|
Author
|
Topic: Elliott Sober's Independent Evidence Requirement for Design
|
William A. Dembski
Member
Member # 7
|
posted 23. October 2002 22:06
In his paper "The Design Argument," Elliott Sober predicts that "human beings will eventually build organisms from nonliving materials."[1] In that case, we could obtain clear evidence that certain organisms resulted from intelligent design whereas earlier we might have thought they were due to a Darwinian process. I consider a similar possibility in chapter 6 of No Free Lunch.
Such examples, however, raise a curious dilemma for the Darwinist. Suppose humans do become sufficiently advanced technologically that they can build life "from scratch." Consider now some organism in the wild. How could we tell whether it was designed from scratch and escaped from a laboratory, or whether it was the result of a Darwinian process operating without human intervention in nature (assuming, for the moment, that a Darwinian process is capable of generating the needed biological complexity)? Perhaps there will be tell-tale signs. Perhaps evolved biology will be more efficient and less labile that designed biology. Or perhaps the opposite will be true. Perhaps, the difference will be a matter of style or art -- like the difference between Egyptian and Greek art.
But consider a related thought experiment. Tomorrow space travelers show up loaded with unbelievably advanced technology. They tell us (in English) that they've had this technology for hundreds of millions of years and give us solid evidence of it (perhaps by pointing to some star cluster hundreds of millions of light years away whose arrangement signifies a message that confirms the aliens' claim). Moreover, they demonstrate to us that with this technology they can atom by atom and molecule by molecule assemble the most complex organisms.
Suppose we have good reason to think that these aliens were here at key moments in life's history (e.g., at the origin of life, the origin of eukaryotes, the origin of metazoans, and the origin of the animal phyla in the Cambrian). Suppose further that in forming life from scratch the aliens would not leave any evidence (their technology is so advanced that they clean up after themselves perfectly -- no garbage or any other signs of activity would be left behind). Suppose, finally, that none of the facts of biology are different from what they are now. Should we think that life at key moments in its history was designed?
How would this thought experiment differ in its logical and probabilistic structure from the following gambling example: A gambling casino has a (quantum) roulette wheel that outputs the numbers 0, 00, 1, ... , 36 with independent and identically distributed probabilities of exactly 1/38, but which can be overridden by the casino owner, who can make the numbers come out as he or she wills. Suppose the spins of the roulette wheel on a given day (April 1, 2003) produce great literary texts such that the numbers 1 to 26 correspond to the letters of the alphabet (1 --> A, 2 --> B, etc.) and the remaining numbers correspond to punctuation (e.g., 0 --> period, 00 --> comma, 27 --> space, etc.). Suppose that there is no and need be no direct evidence of the casino owner tampering with the roulette wheel (the casino owner can tamper away without anyone's direct knowledge, the only give-away being the sequences of rolls of the roulette wheel itself).
It seems that in both instances, a relevant consideration for precluding design is whether some unguided natural process could with reasonable probability have generated the output in question (living forms in one case; great literary texts in the other). What do I mean by "reasonable probability"? I mean events that one is quite likely to see at least once given the time and opportunity allotted (i.e., with probability considerably greater than 1/2).
But what if that's not the case? What if our best probabilistic analysis of the biological systems in question tells us that unguided natural processes could not have produced them with anything like reasonable probability? Could the design of life in that case become more probable than a Darwinian explanation (probabilities here being interpreted in a Bayesian or likelihood sense) simply in virtue of there being independent empirical evidence attesting to designers with the capacity to produce biological systems?
This, however, should raise a worry for Darwinists. The facts of biology, after all, have not changed. Yet design would be a better explanation if designers capable of, say, producing the animal phyla of the Cambrian could be attested through independent evidence. Note that there's no smoking gun here (no direct evidence of alien involvement in the fossil record, for instance). All we know is that beings with the power to generate life exist and could have acted. Would it help to know that the aliens really like building carbon-based life? But how could we know that? Do we simply take their word for it? The data of biology and natural history, we assume, stay as they are now.
But if design is a better explanation simply because of the independent attestation of technologically advanced space aliens, why should it not be a better explanation absent their independent attestation? If Darwinism is so poor an explanation that it would cave the instant space aliens capable of generating living forms in all their complexity could be independently attested, then why should it cease to be a poor explanation absent those space aliens? Again, the facts of biology themselves have not changed.
