|
Author
|
Topic: Is "complexity" relative or absolute?
|
Evan
Member
Member # 164
|
posted 29. March 2002 09:10
As I read this thread, I see the problem that I have tried to address in the Evolution and Design thread: that ID advocates primarily focus on establishing what evolution can’t do, and seldom on providing enough specificity about ID that one could study what ID can do.
For instance, Drosera writes in one post “Hi Paul, ...could you please define what you mean by 'ID'?” and then goes on to discuss several different views on exactly what some ID advocates say evolution can’t do.
Later Drosera also writes, “the real question for me is: independently of how RM/NS could have originated IC, how does ID propose it originated? Until some actual mechanism is proposed, there is no way of fairly comparing the explanatory powers of the 2 theoretical systems.”
The problem, then, seems to be that even if there are difficulties in thoroughly understanding how natural evolutionary processes can create biological novelty, the absence of any details about an alternative ID theory keeps us from establishing any tests to determine whether ID is a better explanation. In fact, unless ID has some details, it doesn’t seem that it is an explanation at all, but rather just a statement that something else other than natural evolutionary processes happens.
In reply to Drosera’s remarks, Paul Nelson writes, “Drosera, for the sake of discussion, let's us the best-articulated theory of design inferences (ID) currently available, namely, Dembski's. Let me know if you have a copy of No Free Lunch, and we'll take it from there.”
But my understanding of NFL is that it is primarily (or maybe exclusively) an argument about what natural evolutionary processes can not do. I wasn’t aware that there was an actual theory of design described in NFL, other than CSI has to come from somewhere other than nature.
So I ask this question of Paul: could you summarize in a few brief sentences what theory of design Dembski’s offers in NFL, or offer one of your own?
For comparison sake, the hypothesis about ID that I offered in Evolution and Design contained the following components:
1) accepts the geological age of the earth, the progression of life forms laid out in the fossil record, and common descent (at least from the Cambrian on.)
2) hypothesizes that the designer acts upon the genome at conception - that is the place and moment that design enters the world
3) hypothesizes that the designer acts in a non-miraculous way (without contravening any laws of nature) and yet also in a way that is non-detectable. These ideas are spelled out in a little more detail in ID Coming Clean by Dembski. That is, the designer uses exactly the same mechanisms that natural processes use (point mutation, gene duplication. etc.), but manipulates the chance aspect of these so that what would appear random to us is actually, in the long run, a part of a longer designed sequence of changes
4) claims that each act of design is constrained by the fact that the resulting child organism must be enough like the parent that development, birth, and nurturance can take place. (See Wells “Evolution by Design for more details, although Wells denies common descent, and I don’t)
So, again to Paul, two questions:
1) Can you supply a similar summary of a theory of design so that we can talk more specifically about what ID can do, as opposed to what evolution can’t,
2) Or do you think that such specificity is not necessary for some reason, and that this should not be such a large concern?
Thanks
[ 29 March 2002, 09:15: Message edited by: Evan ]
IP: Logged
|
|
Paul A. Nelson
Member
Member # 26
|
posted 29. March 2002 09:36
Charlie wrote:
Therefore, in the absence of falsification or better alternative explanations, we would be crazy not to use RM/NS as our basic model for the evolution of all biological features which are reducible, at least in principle, to progressive pathways along a fitness axis (such as eyes and wings, for instance). At our current status of knowledge, I think it is heuristically just useless to argue that Darwinian evolution is not the way most change is brought forth in nature, since this is in any case the hypothesis tested every time data about a system do become available, and the more we learn, the more opportunities for falsification will abound.
I'm not sure we disagree with each other, at one level, anyway. If one conceives of the explanatory filter on the model of "let's take all the natural causes we know and run them at this event to see if they work," then of course one must test RM & NS in the course of trying to explain the origin of a biological system.
Where we differ, I think, is on the question of falsification (explanatory failure) and positing causes other than RM & NS. Let me use a shopworn but still helpful analogy. Your position -- i.e., stick with RM & NS because they work in some easy cases and may well work, with more knowledge, in the currently intractable cases -- reminds me of a detective who is determined to explain all deaths in terms of strictly natural causes (e.g., illness, accidents). "That woman we found last week had a heart attack. Nobody killed her deliberately. Most of the people who die do so naturally. So let's stick with natural [non-intelligent] causes until we're absolutely certain they won't work."
