculptors from recognizing the form “trapped within”. Since the part outside contains no information bearing on the final form (except in being its mirror image, and hence not independent), removing it will not increase the net amount of information. You might argue, instead, that any increase in information occurred in the mind of the creator before the sculpting even began.
Considered from the block of stone’s perspective, instead of containing an infinite number of possibilities, it has become limited to a single realization; is it not now a much simpler thing?
Moreover, in arguing that natural selection is a process where the finished product already exists, and merely removing some parts gives us what we see, is not Gil actually arguing the case of Michaelangelo’s David, instead of his point being refuted thereby? After all, he is saying, removing parts to get to a certain form is much more straightforward a thing than starting with a lump of clay and building it up piece by piece, where each addition must itself make the whole more “fit” than the previous.
And indeed, does not our actual experience (such as the development of bacterialogical resistance) show that it is the removal or degradation of existing parts – burning the drawbridge, so to speak? Sort of like removing parts on David.
SCeesman: “Since neither the number, or arrangement of these atoms changes from start to finish, how has the “information” contained increased?”
Huh? Of course the arrangement of atoms has changed for a considerable number of the atoms involved: all those atoms that were part of the block of stone at the start, and are not part of it at the finish.
Am I missing something here?
Wait a minute. With every chip removed, the sculpture became something different, right from the start. You could just as accurately say that the current state of the sculpture is just a longer delay between chips removed. And the removed chips must be somewhere, and taken altogether wherever they are, they comprise a different sort of sculpture, and every time any of them moves, this same sculpture changes again.
Your frame of reference is much too narrow.
Sculptors of clay (or sand, etc.) are constantly both building up and cutting back. One could even say that the block of marble got built up, and then cut back. Both adding and removing are going on constantly in biology, and both are essential. It’s not really a good representation to take either one in isolation and point out that, in isolation, it’s insufficient.
Yes, you are missing something. That which is removed is completely defined by (e.g. a mirror-image) of what remains, in other words, it contributes nothing independently describable to the whole beyond a “-1” sign. Because its form is fully dependent on what is interior, it contains no independent information of its own; when you remove that portion there is no net increase or decrease in information, merely a separation. You have, for want of a better term added a translational value to the position of all the exterior points; you have neither increased or decreased the information contained in those exterior points. Since the information of the parts has not changed, then how has the sum?
Different does not mean that anything has been created. And for that part which describes the final statue itself, what has changed?
When you view David, are you consumed by what was taken away? No. That part is irrelevant; it could have extended a thousand miles in every direction or have been an atom thick; you can determine nothing with any degree of certaintly about what has been removed by looking at what remains.
What David really IS is a mathematical surface which we interpret as being equivalent to that of a human being. The material itself if actually irrelevant, whether it has been added or subtracted. How that surface is instantiated is also irrelevant, whether built up from nothing, or sculpted down to from the outside.
Flint, with this I mostly agree. The big difference is that in creating a sculpture the artist has the final form in mind throughout, so that the path to the end is largely irrelevant. In this respect a sculture is a poor model of natural selection, which has no forethought and must at every stage adjust to the local filtness gradient. It is also why the creation of a sculpture is a poor disproof of arguments AGAINST natural selection. Which is why I disagree with it as an argument as to why Gil’s first statement is incorrect.
“That which is removed is completely defined by (e.g. a mirror-image) of what remains”
No. Not if you want to describe the two states (start = block of stone, finish = David of stone) by the arrangement of atoms involved in them. And you seemed to say that that’s what you want to do.
madbat89: Not if you want to describe the two states (start = block of stone, finish = David of stone) by the arrangement of atoms involved in them. And you seemed to say that that’s what you want to do
Let’s grant this desire. At the start we have the information that describes the arrangement of the interior points “A”, and the information that describes the arrangement of the exterior points “B”, which you might argue gives the sum A + B.
Afterwards, the arrangement of the points in “B” is irrelevant, in fact all we now know (because we can only see the remaining statue) is A. So it seems we have lost an amount of information “B”. The loss is not because we don’t know what B is, it is because the current arrangement B can truly be anything – it is no longer specified except to say it is not there – it has gone to zero.
So have we in fact lost an amount B? No. For the same reason I said at the beginning B is just -A, it has no idependent existance, to remove it does not change the total amount of information of the final product, which is in fact the information of the surface separating A and B, and entirely independent of stone itself.