It's not clear, therefore, that independent evidence is doing any useful work here at all. If Sober is saying that independent evidence requires direct "eye-witness" evidence of a designer actually manipulating the designed object in question, then clearly he is ruling out design from biology on a priori grounds. But that would be inconsistent with what Sober claims in his text Philosophy of Biology, where he leaves open that design might be a viable explanation in biology. What's more, it raises the question of a double-standard inasmuch as natural selection and other naturalistic mechanisms that are reputed to be capable of accomplishing biology's design-work are just as unsupported by direct "eye-witness" evidence.
But once independent evidence for design needs merely to establish that there exists a designer with the causal power and opportunity to produce the effect in question (as in the alien thought experiment and in the gambling example), we have exactly the same set of data to explain that we did before we had this independent evidence. The requirement for independent evidence is therefore either vacuous (if independent evidence can be circumstantial) or prejudicial (if required to be direct). And in either case it obstructs inquiry into any actual design that might be embedded in the item under investigation.
[1] http://philosophy.wisc.edu/sober/DAF.PDF
[ 28. October 2002, 13:56: Message edited by: Moderator ]
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 23. October 2002 23:16
I believe that answer to Dembski's question is quite simple. Dembski suggests that intelligent design in nature becomes a better explanation if we can find independent evidence of intelligent designers that can design life forms and thus Demski wonders why should ID not be e better explanation absent such independent evidence.
There are several problems with this. First of all ID without independent supporting evidence is largely appeal to ignorance "Not X thus Y". There is no hypothesis generated which will help us understand if the ID hypothesis is a better one or not. As Sober has so excellently explained in another paper, one cannot determine if a certain observation is supportive or not of a certain hypothesis without knowing the other hypotheses.
So it seems quite clear that in absence of any independent evidence supporting ID, Darwinism has the upper hand in that it proposes hypotheses.
Now to the alternative hypothesis: Assume that Independent evidence of Intelligent Designers were found. Does this make a better case for Intelligent Design in nature? Not the mere existence of such creatures.
So I believe that the argument proposed by Dembski is NOT the argument suggested by Sober. In fact the claim that quote:
This, however, should raise a worry for Darwinists. The facts of biology, after all, have not changed. Yet design would be a better explanation if designers capable of, say, producing the animal phyla of the Cambrian could be attested through independent evidence.
But that is unconfirmed since Dembski still has to propose a hypothesis that can be tested side by side with natural hypotheses.
In fact, ironically, the 'argument' that
quote:
If Darwinism is so poor an explanation that it would cave the instant space aliens capable of generating living forms in all their complexity could be independently attested, then why should it cease to be a poor explanation absent those space aliens? Again, the facts of biology themselves have not changed.
Could be rephrased easily to read "If ID is so poor an explanation that it would cave the instant a natural pathway could be found, then why should it cease to be a poor explanation absent such evidence of pathways? Again, the facts of ID themselves have not changed.
The big difference of course is that Darwinism is not only based on the observed facts but also proposes ways to explain these facts. The existence or absence of intelligent designers has nothing to do with the likelihood of these hypotheses, they do not change. What we now have though is the possibility to define a plausible and supportable pathway for intelligent design. How plausible would such a pathway be? That depends on the claims. But one thing is sure, it will be far better than absence of any independent evidence. But independent evidence is not sufficient, we still need the alternative hypothesis.
Perhaps reading the Sober comment in context will help us realize what Sober had in mind when he made the statement quoted at the beginning of Dembski's article
quote:
I predict that it will eventually become clear that the organismic design argument
should never have been understood in this way. This is because I expect that human beings will
eventually build organisms from nonliving materials. This achievement will not close down the question of whether the organisms we observe were created by intelligent design or by mindless natural processes; on the contrary, it will give that question a practical meaning, since the organisms we will see around us will be of both kinds.26
After all the fact that we know that human 'designers' can use artificial selection to modify plants and animals, has hardly undermined the strength of the Darwinian mechanism.
See note 26 in Sober...
So in conclusion, the existence of advanced life forms may help us understand the existence of additional intelligent designers but it does not help us much yet in determining if they were involved in any meaningfull manner in the evolution of animals and plants. But our knowledge of their existence may be helpful in formulating a scientific hypothesis of ID that involves their methods. Perhaps a simple question 'Did you design life on earth´ would be sufficient to address the issue.