The origin of biological complexity -- what Darwin claimed to be able to explain with RM & NS -- remains unsolved 150 years after the Origin. Why isn't it time to move on, to consider other possibilities?
Evan, you asked:
So, again to Paul, two questions:
1) Can you supply a similar summary of a theory of design so that we can talk more specifically about what ID can do, as opposed to what evolution can't,
2) Or do you think that such specificity is not necessary for some reason, and that this should not be such a large concern?
There is not now, and may not be for some time, a biological theory of ID as well-articulated as neo-Darwinism. When I lecture to university audiences, I encourage students to take a path to ID that passes first through a complete education in evolutionary biology.
In a way, however, the lack of a well-developed biological theory of design doesn't really matter. (Of course it does matter, but I'm being deliberately provocative for a moment.) If organisms were designed, science will discover that, whether scientists want to or not. Indeed I think within the past couple of decades, evolutionary biology has been running up against the limits of RM & NS, and roughly mapping those out.
The purpose of this board, and of ISCID generally, is to discover or construct a testable theory of design. If we already had that theory, instead of the bag of intuitions we currently possess, most of the discussion here would be moot. Instead, we'd be out applying the theory in laboratories and in the field.
Finding out what "evolution can't do," as you put it, is part of the picture. I agree that that is not very satisfying as science, but one has to use the tools available, and right now, neo-Darwinism is the best-articulated theory.
If you don't want to work on ID until it reaches a level of sophistication akin to the body of evolutionary theory accumulated since Darwin, then this board is the wrong place for you. ID theory will be a loud, dusty, dirty workshop with stuff flying around, arguments, uncertainties, frustrations, confusion, for the forseeable future.
But right now, of course, is when all the really fun and deeply creative work needs to be done. After the theory is assembled...not as much fun.
[ 29 March 2002, 09:57: Message edited by: Paul A. Nelson ]
IP: Logged
|
|
Art
Member
Member # 179
|
posted 29. March 2002 10:02
Paul said: quote:
**I looked at the passage you cited, and Dembski writes that "this [the origin of URF13] is still uncomfortably small, but well above the universal probability bound" (p. 219). I haven't seen the Cell paper, or any of the other data, on which this counterexample rests, however, so I might well disagree with Bill about its relevance. If you'd like, I'll track down the original paper and other relevant sources, and we can go into the nitty-gritty. Let me know.
The "nitty-gritty" has been explored at length. Here and, in more detail, here. The second thread is the one to which Dembski refers in his book.
IP: Logged
|
|
James A. Barham
Member
Member # 50
|
posted 29. March 2002 10:04
I would like to second what Paul Nelson is saying.
It is crucial that Charlie D., Drosera, and others keep in mind the distinction between showing that a relatively smooth pathway from one functional form to another exists, on the one hand, and claiming that RM/NS is an adequate explanation of such a transition, on the other. Many of us agree with the first of these propositions (that evolution is the overwhelmingly most likely explanation of the evidence), but dispute the second (that the theory of natural selection sheds much light on evolution).
As James A. Shapiro has recently written: "Rather than being restricted to contemplating a slow process depending on random (i.e., blind) genetic variation and gradual phenotypic chnage, we are now free to think in realistic molecular ways about rapid genome restructuring guided by biological feedback networks." ("Genome System Architecture and Natural Genetic Engineering in Evolution," in L.H. Caporale, ed., Molecular Strategies in Biological Evolution, New York Academy of Sciences, 1999, pp. 23--35).
To be sure, there is a tendency to take whatever new ideas come along, and "assimilate" them to the neo-Darwinian perspective, and I am sure that is what many will try to do with the revolution now occurring in molecular genetics. But it seems to me that the idea that organisms are able to intelligently guide their own genetic restructuring, and indirectly their own evolution, contradicts the spirit of Darwinism, if anything could be said to contradict that notoriously vacuous set of ideas. We are in new intellectual territory here, and we ought to be brave enough to recognize that fact.