SCheesman: At the start we have the information that describes the arrangement of the interior points “A”, and the information that describes the arrangement of the exterior points “B”, which you might argue gives the sum A + B.
If by “information” you mean: the arrangement of atoms, then I agree.
Afterwards, the arrangement of the points in “B” is irrelevant, in fact all we now know (because we can only see the remaining statue) is A.
I assume you mean: for the description of the second state (David of stone), the arrangement of atoms in “B” is irrelevant… which is of course false, because there is at least one important fact about the arrangement of atoms in “B” that is very relevant: their arrangement differs from their arrangement in the first state (block of stone).
The loss is not because we don’t know what B is, it is because the current arrangement B can truly be anything – it is no longer specified except to say it is not there.
Again – the fact that it is no longer there is the relevant “specification” – it cannot truly be anything.
So have we in fact lost an amount B?
I don’t think “lost” is a meaningful word to use here. We have made a relevant change in the arrangement of atoms in B. That’s how we got from the first state to the second state.
madbat089: “different” does not mean anything has been added or subtracted in terms of information.
Words can be rearranged without any change in the amount of information conveyed.
Without any change in the amount of information conveyed, words can be rearranged.
SCheesman, You’re bending over backwards to defend what is just one more sloppy argument from Gil. Take another look at Gil’s words:
He is saying that natural selection has no creative power because “throwing stuff out” has no creative power.
As you acknowledge, “throwing stuff out” has tremendous creative power if it is done with foresight. You argue that what hobbles natural selection is not the fact that it throws stuff out, but rather that it does so without foresight and in a way that is constrained by the local fitness gradient.
Gil’s claim is, as he puts it, “simple and utterly trivial to figure out.” It is also clearly wrong. Your claim is neither simple nor trivial, and it is impossible to demonstrate without having a detailed model of the fitness landscape and how it evolves over time.
Gil’s claim is that “throwing stuff out” is what makes natural selection non-creative. Your claim is that what matters is natural selection’s lack of foresight.
Gil’s statement is disproven by Michelangelo. Yours is not.
Your claim depends on the nature of the fitness landscape. Gil’s does not.
Gil claims that the idea of natural selection is “transparent lunacy”. You seem to acknowledge that it is not, and that a case can be made for it by sane and dispassionate thinkers.
You are willing to engage in actual argument. Gil is not.
SCheesman: “different” does not mean anything has been added or subtracted in terms of information.
This does not make any sense if by “information” you mean the arrangement of atoms.
Words can be rearranged without any change in the amount of information conveyed.
Change without conveyed information can be of any amount in the rearranged words.
No, the difference is that sculpting a marble statue only involves removing stuff, whereas evolution involves far more.
The sculpture was not in any way intended to be a model of natural selection, but rather a trivial illustration that Gil’s assertion was incorrect.
The statue illustrates that Gil’s first argument was clearly wrong. Since his argument rests on an incorrect premise, it’s a wrong argument.
Your argument seems to be that creativity requires intent. And that argument needs some fairly extensive decoding, because your words have denotations and connotations that may not be entirely shared.
Is (for example) mountain building a creative process? Well, it creates mountains, it creates skylines, it creates photogenic scenery, it involves both building up and tearing down acting simultaneously so it’s not a simple process. So in that case would it be better to say that natural processes can create without intent, or would it be better to say that since this process clearly creates, it implies intent and we must gin up an imaginary entity to possess that intent?
When Natural Selection “throws stuff out”, a separate process, reproduction, copies the survivors. Those survivors, being drawn from that fraction that was not thrown out, are more likely to possess ‘better’ qualities and less likely to possess ‘worse’.
What seems to trouble many opponents of ‘natural’ evolution is the failure to see the symmetry between ‘throwing stuff out’ and ‘keeping the good stuff’. You cannot dilute a population in Quality A without enriching it in Quality Not-A. Any differentials in fitness between any variant and its rivals, novel or ancient, will tend to cause enrichment in the fitter variant – generally as a consequence of “throwing stuff away” (everything gets thrown away in the end, just some more readily than others).