But then again, we are talking in pure hypotheticals here not testable hypotheses yet. But there is at least some hope that such a hypothesis may be forthcoming. And that is all Sober was trying to say imho.
I think this becomes quite clear when reading Sober's main text
quote:
The fact that there is no probabilistic modus tollens has great significance for understanding the design argument. The logic of this problem is essentially comparative. To evaluate the design
hypothesis, we must know what it predicts and compare this with the predictions made by other
hypotheses. The design hypothesis cannot win by default. The fact that an observation would be very improbable if it arose by chance is not enough to refute the chance hypothesis. One must show that the design hypothesis confers on the observation a higher probability, and even then the conclusion will merely be that the observation favors the design hypothesis, not that that hypothesis must be true.8
and
quote:
What they mean is that certain outcomes
are very improbable according to the evolutionary hypothesis. Even this more modest claim needs to
be scrutinized. However, if it were true, what would follow about the plausibility of creationism? In a word – nothing.
So is a hypothesis of a designer valuable? Only if we can restrict what a designer has in mind.
quote:
The problem is that the design hypothesis confers a probability on the observation only when it
is supplemented with further assumptions about what the designer’s goals and abilities would be if he
existed. Perhaps the designer would never build the vertebrate eye with features F1 .... Fn, either because he would lack the goals or because he would lack the ability. If so, the likelihood of the design hypothesis is zero. On the other hand, perhaps the designer would want above all to build the eye with features F1 ... Fn and would be entirely competent to bring this plan to fruition. If so, the likelihood of the design hypothesis is unity. There are as many likelihoods as there are suppositions concerning the goals and abilities of the putative designer. Which of these, or which class of these, should we take seriously?
So it is not independent evidence of the existence of a designer that Sober is arguing but
quote:
One needs independent evidence as to what the designer’s plans and abilities would be if he existed; one can’t obtain this evidence by assuming that the design hypothesis is true (Sober 1999).
The mere existence of designers does no make the hypothesis of ID more likely until we can understand the plans and abilities. Absent any independent evidence of designer the likelihood of such is zero, at least with an identifiable designer there is some chance that we can formulate a testable and useful ID hypothesis.
Sober also addresses the sometimes used criticism that SETI is looking for intelligent design. But what is SETI really looking for?
quote:
The SETI engineers search for this signal,
not because it is “complex” or fulfills some a priori criterion that would make it a “sign of intelligence,” but simply because they think they know what sorts of mechanisms are needed to produce it.12 This strategy may not work, but it is hard to see how the scientists could do any better. Our judgments about what counts as a sign of intelligent design must be based on empirical information about what designers often do and what they rarely do. As of now, these judgments are based on our knowledge of human intelligence. The more our hypotheses about intelligent designers depart from the human case, the more in the dark we are as to what the ground rules are for inferring intelligent design. It is
imaginable that these limitations will subside as human beings learn more about the cosmos. But for
now, we are rather limited.
[ 24. October 2002, 00:10: Message edited by: Frances ]
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 24. October 2002 00:29
In the OP Dembski asked quote:
But if design is a better explanation simply because of the independent attestation of technologically advanced space aliens, why should it not be a better explanation absent their independent attestation?
The short answer is that the aliens have not merely made a claim. They have identified the hypothesized designer(s) (themselves) and provided a mechanistic explanation for the actual creation of the cited events: quote:
"Moreover, they demonstrate to us that with this technology they can atom by atom and molecule by molecule assemble the most complex organisms."
That's precisely the sort of mechanistic explanation that has been ruled out by contemporary intelligent design theory. Current intelligent design theory actively resists conjectures about the identity of the designer(s) and the mechanics of its/their execution of the design(s).
The issue raised in the OP is not whether evolutionary theory is so "poor an explanation," but is rather given an identified designer with demonstrated technical means of instantiating designs in biological media and some evidence for the designer(s) having had the opportunity to execute the designs in the past, would design then be a viable alternative hypothesis?
The answer is, Sure! In the thought experiment, the "data" really include all the data of biology then extant and the identity of the hypothesized designers and the demonstrated ability of the aliens to produce the biological examples cited and some evidence (those stars) that they could have had the opportunity. Given the totality of data then available, design would be a viable hypothesis though it would by no means be established. Among other things, there are a whole lot of biological phenomena other than the examples cited still to be explained. Nevertheless, design would gain significant strength, not primarily because of the aliens' claim (anyone can make wild claims) but because of their demonstration of the mechanisms of creation, technological methods of accomplishing what they claim to have done. They had both means and opportunity.