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 29. March 2002 10:27
I appreciate the Paul’s candor in distinguishing concerns about what evolution can’t do from alternative proposals.
However, I am puzzled when Paul responds to me by writing,
quote:
The purpose of this board, and of ISCID generally, is to discover or construct a testable theory of design. If we already had that theory, instead of the bag of intuitions we currently possess, most of the discussion here would be moot.
If you don't want to work on ID until it reaches a level of sophistication akin to the body of evolutionary theory accumulated since Darwin, then this board is the wrong place for you.
But it seems to me that I, as much as anyone on this board, am trying to construct a testable theory of design. Of course, I don’t expect it to be as “well-articulated as neo-Darwinism.” But we have to start somewhere, it seems.
So, in the interest of trying to explore specificity, I would ask Paul to comment on the four components of the hypothesis I have made. Do you think any or all of them are reasonable places to begin?
Of course, I don’t mean do you think they are true - that would be the goal of investigating them scientifically. But are they the type of hypothesis that needs to be considered in order to move towards a testable theory of design?
Thanks
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 29. March 2002 10:57
In trying to keep the conversation focused, this response is now to James:
I believe that James’s perspective is similar to my hypothesis in that he believes we might find ways in which “organisms are able to intelligently guide their own genetic restructuring.”
A crucial distinction, however, that we need to consider is whether this is done by a “something” outside of the organism - a designer - as my hypothesis supposes; or whether it is done entirely within the organism by the exercise of properties that we do not now understand or even acknowledge.
Also, it is often useful, in exploring differences, to clearly state places where there is agreement. I think James would agree (but I ask that he correct me if I am wrong), that we are in agreement about:
a) the age of the earth and the progressive existence of organisms as laid out in the fossil record,
b) common descent (a continual chain of ‘parent-child’ relationships),
c) genetic change (although not necessarily at the moment of conception) as the location of the action of intelligent design, and
d) that all the other external factors that bear on survival (such as comets hitting the earth, or whether a particular animal survives to reproduce) are adequately explained by a non-intelligent combination of law-and-chance.
So, James, would you agree with these statements?
IP: Logged
|
|
Paul A. Nelson
Member
Member # 26
|
posted 29. March 2002 11:16
Evan asked for comment on the following propositions about a possible ID theory. Very quickly:
1) accepts the geological age of the earth, the progression of life forms laid out in the fossil record, and common descent (at least from the Cambrian on.)
Of these (like Wells), I doubt common descent.
2) hypothesizes that the designer acts upon the genome at conception - that is the place and moment that design enters the world
Not sure what you mean by "conception" -- please explain. (I am sorry if you already did so in another thread, and I missed it. If so, please give the link.)
3) hypothesizes that the designer acts in a non-miraculous way (without contravening any laws of nature) and yet also in a way that is non-detectable. These ideas are spelled out in a little more detail in ID Coming Clean by Dembski. That is, the designer uses exactly the same mechanisms that natural processes use (point mutation, gene duplication. etc.), but manipulates the chance aspect of these so that what would appear random to us is actually, in the long run, a part of a longer designed sequence of changes
Did you really mean to write "non-detectable" above?
4) claims that each act of design is constrained by the fact that the resulting child organism must be enough like the parent that development, birth, and nurturance can take place. (See Wells “Evolution by Design for more details, although Wells denies common descent, and I don’t)
Organisms have to be viable, of course, but I don't see why the designer has to be constrained by what already exists. In other words, radical discontinuities are possible (cf. the discontintuities between human artifacts).
I've got to turn to other business for the remainder of the day, but will return to this thread tomorrow morning to see what's been posted.
IP: Logged
|
|
James A. Barham
Member
Member # 50
|
posted 29. March 2002 11:54
Evan:
Yes, I agree with your points a), b), and d). I think that all three schools---certainly Darwinians and self-organization folks like me, and probably many ID theorists, as well----can all agree on these three points. It is your point c)---the point at which intelligence operates---that separates all of us.
