|
Author
|
Topic: Wheels of Life
|
Rex Kerr
Member
Member # 632
|
posted 24. June 2003 23:13
Mike,
Rather than generate an entire developmental genetic pathway, when no such pathway has been worked out for extant organisms, I shall simply restrict myself to using structures that are observed elsewhere in biology.
Here is a sketch of a biowheel:
It uses fairly standard tissue types. The only "interesting" part molecularly is the free rotation joint, which could be generated by production of a gene product by both the axle limb bone and the tyre epidermal layer; cells receiving a high level of both would undergo apoptosis, while those on either side of apoptosing cells would differentiate into epidermal cells that released smooth, tough epidermis lubricated by oils. Pressure-sensitive neurons would release growth factors when understimulated and necrosis factors when overstimulated in order to maintain a smooth interface. The fluid-filled cavity at the end of the axle bone allows the embedded wheel artery to stay fixed at the wheel bone but rotate with the edge of the wheel.
A tough band of connective tissue and muscle would lie on the inside of the wheel and press against the bulb on the end of the axle bone, to keep the organism from losing a wheel.
How is this any more difficult than the differentiation of, say, a hand?
IP: Logged
|
|
charlie d.
Member
Member # 159
|
posted 24. June 2003 23:25
quote:
Mike:
The point remains that the criteria, “complex-looking thing that does something reasonably well,” is insufficient for predicting more wheels, even hypothetical class known as molecular threaders. If it’s simply about complex-looking things to do something reasonably well, a bag of biomolecular soup would qualify. Yet if this original perspective of the cell turned out to be correct, I would not have the same degree of suspicion about design.
I'll have to take your word for it. My suspicion is, however, that many a committed IDist would say instead: "Look what a marvel of engineering a bag of soup is, how many things it can do based on such a simple principle. If I were a bioengineer, I'd make cells exactly like a bag of soup, because that's how they'd work best." quote:
As for detecting wheels because we make them, that’s the point. As long as our design science overlaps with the design science of the cell’s designers, detection remains possible. If something about the cell was modeled after “moleculat protonic snorflblatts, there would be no overlap and we’d thus score a false negative or miss an important part of the picture.
Perhaps, or as I said one could say that molecular protonic snorfblatts are the best things since water-cooled reversible molecular outboard motors, and the best way to engineer one is the only one they know about, standing right in front of them. quote:
As for fish and aquadynamic principles, are you suggesting there is something about environmental constraints that imposes wheels into cells? The flagellum is a great example, as we are often reminded that bacteria don’t need flagella to be motile. Clearly, “environmental constraints” didn’t impose the flagellar wheel on bacteria.
Well, considering that flagella of different fashions have evolved at least three times, I see 3 possibilities: 1. the bioengineer wanted to try different models, even if some critters would clearly get the sub-par one; 2. there is some constraint that makes flagella better than other solutions; 3. it's not that difficult to evolve flagella, possibly because the starting material (filamentous polymeric proteins, extrusion/secretion systems, and mechanisms to convert chemical energy stored in ATP or ionic gradients into allosteric protein structure changes) are essentially ubiquitous in nature.
But of course one should also note that not all fish are very aquadynamic. And many bacteria are happy being non-motile, happily bounced around by brownian motion, and many others use non-flagellar locomotory systems. Do we know that the flagellum is indded the best engineering solution? Last time I heard, it converts less than 10% of its use energy into motion. Are we sure that, ATP-for-ATP, squirming (or something else) is not better? Are we sure there in fact is not a betetr engieering solution that is not being used in nature at all? It could well be that flagella are just a big waste of energy (to build and to run), and the poor fellas are just stuck with them.
One can't just assume good engineering, works backwards from there, ignoring alternatives and real engineering parameters, and reach the conclusion of good engineering. If that's where ID wants to go, than ID advocates should train as biophysicists and put some meat on their good engineering principles assumptions. I think that would be a worthy research project.