If one adds Dembski's condition that someone has somehow estimated probabilities of occurrence of the cited biological examples as not being large, where those estimates are based on appropriate PDFs and on thorough knowledge of the biology in question, then the 'alien design' hypothesis would gain strength. As with any hypothesis it cannot be proved, but surely in the hypothetical situation described, it would be strong. Naturalistic evolutionary theory would have a powerful competitor given the aliens described in the thought experiment.
Now delete all knowledge of the aliens from the thought experiment. All we have left is some probabilities that are not "considerably greater than 1/2", and all the extant data of biology (and associated disciplines) that admit of naturalistic explanations along with some for which we don't yet have satisfactory naturalistic explanations. We have no evidence for design like that provided by the identities and technology of the hypothetical aliens.
Probabilities alone, however they are estimated or how small they might be, cannot distinguish between design and ignorance; they cannot discriminate between unknown designers and unknown natural causes. In this case, evolutionary theory has no competitor: no alternative has been offered that is analogous to the aliens of Dembski's thought experiment. The two cases (aliens versus no aliens) are not parallel; our knowledge states are very different in the two, hence our weighting and interpretation of the totality of the evidence is different.
RBH
Edited to set off a quotation.
[ 24. October 2002, 00:46: Message edited by: RBH ]
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 24. October 2002 00:59
A postscript:
As I think about it further, Dembski's thought experiment makes it crystal clear why hypotheses about the identity, nature, and technical capabilities of the intelligent designer(s) are absolutely necessary to any theory of the intelligent design of biological phenomena. Deleting all mention of the aliens and their properties and capabilities from the thought experiment highlights how very little current IDT offers by way of an explanatory alternative to evolutionary theory.
RBH
[ 24. October 2002, 01:01: Message edited by: RBH ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 24. October 2002 13:24
It is useful to note that the issue here is not ‘design by a designer’ but ‘design not explained by known materialistic processes’. Designs, no matter of what type or from what source are designs produced by ‘The Designer(s)’. The issue of what form The Designer takes is a question for theology not science.
IMO, there would be no scientific issue of ‘design by a designer’ if academic/scientific community did not impose artificial, arbitrary non-scientific limitations of ‘design by materialistic processes’. The academic/scientific community supports some variation on the claim ‘Some form of Darwinian evolutionary theory is compatible with, logically consistent with known facts regarding biological design produced by evolutionary process" . If, and IMO only if, you accept academic/scientific validation of Darwinian theory, then you can conclude that ‘evidence falsifying Darwin’ is evidence supporting design by an non-materialistic designer.
It is also useful to note that ID is based on two observed phenomena- 1)measurable, observable designs created by humans and 2)measurable, observable biological designs. IMO, if you wish to formally analyze this interesting observation you have two basic choices.
First, you can accept the current ‘scientific’ concepts/techniques/standards/principles of evolutionary biology. If you accept these standards then findings of ‘can’t be explained by Darwinian theory’ can be interpreted to mean ‘might be explained by non-materialistic processes’. There seems to be a reasonable argument that, accepting current standards, the bacterial flagellum can be interpreted as ‘might be explained by non-materialistic designers’.
But, IMO, if you accept current evolutionary biology standards, the flagellum is only the ‘tip of the iceberg’ of what can be interpreted as might be explained by non-materialistic designers. If you impose the standard, ‘must actually demonstrate that it can be modeled, simulated and explained’, then you will find a wide range of phenomena that can be interpreted as ‘explained by non-materialistic designers’.
The second option in formally analyzing the similarities between human and biological designs is to reject the current evolutionary biology concepts/techniques/standards/principles and essentially start with the basic concepts and principles of scientific analysis. In somewhat simplistic terms this involves:
1. Develop techniques for precisely defining occurrences of both biological design and human design.
2. Develop techniques for identifying and precisely defining occurrences of change in biological and human design.
3. Identify a set of mathematical/logical algorithms and processes which could model and simulate observed design processes.
4. Identify mathematical algorithms from set of possible algorithms which can used to formulate theory of observed design processes.