For my part, I see intelligence operating, not just at the moment of the creation of new functions, but throughout the life of each organism. That is, I see diachronic, phylogenetic change as in principle not that different from synchronic intelligent action or rational agency as such---or perhaps I should say, not that different from ontogenetic learning, to compare diacrhonic processes and not mix apples and oranges.
Of course, phylogenetic change involves genetic re-organization in a way that synchronic action and ontogenetic learning do not, but I still think it can only ultimately be understood in the context of a general and fundamental theory of the dynamics of rational agency, which of course we currently lack (although there are many hints in the literature about possible fruitful directions for research to pursue).
So, we do differ substantially on the main point, it seems.
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 29. March 2002 12:05
Paul: thank you for replying.
Breaking 1) into two parts, I will assume Paul “accepts the geological age of the earth, the progression of life forms laid out in the fossil record,” although he doesn’t specifically say this. I will return to the common descent issue later.
2) I wrote “hypothesizes that the designer acts upon the genome at conception” and you wrote “Not sure what you mean by "conception" -- please explain.”
My original hypothesis was the designer intervened at the moment of conception (for sexual reproduction) or at the moment of cell division (for asexual reproduction.) But, this is perhaps overly specific, particularly in trying to find common ground with James Barham.
So I will amend this to be the following hypothesis: Design (either externally imposed by a designer or internally generated by an intelligence inherent at the cellular level,) works by acting on the genome of those cells in the organism that are involved in passing genetic information to the next generation.
This hypothesis, by specifically pointing to where design does happen is meant to implicitly point to places where design doesn’t happen. I am hypothesizing that the designer, or design process, has no control of events outside the organism, and that, once conceived, an organism develops according to the genetic structure present at the moment of conception.
Concerning 3: I wrote, “3) hypothesizes that the designer acts in a non-miraculous way (without contravening any laws of nature) and yet also in a way that is non-detectable,” and Paul responded “Did you really mean to write ‘non-detectable’ above?”
Yes, I did, although I see I’d better explain.
In Dembski’s “ID Coming Clean,” he explains, and I am agreeing with him, that design does not posit miraculous interventions which cause breaks in the causal chain. Rather, he says, information can be inserted into the world without “moving particles.” This insertion of information, he suggests, and I hypothesize for the sake of this discussion, is done by manipulating events that would otherwise be considered truly random (such as at the quantum level.)
So any particular act of design would be indistinguishable from a random event. It is only in the cumulative effect of many events that we might find the evidence for an inference of design.
This is an important point, and I will provide a more thorough example in a separate post.
4) and the common descent part of 1: Paul writes, “Organisms have to be viable, of course, but I don't see why the designer has to be constrained by what already exists. In other words, radical discontinuities are possible.”
This is a particularly important point to discuss, I think.
When Paul write “I don't see why the designer has to be constrained by what already exists. In other words, radical discontinuities are possible,” he is making some implicit assumptions about the designer that really ought to be made explicit. In particular, he seems to be assuming that the designer can miraculously intervene in the sense of breaking (in fact, radically breaking) the chain of natural cause-and-effect.
In fact, if new organisms don’t come about by common descent, the only alternative I can think of (and I would be glad to hear of other alternatives) is an immediate materialization into existence.
Is this a viable hypothesis? What does this say about the designer?
I think there is a vast gulf between the type of inherent cellular intelligence that James Barham discusses (for which the existence of an external designer is not even necessary) and the idea that there is a designer who could, and has, materialized new organisms into existence.
So I would ask you, Paul, is some type of immediate materialization of new organisms the alternative to common descent that you would offer as an hypothesis, and do you therefore hypothesize a designer with the power to perform such miraculous actions?
IP: Logged
|
|
Jesse
Member
Member # 112
|
posted 29. March 2002 12:34
Paul A. Nelson:
If RM & NS cause CSI, defined by Dembski's universal probability bound, then we should have plenty, or at least some, examples of that happening.
You're setting up an impossible task here. Any feature of the biological world that has a reasonable probability of being generated by RM&NS will not exhibit high CSI, by definition. The question is not whether RM&NS can generate specified complexity, but rather whether any structures in the biological world actually exhibit specified complexity in the first place.