IP: Logged
|
|
Rex Kerr
Member
Member # 632
|
posted 24. June 2003 23:28
Nelson, I'd say that this Pratt & Whitney PT6 turboshaft
is every bit as complex as a flagellum, if not more so. It is larger, though, so it is easier for us to understand.
Further, wheels would be useful in terrain that was primarily flat, such as ice/snow and desert conditions (especially salt flats). Since you could put wheels on the ends of short limbs, the organism could walk around if necessary. (My sketch above is for an unpowered wheel, but there are a variety of ways to get a powered wheel, such as differentiating a friction band at the outer edge of the free rotation joint, and clamping it down with muscles and twisting the axle arm. Or it could simply kick with its hind legs and then glide on its front wheels.
[ 24. June 2003, 23:29: Message edited by: Rex Kerr ]
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 24. June 2003 23:42
Rex,
I think the fact that it's made of human made parts alone makes it much less complex than a bacterial flagellum. Especially since it can self-assemble. As far as it's design, I don't know too much about it to say, but the two factors that I cite alone make it much less complex than the flagellum. I forget the exact number (I wrote it and referenced at ARN) but if we enlarged the flagellum it would reach speeds that we probably have never seen before.
Maybe I'll comment on the actual design of it and your biowheel in the near future.
I don't think that wheels are useful on snow/ice or desert (last winter made me permanently store my car in my driveway, and I can't even ride a bicycle on the beach). Salt flats are a different story but they aren't very hospitable.
[ 24. June 2003, 23:46: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
Rex Kerr
Member
Member # 632
|
posted 24. June 2003 23:50
Can you give reasons why being constructed from metal makes something less complex?
And I can ride my mountain bike just fine on wet sand and most of the nearby desert. Dry beach sand is a bit difficult, but with much fatter tires it might be okay. I agree that there wouldn't be a huge population of wheeled organisms, but that needn't matter to a designer. It might matter to evolution.
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 25. June 2003 00:07
Rex,
I don't have time for the moment, but do you really think that metal parts are more complex than protein parts? Have you ever seen that thing you link to up there: carry out the feat of building all it's parts, delivering the parts to the construction site, and moving the correct parts to the upper floors while adhering to the design specifications with a high degree of accuracy? When the appropriate length is reached, does it turn off the assembly of one part of the structure and switch on the assembly of the next part? Are there checkpoint mechanisms that deterimine whether one component has been completed and that it is okay to start construction of the next component? Do you see this information conveyed to the expression of the metal parts?
Of course this was derived from a Shapiro quote (some of the wording is identical to Lucy Shapiro's mini review), but you see what I mean by more complexity of the flagellum even without comparing the designs. There are a hundred reasons to think that protein parts are more complex than mere metal parts, but thats just my opinion.
I am a little skeptical of the claim that it is simpler to ride a bike through any kind of sand then to just walk it using two feet. But then again, there are plenty of desert creatures that shed skin that can make a simplistic wheel and get some use out of it. I don't see why we don't see any. Maybe because it's just useless.
[ 25. June 2003, 16:19: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
nobody
Member
Member # 145
|
posted 25. June 2003 02:39
Mike Gene says:
quote:
1. Cells are an example of nanotechnology, while the products of our engineering are not.
2. Design principles that may apply at one scale do not necessarily translate to another.
3. The cells we study are not the direct products of a designer, but the descendents of billions of years of evolution.
4. It is reasonable to assume that if cells were designed, a design science far more advanced than ours was used.
1. Correct. Cells are nanotechnology. However you would allow that we are making some primitive steps in the nanotechnology field, wouldn't you? I just recently heard that the worldwide investment into this field is in the billions of dollars.
2. That's true. But it's not only the scale that's a factor. On top of that, different materials can also force different design parameters.