I mention these procedures in order to note that the ‘abandon evolutionary biology’ approach to design departs rather quickly and dramatically from the ‘build on evolutionary biology’ approach. Not only does the abandon EB approach avoid the controversial ‘explain by designer’ issue, but it suggests that 1)biological design and biological design processes are fundamentally different from what is traditionally defined as evolutionary results and evolutionary change processes and 2)the range of mathematical processes and algorithms capable of modeling and simulating design processes is much greater than the range of processes generally recognized in evolutionary biology.
SUMMARY
Depending on your perspective, ID analysis does or might suggest the work of an external designer. IMO, this finding either suggests 1)the designer decided to let us demonstrate his existence, or 2)there is some fundamental flaw in the logic or assumptions being used. Again IMO, it is more likely that ID analysis shows ‘there is a basic flaw in evolutionary biology’s concepts/techniques/standards’. IMO, if we start from scratch in analyzing the similarities between human and biological design, the flaws in the current approaches are fairly easy to spot.
IP: Logged
|
|
Jack Foster
Member
Member # 79
|
posted 24. October 2002 13:45
I think natural explanations might be better explanations for biological origins if you only consider the ability to investigate. Natural explanations, however, are not automatically better explanations when you consider everything that there is to consider.
Frances wrote:
quote:
So it is not independent evidence of the existence of a designer that Sober is arguing but
quote:
------------------------------------------------------------------------
One needs independent evidence as to what the designer’s plans and abilities would be if he existed; one can’t obtain this evidence by assuming that the design hypothesis is true (Sober 1999).
------------------------------------------------------------------------
The mere existence of designers does no make the hypothesis of ID more likely until we can understand the plans and abilities. Absent any independent evidence of designer the likelihood of such is zero, at least with an identifiable designer there is some chance that we can formulate a testable and useful ID hypothesis.
So independent evidence isn't enough; we need independent evidence as to what the designer's plans and abilities would be if he existed. We need lab notes and the technical understanding of the designer, or else we can't conclude (or even consider?) design.
Sober is asking for the impossible. If Louis Pasteur had a calculator beamed to him across time, he would be forced to conclude that its origins were of natural causation?
It's clear that there's a logical problem associated with this view. It ignores the fact that design is sometimes the better explanation.
[ 24. October 2002, 13:49: Message edited by: Jack Foster ]
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 24. October 2002 13:55
Jack:
How do we know that design is sometimes the better explanation and how does that help us when we do not know this?
Sober states that in order to know which hypothesis is better you need alternatives. In order to constrain what a designer could and could not do you need to know more about the designer.
That's how archaeology, forensics and even SETI work.
Sober also addresses this in his excellent paper
quote:
My claim is not that design theorists must have independent evidence that singles out a
specification of the exact goals and abilities of the putative intelligent designer. They may be uncertain as to which of the goal-ability pairs GA-1, GA-2, ..., GA-n is correct. However,
since
Pr(the eye has F1 ... Fn * Design) =
Pr(the eye has F1 ... Fn * Design & GA-i)Pr(GA-i*Design),
they do have to show that
Pr(the eye has F1 ... Fn * Design & GA-i)Pr(GA-i*Design) >
Pr(the eye has F1 ... Fn * Chance).
I think that Maynard Smith in his warehouse and archaeologists in their museums are able to do this.
They aren’t sure exactly what the intelligent designer was trying to achieve (e.g., they aren’t certain that
GA-1 is true and that all the other GA pairs are false), but they are able to see that it is not terribly
improbable that the object should have the features one observes if it were made by a human intelligent
designer. After all, the items in Maynard Smith’s warehouse were symmetrical and smooth metal
containers that had what appeared to be switches, dials, and gauges on them. And the “artefacts of
unknown function” in anthropology museums likewise bear signs of human handiwork.
Sober provides a much more detailed analysis in Intelligent Design and Probability Reasoning
Elliott Sober with Source
In this paper Sober points out some of the problems with the design inferences as proposed by Behe, Dembski and others.
quote:
Behe’s first sentence says that irreducible complexity cannot arise by Darwinian processes; however, the next two assert, more modestly, that irreducibly complex features are improbable on the Darwinian model and that they become more improbable the more complex they are. I hope it is clear from what I’ve said earlier why this shift is important. If evolutionary theory really did have the deductive consequence that organisms cannot have features that are irreducibly complex, then that theory would have to be false, if such features exist. But what if the theory merely entailed that irreducibly complex features are very improbable? Would the existence of such features show that the theory is improbable? Would it follow that the theory is disconfirmed by those observations? Would it follow that these features provide evidence in favor of intelligent design? The answers to all these questions are the same – no. There is no probabilistic analog of modus tollens.