IP: Logged
|
|
charlie d.
Member
Member # 159
|
posted 29. March 2002 12:52
quote:
Paul:
Where we differ, I think, is on the question of falsification (explanatory failure) and positing causes other than RM & NS. Let me use a shopworn but still helpful analogy. Your position -- i.e., stick with RM & NS because they work in some easy cases and may well work, with more knowledge, in the currently intractable cases -- reminds me of a detective who is determined to explain all deaths in terms of strictly natural causes (e.g., illness, accidents). "That woman we found last week had a heart attack. Nobody killed her deliberately. Most of the people who die do so naturally. So let's stick with natural [non-intelligent] causes until we're absolutely certain they won't work."
But that's exactly what detectives do! If there is no evidence of foul play, every death is considered a natural death.
It doesn't matter how unlikely it is for, say, a 25 year old athlete to die of natural causes: if there is no evidence of murder, the default explanation must be natural death. This is true even if we know that murderers are indeed out there and how they usually implement their murderous designs (neither of which of course we know about biological Designers).
Note that this doesn't mean that you can't look at individual deaths for signs of foul play, but that the police just don't take the methodological position that they have to actively rule out foul play in every death.
You seem to argue that one must assume all deaths are murders until convincingly proven otherwise, including deaths occurring in rooms locked from the inside, because supernatural (or extra-natural) murderers have not been definitively ruled out yet.
[ 29 March 2002, 14:21: Message edited by: charlie d. ]
IP: Logged
|
|
Mike Gene
Member
Member # 149
|
posted 29. March 2002 16:26
Edmund: The fingerprint is not just "NS produces systems which serve to reproduce the organism's genes." It is "NS produces ONLY systems which serve to reproduce the organism's genes." If we discovered systems with any other complex, specified attributes, we could be sure that those attributes weren't put there by NS.
Fingerprints are supposed to be distinctive clues. That a biological feature serves to reproduce the organism's genes is not a distinctive clue that indicates the feature in question arose by NS. However, there is an asymmetry here, as you are right about finding fingerprints of a mechanism of propagation that does not involve NS. I'm still unclear about the fingerprints you see.
Evan: As I read this thread, I see the problem that I have tried to address in the Evolution and Design thread: that ID advocates primarily focus on establishing what evolution can’t do, and seldom on providing enough specificity about ID that one could study what ID can do.
True, but recall that Edmund specifically requested that ID theorists focus on what evolution can't do.
As for mechanisms, I have a preliminary discussion here . If someone were to ask about the mechanism that brought a Ford automobile into existence, what should we tell them? I should point out, Evan, that I am indeed quite open to your model. My focus has intelligent intervention at the origin of the first cells and their design has then influenced evolution since. However, I would not be presumptuous enough to rule out the further interventions that appear to be part of your model.
James: To be sure, there is a tendency to take whatever new ideas come along, and "assimilate" them to the neo-Darwinian perspective, and I am sure that is what many will try to do with the revolution now occurring in molecular genetics. But it seems to me that the idea that organisms are able to intelligently guide their own genetic restructuring, and indirectly their own evolution, contradicts the spirit of Darwinism, if anything could be said to contradict that notoriously vacuous set of ideas. We are in new intellectual territory here, and we ought to be brave enough to recognize that fact.
Well stated. When it becomes clear just how much organisms guide their own genetic restructuring, the neo-Darwinian perspective will simply claim that this puts such organisms at an advantage over others that could not, and thus it too will be absorbed.
IP: Logged
|
|
James A. Barham
Member
Member # 50
|
posted 29. March 2002 17:51
Mike:
I am sure you are right---alas.
IP: Logged
|
|
charlie d.
Member
Member # 159
|
posted 29. March 2002 18:08
Mike,
your discussion of mechanisms sounds to me like an intellectual cop-out. First of all, it is ID which claims darwinian mechanisms are insufficient to produce biological complexity, so I think it is just logical to expect ID to provide some alternative plausible, if not outright better, testable mechanisms. (Unlike what you say, ID per se is not a mechanism, it's just a theoretical concept; ID without a mechanism is just as heuristically useless as "natural selection" without a mechanism).