3. Life might have been here on Earth for billions of years, but after reading your comment on the first page, "Given that there is growing evidence that DNA replication occurs in fixed protein factories, I expect this model to pan out. This would highlight just how artificial-like life is at its core. Here we’d have protein wheels being loaded on the DNA in their proper orientation and spinning down the DNA until they reached a proper location where they become anchored at fixed sites. Now, as they spin, they begin to thread the DNA back into replisome factories. And it’s worth mentioning that the same MCMs also license the DNA for replication." I do have a question. Do you think life is possible without DNA replication? My point being that early life was also high technology. Also, the more we look into life the more high technology we find.
4. Not just reasonable, but mandatory. Although they designed and built some very impressive pyramids we can immediately rule out the technology of the ancient Egyptians, to choose just one example.
Mike Gene also says:
quote:
What all these proposed wheels have in common is that they form ring structures and handle polymers in an ATP dependent fashion. I’ll label these machines as the molecular threaders. In a sense, cells do have “cog wheels,” only they are more sophisticated than Paley’s watch.
Paley would have been shocked at the current level of human technology, yet we are still not able to match the design, assembly and programming of the nanotechnology called life. And, since he used a watch as an obvious example of design, I wonder what he would have thought about God's high tech watch, the circadian rhythm? Maybe I should start a seperate thread on that.
IP: Logged
|
|
yersinia
Member
Member # 324
|
posted 25. June 2003 02:48
Another bit on wheels:
quote:
Some thoughtful points provided by Dawkins, but it should be pointed out that other macro-scale features like horns and armor don't require vessels or nerves. Dawkins asks this question for the macro-scale wheel:
"But what would the evolutionary intermediates have looked like?"
The same question remains in play for the micro-wheel.
Just to be picky, antlers at least (e.g. deer antlers) do require blood vessels to grow (deer scratch off the velvet before breeding season and then the horns fall off). Other forms of armor etc. I'm not so sure, it depends if they are derived from hair or bones or what.
But yes, it's the intermediates that are the key. The twisting blood vessels seem to be impossible to gradually convert into vessels with rotary joints.
As for whether car-style wheels, if intelligently designed into a critter, would even be favored vs. legs, I'm not so sure. The african savanna and similar plains areas seem like the most likely places (salt flats and beaches etc. are pretty restricted environments).
That appears to be the primary conclusion of LaBarbera, who actually got an article out of this bit of speculation we're doing:
http://www-news.uchicago.edu/resources/quantrell/labarbera.html
quote:
“It was one of those curiosity questions that you get asked after a lecture that has nothing to do with the lecture,” LaBarbera said. The student came up and asked, “Why don’t animals have wheels?”
LaBarbera gave the stock answer. Animals with wheels would have to evolve a circulatory system that could pump fluids through a rotating joint, an unlikely though not impossible development. But LaBarbera further wondered if wheels would really help an organism. The answer is no.
“Wheels are useful on manmade, artificial terrain like streets and parking lots, but they’re not terribly useful on natural terrain,” LaBarbera said. “If you put on a pair of Rollerblades, you can go like crazy down the sidewalk, across the street and around the parking lot, but don’t try it on the grass. It doesn’t work.”
LaBarbera also asked himself in what natural environments would wheels actually be useful? It turned out that rolling forms of locomotion––the equivalents of wheels––have evolved in those environments where the vegetation is sparse and the landscape is relatively flat. On the African savanna, for example, dung beetles make little balls out of elephant dung, which they use to feed their larvae.
“The student forced me to think about the question in a different way than I ever had and indeed than anyone had ever thought about it,” he said. “The real answer turned out to be a lot richer, I think, than the standard answer.”
And the American Naturalist article that the question inspired continues to be LaBarbera’s most frequently requested reprint.
The article is online in JSTOR if you have access:
quote:
Why the Wheels Won't Go
Michael LaBarbera
American Naturalist, Vol. 121, No. 3. (Mar., 1983), pp. 395-408.