Which raises the following criticism
quote:
In addition to rejecting evolutionary explanations, Behe advances the positive thesis that the biochemical systems he describes in loving detail “were designed by an intelligent agent” (p. 204). However, for these details to favor intelligent design over mindless evolution, we must know how probable those details are under each hypothesis. This is the point of the Law of Likelihood. Behe asserts that these details are very improbable according to evolutionary theory, but how probable are they according to the hypothesis of intelligent design? It is here that we encounter a great silence. Behe and other ID theorists spend a great deal of time criticizing evolutionary theory, but they don’t take even
the first steps towards formulating an alternative theory of their own that confers probabilities on what we observe.
Sober's analysis supports my argument that it is not enough to argue that the probability of evolution is small but one also has to show that the probability of the alternative is larger. Thus a hypothesis based on elimination can never be used to provide support of the hypothesis wrt other hypotheses. In fact, one cannot even eliminate alternative hypotheses unless one has a hypothesis of ID. See also the "evolution of complexity" thread in which I make my arguments for what I consider to be a better EF.
Thus when Dembski states
quote:
It's not clear, therefore, that independent evidence is doing any useful work here at all. If Sober is saying that independent evidence requires direct "eye-witness" evidence of a designer actually manipulating the designed object in question, then clearly he is ruling out design from biology on a priori grounds.
It seems clear to me that this is NOT what Sober is arguing. All he is arguing for is a viable ID hypothesis.
[ 24. October 2002, 15:54: Message edited by: Frances ]
IP: Logged
|
|
Jack Foster
Member
Member # 79
|
posted 24. October 2002 18:01
quote:
Sober's analysis supports my argument that it is not enough to argue that the probability of evolution is small but one also has to show that the probability of the alternative is larger.
But given law, design is the logical complement of chance. If you can show that it is impossible for life to have arisen by chance, for instance, what's left is design. In this instance, Sober's independent evidence for design as Bill Dembski puts it, does "no useful work" at all. It's extraneous to the analysis.
That's why I put the calculator in Pasteur's hands. Obviously, he has no independent evidence regarding the designer's plans and abilities. Still, via rational analysis, he should be able to determine that the calculator is not a product of natural law and chance, and therefore must be a product of design.
The same could be true for life; life could be the result of extremely advanced biotechnology, and as our own biotechnology improves we recognize how advanced life's technology is. Since it is so advanced, why should we have to speculate about the unknowable, before we are allowed to conclude design? Sober's analysis is illogical.
That being said, I agree that design should not be the default position. But if we were able to show that naturalistic abiogenesis (for example) is impossible, you and Sober seem to be saying that we still can't conclude design. This is sheer folly! What else are we going to conclude?!
[ 24. October 2002, 18:02: Message edited by: Jack Foster ]
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 24. October 2002 18:17
Jack:
Design may be the logical complement of chance (and regularity) but both chance, regularity and design are not single hypotheses. Furthermore one cannot eliminate a hypothesis merely by showing that it has low probability. Sober is not asking for independent evidence as much as a way to formulate one or more design hypotheses. Without any knowledge about designers how can one constrain the design hypothesis.
I think you need to differentiate between impossible or low possibility. They are not equivalent notions. And how are we going to determine that all naturalistic pathways are impossible? At most we can do is show that the ones we can think of are of low possibility but how does this help us infer design?
Life may very well be the result of extremely advanced biotechnology but where is the hypothesis? How would one want to test such a hypothesis?
Sober's analysis is quite logical although the inevitable conclusions may not be ready for acceptance.
To conclude: Unless one can show that something is impossible, not that it has low probability, how can one determine if the evidence is in support or non-support of a certain hypothesis?
In the above linked Intelligent Design and Probability Reasoning Sober shows how modus tollens arguments have no probabilistic analog. See for instance his analysis of the IC claims. If it can be shown that IC is logically impossible under Darwinian evolution then finding an IC system would likely falsify Darwinian evolution. But Behe does not argue that Darwinian pathways are impossible, merely that they are improbable.