After all, if explanatory mechanisms for everything are not important, as you say, what's the problem with darwinism? At least, darwinian explanations, unlike ID's, do explain something (if only, say, minor variation within populations, or adaptation in rare cases such as those mentioned above), and help us make sense of the world as we see it.
Basically, if you have it your way, there's no doubt that, on a scientific level, darwinism just wins by default (again, I am not arguing the logical and philosophical coherence of ID's arguments, just their scientific validity).
Second, you have repeatedly stated that science cannot study, or even detect, teleological mechanisms ("To detect teleological mechanisms requires a new way of thinking"). Let me repeat this once more, this is not the case: there are many examples of apparently teleological mechanisms that have been (and still are being) tested scientifically. Adaptive mutations in bacteria are the best characterized example; see for instance this paper, or this one (sorry, abstract only available for ther latter): can the issue be spelled any more clearly? (I should note that the story is far from settled, see here or here) "Lamarckian" and certain epigenetic mechanisms with teleological implications have also been proposed and are being studied.
Personally, I think you are right: quite possibly, we'll never know how the flagellum evolved, or whether it was designed. So, as much as we all admire its perfection, it may prove entirely inconsequential for the establishment of ID as viable scientific theory. On the other hand, some microbiologist running real experiments with bacteria in a Petri dish may stand a good chance to open the door for a science of ID.
[edit: some crap cut out, some other crap explained]
[ 29 March 2002, 19:29: Message edited by: charlie d. ]
IP: Logged
|
|
Drosera
Member
Member # 139
|
posted 29. March 2002 21:44
Hi Mike, can you give us the reference for your Leigh quote? Thanx.
This thread is going in about 18 different directions, e.g. the definition of ID, the standards of evidence regarding positing natural selection as the explanatory force in both specific cases and as a general explanatory theory, the origin of information/"CSI", what Dembski actually means, what specific examples mean, what specific examples say, what NS can & can't do, etc.
Here was the original question from the beginning of the thread:
quote:
The way in which the word "complex" is sometimes used with reference to biological systems in the ID literature and fora such as this one troubles me. It is not the absolute complexity of the systems which matters. It is how unlikely the biological systems are *given the natural processes we know about*. Information only becomes complex information when we know that our "known processes" are statistically unlikely to produce that information.
From this point of view, Dembski's claim (in his Law of Conservation of Information) that natural processes cannot produce CSI runs the risk of becoming a tautology. If our declared set of "known processes" consists of all natural, non-intelligent processes, then of course they cannot produce CSI-- by definition! CSI is defined as a specified pattern which is highly unlikely to be produced by the set of "known processes". The only question that remains is: what is the probability distribution generated by "all natural non-intelligent causes"?
To put this in simple terms: what are natural causes likely to produce? In the biological context, these causes mean evolutionary processes, especially natural selection. It seems to me that we can only declare biological organisms CSI as opposed to mere SI if we can show conclusively that the patterns of these organisms are unlikely *given* evolutionary mechanisms. Which suggests that the important questions are the ones which evolutionists and creationists always dicker about:
1) What can natural evolutionary mechanisms produce, and what can they not produce?
2) How do we know that this is true?
As these are very broad questions, perhaps we can boil them down to something more specific that will also do justice to the various issues that have been brought up by various posters. E.g.:
1) Do genes exhibit Dembski's CSI?
- If they do, proceed to #2.
- If they don't, then isn't CSI a pointless concept?
- If question #1 is indeterminate, and depends on the answer to #2 (i.e. likely to come about by evolution = not CSI, not likely = CSI) then Dembski's central argument is indeed just a tautology.
2) Assuming that the answer to #1 is "yes", can natural processes (mutation and selection and the related well-known mechanisms) produce new genes, and therefore CSI?
It seems to me that unless the answers to #1 and #2 are determined to be "yes", then there is no point in moving on complex multi-gene structures yet.
Drosera
IP: Logged
|
|
|
Topic Closed
Printer-friendly view of this topic