Abstract
The scarcity of rotating systems in nature is a function primarily of the limited utility of such systems in natural environments; constraints intrinsic to biological systems (such as physiological problems of nutrient supply) are of secondary importance. In aquatic environments, rotating systems are advantageous only at low Reynolds numbers; in terrestrial environments, rotating systems are feasible as a form of transportation only on relatively flat, open terrain and become less useful as the size of the rotating element decreases. Prokaryotic flagella are popularly believed to be the only rotating system in nature, but dung beetles and tumbleweeds also use such systems for transportation. Whenever rotating systems are a feasible mode of transportation, organisms have evolved that use these systems.
IP: Logged
|
|
Jack Foster
Member
Member # 79
|
posted 25. June 2003 11:47
Hi RBH:
quote:
The principal result of interest from the hardware evolution work is that given an environment that rewards some functionality and given some standard evolutionary operators, hardware evolves novel ways of performing that function. And that outcome is independent of whether the selective environment was imposed by humans interested in studying evolution or by the 'natural' world.
The outcome itself is also dependent upon the evolutionary operators and has internally-imposed phase space and potential trajectory; why focus exclusively on the selective environment? An environment is not sufficient to evolve anything. What are the distinguishing characteristics of the operators of biological life's evolution that allowed for life's remarkable result? I recognize the importance of selective environment, but that's only part of the story.
Remember that Rex said to Mike:
quote:
But if this is an issue only of teleology vs. non-teleology, it seems as though experimental methods to address the question may be difficult--you seem to be suggesting that teleological design plus evolution may look an awful lot like non-design.
I guess I just want to truly find an example of non-design. What does "non-design" mean? If an evolutionary system is designed, then aren't the products of the system an example of design, and not of non-design? So I asked:
quote:
Are you aware of any non-biological evolutionary system that is not a product of teleological design?
I think this is a valid question. In every reference that you provided, RBH, the evolutionary system had a designer. Is anybody aware of any "accidental" evolutionary system that has chanced into being? If not, then that absence is significant. The inductive inference is that evolutionary systems require designers.
Here's an interesting class project for some class of artificial evolution. Student's task: to design a form that will evolve a wheel! With the right technology, with the right operators, . . . I think the task could be easily accomplished. But without them, the task is impossible regardless of selection.
The investigation of potential teleology in life is fascinating. We have a great model, a great example: our own pursuit of evolutionary design. I don't see why the investigation should be limited to direct teleogy, and the more Science that can be brought to the table to further the investigation, the better.
quote:
Jack: Some thoughtful points provided by Dawkins, but it should be pointed out that other macro-scale features like horns and armor don't require vessels or nerves.
Nick: Just to be picky, antlers at least (e.g. deer antlers) do require blood vessels to grow . . .
But they don't require blood vessels to fulfill their function as weapons or armour. That's my point with wheels; organisms can grow wheels that don't require the continued access to blood vessels.
IP: Logged
|
|
John Bracht
Member
Member # 5
|
posted 25. June 2003 11:55
Rex,
Very cool diagram of a biowheel. I have a question, though: wouldn't the blood vessel (in green) still get all twisted up inside, even though it comes out of the bone into a fluid-filled cavity? As best I can tell from your diagram, the vessel is one continuous entity from the end of the bone to the circumferential vessels in the wheel.
It seems like you want the vessel to empty into the fluid filled cavity (maybe this is what you meant) and then the cavity has ducts connecting with circumferential vessels.
Here's a question: wouldn't the rotation of the wheel prevent proper circulation to tissues in the wheel, simply because of centripetal force pushing the blood to the edges of the wheel?
One final nit-picking point. It looks to me like your tough bands of connective tissue go the wrong way to keep the wheel from popping off. You'd need something which connects on the other side of the wheel and somehow attaches to the bone. Or maybe I'm misunderstanding your idea and you meant that the connective tissue is rod-like and stiff, in which case pushing against the bone knob might work to hold the wheel on.