An “NS David” analogue would involve reproductive variants of an original marble block being eliminated for looking a bit less like David than others. They replicate, with variation – almost chips off the old block – and the process reiterates (dangerously close to Weasel, of course, in having a goal). What is thrown away is actually the whole genomes that generate the less-David-like form compared to what the current population holds, rather than the bits of any particular not-quite-David – but the result is equivalent.
In defense of Gil, he was speaking to natural selection, so it is not a stretch to propose that his comment about throwing things out might also refer to the process of natural selection as well, not necessarily the universe of possibilities of throwing things away. I think there was some hyperbole in his assertion, which naturally left him open to attack. I, on the other hand would have thrown in some qualifications that would have left me open to the charge of imprecision. Each to his own.
I was not trying to make any argument so strong as that creativitiy requires intent, only that in the case of the sculpture it did; I try very hard in these conversations to argue the specific instead of the global. Making blanket statements is an invitation not only to contradiction, but to diversion from the point being discussed.
Is mountain building a creative process? I would say only in the sense that something is being created. The creation of David is qualitatively different in that it is a complex specification; it is the reproduction of the physical outline of a recognizable, historical figure. The configuation space for surfaces is sufficiently large that we can state with little fear of contradiction that to find such a representation on a distant planet we would conclude it is the result of intelligent agency. Mountains may be complex, but lack specification. They are easily explained through the action of physical processes readily observed or modelled. The same cannot be said for statues of historical individuals.
Yes, it is really a mistake to separate the reproductive side of natural selection from the selection part. The process needs to be considered as whole.
How a sculptor creates a statue really has nothing to do with the process of natural selction; your NS analogy pretty well proves the point. The allusion to Weasel is apt; I was thinking of bringing that up, since the goal is always in mind.
—-Yes, it is really a mistake to separate the reproductive side of natural selection from the selection part. The process needs to be considered as whole.—-
Yes, this is exactly the point. NS consists of several independent processes acting interdependently. It is an iterative feedback process. If ALL PARTS of the process are not considered together, then one is not properly addressing the process itself.
Or you can be like Gil, looking individually at each part, dismissing it because it’s not the whole, and then concluding that since none of the parts are the whole, the whole itself is “ludicrous”. Which he knew before he started, of course!
Goals and specifications are beside the point, a red herring. What we have is a process that is continuously creative, producing “endless forms most beautiful” without any goals, intentions, motivations, purposes, or specifications.
This is of course NOT to say that intentional process cannot also be creative; they can. So can accidents.
If we were to find a sculpture on a distant planet, we would have NO WAY to determine intent, since intent (as you say) requires a specification, and the specification comes from our own prior contextual knowledge of what we’re observing. We know that David is intentional because we know a great deal about its history. If we knew absolutely nothing about an object, it is THEREFORE unspecified, by definition. We could only assess its complexity. But nonetheless we could conclude that SOME process created it.
Cheers; nice summary. Nothing more to add.
Whose mind? 🙂 The key is producing more offspring than rivals in the current selective environment. If the current selective environment includes a designer with a long-term objective (eg “Methinks it is like a weasel”), then you will have more offspring if you conform to that long-term objective. If it includes no intentional agent, something else can still cause one variant to have more offspring on average than another. It is a learning algorithm of sorts – ‘information’ about more-optimal survival strategies in recent environments is integrated into surviving genomes, thanks to the strategy of “throwing stuff away” and reproducing the remainder with variation.
The real problem for would be designers is knowing what variants will be successful (and how to make functional variants). I’d like to see a design advocate explain how the designer knows, and how He acquired the knowledge.
Yes – I finally caught up last night with a BBC program I taped a couple of years ago (The Secret Life of Chaos). Torstein Reil of games company Natural Motion described how he used a GA to start with very simple, and pretty useless, stick figures that kept falling over, select the better movers and breed from them, with undirected variation. His goal was obviously to produce the ‘best’ movers he could, but the long-term strategy is served by the simple, immediate selection process of sifting a current population by better and worse. He ended up with motion that was uncannily, and artistically, realistic – and had no idea how, programmatically, it was achieved. As good an example as I have seen of how creative “throwing stuff away” can be.
He simply selected the better variant algorithms available at any given time. Any Designer with any sense would use Selection and random variation. After all, what are the organisms ‘for’? As far as we can tell, simply survival in the environment. Let the environment pick ’em, then!