(added in edit): I just went back to the diagram and realized that the connective tissue bands have another problem: they are entirely intra-wheel, that is, inside the rotating part of the wheel. Yet to hold the wheel onto the axle, they need to connect the rotating part with the stationary part (axle) somehow. But this re-introduces the whole twisting problem because the connective bands would quickly get terribly twisted and eventually stop the wheel's rotation. Maybe connective tissue bands to hold the wheel on just won't work.
Thanks for making the diagram, it's useful to think about biological design this way!
John
[ 25. June 2003, 12:03: Message edited by: John Bracht ]
IP: Logged
|
|
zygotecowboy
Member
Member # 220
|
posted 25. June 2003 12:49
quote:
Nelson said:
I think the fact that it's made of human made parts alone makes it much less complex than a bacterial flagellum.
Huh? I thought complexity is defined as the inverse of the probability that the object originated via chance and natural law. If something is made of human-made parts alone, then I would argue that the object is VERY complex, at least in the Dembskian sense. The probability for the origin of such an object certainly falls well below the universal probability bound of 10^-150. In fact, since evidence overwhelmingly suggests that objects made of human-made parts alone are designed, the probability is effectively 0.
Although in NFL, Dembski arrives at a very low probability for the origin of the flagellum as well. Perhaps we need to do the same calculations for the object Rex linked to...
zc
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 25. June 2003 16:25
Jack F.,
You're right: The "evolutionary system" in the hardware evolution examples is designed by humans. Part of my point was that designed systems like those produce outcomes that cannot be anticipated. Indeed, one main argument of Thompson's "Notes on Design" article is that there is a class of problems that are intractable to forward or inverse modeling, and thus an algorithm that specifies the steps the construction process must take is impossible to write. One cannot in principle specify the steps, either backward or forward, between initial conditions and outcome.
What does it mean to say of an outcome that it is "designed" if the designer (the guy who wrote the system) cannot anticipate what that system will produce? Since the designer (programmer) cannot anticipate the outcome, that outcome cannot have been intended except perhaps in the most generic sense - 'I anticipate something or other will evolve that can accomplish the task(s) defined (implicitly or explicitly) by the conjoining of a set of evolutionary operators, a selective environment, and a fitness function.' To return to the Lenski paper for a moment, recall that 147 different assembly language programs evolved to perform EQU. That some program would evolve to perform EQU is a generic expectation; that any specific program would evolve is a different kettle of analysis. If one knew what program would evolve, or what circuit would evolve, then the hardware design people would not be interested in EAs: They'd just write the program down, or wire up the circuit. It is precisely because one does not and cannot know what will be produced by evolution that one uses an EA.
In an EA like those under discussion, the intelligent agent does not (and in many cases cannot in principle) specify either the form of the outcome or the path from the initial conditions to the outcome. The designer does define a search space, though in platforms like Avida that allow insertions and variable-length genomes, that space is dimensionally plastic. The designer borrows the evolutionary operators of evolutionary theory - mutations of various kinds, recombination (often but not always), competition for limited resources, differential reproduction as a function of fitness, and mortality of individuals. But because of the intractability of the problem to analytic modeling, no specific outcome can be 'aimed' at. The intractability to modeling also implies that given just the product, one cannot specify with high confidence the path back to the initial conditions. Neither forward nor backward modeling is possible for some systems. One might be able to hypothesize tentative or even plausible rough pathways, but reconstructing the precise path from only knowledge of the structure of the final product is impossible.
I conclude that in those cases where a problem is intractable to forward and inverse modeling but where an EA produces a satisficing solution to an adaptive problem, the designer could not in principle have intended that solution to occur! "Intention" is not appropriately applied to the specific outcome of an evolutionary process running in a context intractable to forward and inverse modeling because the designer has no grounds on which to expect any specific outcome. There may be 'intelligent' design associated with the (meta-)structure of the system - say the conjoining of operators mentioned above - but there can be no intentionality associated with the outcome: the result cannot have been "intended" by the designer.
It follows that intelligent design cannot be inferred from the properties of, or even the existence of, specific products of an evolutionary process, if "intelligent design" includes some notion of intentions of the designer with respect to specific products, because "intended" cannot apply to those products.
It follows further that the intelligent design movement must argue that it is the larger system that was designed, and not any specific product or constituent of it. That's actually Dembski's move in the "referral" business, where information in biological systems is 'referred' to the environment in which they evolve. In this view, the evolution of a population is a transcription process, transcribing information from environment to a biological population. Even there, no specific product of evolution can be said to be "intended." What is "intelligent design" without intention?
RBH
Late Edit for a bloomer of a spelling error.
[ 25. June 2003, 17:26: Message edited by: RBH ]
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 25. June 2003 20:10
Zygote,
My post did not imply that the human designed turboshaft was not complex, just that it is not as complex as the flagellum. Furthermore, we can know that something is designed and still calculate the probability for chance and law to build it. The probability can approach zero but it will probably (pun intended) never be zero, whether we know it was designed or not.
[ 25. June 2003, 21:22: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
John Bracht
Member
Member # 5
|
posted 25. June 2003 20:28
RBH wrote
quote:
What does it mean to say of an outcome that it is "designed" if the designer (the guy who wrote the system) cannot anticipate what that system will produce? Since the designer (programmer) cannot anticipate the outcome, that outcome cannot have been intended except perhaps in the most generic sense - 'I anticipate something or other will evolve that can accomplish the task(s) defined (implicitly or explicitly) by the conjoining of a set of evolutionary operators, a selective environment, and a fitness function.' To return to the Lenski paper for a moment, recall that 147 different assembly language programs evolved to perform EQU. That some program would evolve to perform EQU is a generic expectation; that any specific program would evolve is a different kettle of analysis.
and later on,
quote:
Intention" is not appropriately applied to the specific outcome of an evolutionary process running in a context intractable to forward and inverse modeling because the designer has no grounds on which to expect any specific outcome. There may be 'intelligent' design associated with the (meta-)structure of the system - say the conjoining of operators mentioned above - but there can be no intentionality associated with the outcome: the result cannot have been "intended" by the designer.
Maybe this is just semantics, why can't we say that the result of an evolutionary process (like SELEX, or the Lenski simulation) were intended? Indeed, it seems a bit odd to suggest that the outcome of the Lenski simulation was entirely unintentional. Rather, a specific goal was intended by the designers of the program, who set up the conditions such that the goal (evolution of organisms capable of carrying out the EQU function) was met. That's intentionality if I ever saw it.
I think you're making a distinction that really doesn't matter, and you're grasping at straws to show why it makes a difference whether you can predict the specific outcome. It seems to me like a distinction without a difference. It's still intentionality, it's still goal-directedness, it's still design.
John
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 25. June 2003 22:22
John wrote quote:
Maybe this is just semantics, why can't we say that the result of an evolutionary process (like SELEX, or the Lenski simulation) were intended? Indeed, it seems a bit odd to suggest that the outcome of the Lenski simulation was entirely unintentional. Rather, a specific goal was intended by the designers of the program, who set up the conditions such that the goal (evolution of organisms capable of carrying out the EQU function) was met. That's intentionality if I ever saw it.
A couple of remarks. First, the question the Lenski, et al. study asked was whether under the conditions specified, a program capable of evaluating EQU would evolve. As far as I can tell, they didn't know the answer in advance and so couldn't "intend" such a program, unless to ask a question is to "intend" its particular answer, whatever that answer might turn out to be. That's a peculiar usage, in my opinion. In fact, if Lenski, et al., "intended" to evolve programs to perform EQU they did a bad job of it: More than half the evolutionary runs in all conditions failed to evolve a program capable of performing EQU in the 100,000 cycles allowed.
None of the specific outcomes - the particular programs that evolved to perform EQU - was "intended." That is, the experimenters didn't set up the initial conditions such that a particular lineage would lead to a particular program. They couldn't.
John's remark that the "experimenters set up the conditions such that the goal ... was met" presupposes that the experimenters knew the conditions that would lead to the evolution of a program that could evaluate EQU. I'd like to see some specific grounds for that remark. Is there something in addition to the experimenters supplying raw materials (primitive instructions) and a fitness function that rewarded the performance of simpler logic functions? If so, the charge is vacuous. They ran appropriate controls to more closely define the conditions under which such a program would evolve and report those, too. To argue that a program that evolved to evaluate EQU was "intended" is to argue that the result of every experiment ever performed was "intended." That renders the notion of intentionality vacuous.
John's quoted remark amounts to a concession that there are conditions under which evolutionary operators can produce irreducibly complex objects, and those conditions include the absence of a direct incremental pathway from randomness to the final outcome. His objection must appeal to the mapping of that conclusion into biology from the Avida simulation; it cannot be sustained in the context of the Lenski study, nor, for that matter, in the hardware evolution studies. There has to be something special about biology that renders those studies irrelevant to questions of evolution of IC structures in biology. We see multiple "solutions" to similar adaptive problems in biology, just as we do in the Avida simulation. There are lots of kinds of bacterial motility adaptations, and in fact a number of different kinds of flagella. Why does that not speak to how those multiple solutions may have come about? That's what I want to hear John describe.
John's concession that there are conditions under which irreducibly complex (by Behe's operational definition) can evolve fairly easily shifts the burden to John to show that cannot be generalized to biological evolution. Critics of ID argue that cooption and indirect pathways are sufficient to evolve IC structures and processes, and the Avida simulation demonstrates that the argument holds water. IC defenders say that's unlikely, but obviously there are conditions under which it is not at all unlikely. There are also instances in the molecular biology literature (see the blood clotting work cited here) where what Behe advertised as an IC system has extant versions that lack one or more of the allegedly indispensable components - if the mammalian version is IC, it could well have evolved from precursors like those identified in that literature. If it is not IC, then another gap is filled and 'IC of the gaps' is forced back further.
I don't think asking what is intended is merely a "semantic" question in a trivial or pejorative sense, but it is "semantic" in a very important sense. Unless what is intended can be identified with some precision, to say whatever comes out the other end of an experiment is what was intended is to live in waffle city. I'm reminded of my cat, who when she falls off a narrow branch instantly looks up as if to say "I MEANT to do that!" If we attribute "intentionality" to the outcome of every experiment we run, then it is a truly vacuous notion.
I repeat the assertion from my posting just above: quote:
I conclude that in those cases where a problem is intractable to forward and inverse modeling but where an EA produces a satisficing solution to an adaptive problem, the designer could not in principle have intended that solution to occur!
The intelligent designer - the programmer - may hope that some solution or other to the problem being addressed will evolve, but that is exactly the question addressed by the Lenski, et al., research: 'Are these conditions sufficient to allow a (not "the") solution to evolve?' The answer was yes. So now we know more about the conditions under which an irreducibly complex object can be produced by blind evolutionary processes.
But once again, what can be anticipated is that some sort of solution or another may evolve; the specific solution cannot be anticipated under the conditions I described in my earlier posting. That inevitably means that one cannot 'set up' conditions guaranteed to produce the "intended" outcome (the Avida platform is very different from MESA), unless "outcome" is defined generally as "something or other that will allow the organism to adapt to these circumstances". And that means that an intelligent agency cannot intentionally 'set up' evolution to produce a particular biological structure or process or indeed, a particular species of organism. Not Lenski, et al. nor Dembski's unembodied intelligent agency can do that any more than either of them can provide an analytic solution to the multi-body problem in celestial mechanics.
I have no need to grasp at straws. The experiments are designed, but the outcomes are not. The experimenters intended to find an answer to a question; the particular answer found was not the goal, an answer was.
RBH
[ 25. June 2003, 22:26: Message edited by: RBH ]
IP: Logged
|
|
|
Topic Closed
Printer-friendly view of this topic