Design is infered in reference to ones subjective knowledge which makes the inference so likely to be false positives. For instance the thunder or lightning by so many older cultures assigned to intelligent design (Gods)? Or circles of mushrooms assigned to fairies. Now we know better...
[ 24. October 2002, 18:55: Message edited by: Frances ]
IP: Logged
|
|
charlie d.
Member
Member # 159
|
posted 24. October 2002 18:34
quote:
That's why I put the calculator in Pasteur's hands. Obviously, he has no independent evidence regarding the designer's plans and abilities. Still, via rational analysis, he should be able to determine that the calculator is not a product of natural law and chance, and therefore must be a product of design.
Actually, that's because Pasteur's technology was in fact advanced enough for him to make that inference. He would have known numerals and mathematical notations, recognized keys, probably he would have had a sense of how the casing, screws, glass display could be constructed. (In other words, he would have based his inference on the hypothesis that it was some advanced product, from a human-like technology, built for mathematical purposes).
Give a calculator, with no info, to any member of any stone-age culture, and they'd just think it's a weird rock or animal (depending on whether it's off or on).
IP: Logged
|
|
Emyr Macdonald
Member
Member # 515
|
posted 25. October 2002 07:49
In reply to Frances:
If the SETI project detected a significant signal which passed the statistical criteria for a message, would we wait until we knew all about the nature and identity of the transmitting agent before inferring extra-terrestial intelligence?
I suspect that it would be all over the news networks immediately. It would be likely that science would never be capable of determining the identity of the intelligent transmitting agent - but it would not invalidate the inferrence of intelligence.
Emyr
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 25. October 2002 14:08
Emyr MacDonald wrote quote:
If the SETI project detected a significant signal which passed the statistical criteria for a message, would we wait until we knew all about the nature and identity of the transmitting agent before inferring extra-terrestial intelligence?
I suspect that it would be all over the news networks immediately. It would be likely that science would never be capable of determining the identity of the intelligent transmitting agent - but it would not invalidate the inferrence of intelligence.
Actually, should the SETI project detect an extraterrestrial artificial signal, much more than merely the fact of detection would immediately be known. The SETI detection and analysis technology does much more than merely 'detect.' The location of the source star would be known, and exploration of the implications of the stellar type for the nature of technology-capable life forms would begin at once. Frequency variations of the signal would be analyzed to infer the relative motion and orbital period of the source. The planet-finding technology that has already developed would be deployed in a search for a planetary system there. Natural radiation from the source would be analyzed at every possible wavelength, and other research that I, a non-astronomer, can't imagine would undoubtedly be focused on that source. Hypotheses would be offered and tested against the observable phenomena.
An array of properties of the signal itself would be available for analysis (signal strength, S/N ratio, signal coding method(s), and so on), and some information about the content would be known. Analyses of the properties and content of the signal would be focused on inferring what we can about the originators of the signal: What did they transmit? How did they transmit it? What must the nature of the transmitter be in order to produce the observed properties of the signal? What do the properties of the signal and the inferred transmitter and the contents of the signal tell us about the technical capabilities of the originators? What does the content of the signal imply about the knowledge and intentions of the originators? Is it purely mathematical, or is there other content that implies a wider interest in communicating beyond merely signalling the bare fact of their existence? All of those questions and more would immediately come under intense study. Again, hypotheses would be offered and would be tested against the observed phenomena. And it is the developing set of coherent empirical hypotheses about those questions that would form an explanation, not the signal itself.
The SETI methodology is not designer-neutral. The signal reception hardware and analytic software embody a number of assumptions about the (potential) signal producers. For example, they embody the assumption that potential signal producers will produce signals that have properties that human signals might have if we were transmitting the signal: SETI looks for what humans might do.
Similarly, the design detection technology embodied in the kind of probabilistic analyses that are suggested to detect evidence for a designer in biological phenomena embody assumptions about the designer(s). Most notably, that methodology embodies the critical assumption that the putative designer(s) did not act undetectably; that the designer(s) left evidence of interventions and therefore either could not conceal, or chose not to conceal, evidence of the biological design and production activities. That methodology makes a strong assumption about either the limits of the capabilities of the designer(s) or the intentions of the designer(s).
The question in the OP was quote:
But if design is a better explanation simply because of the independent attestation of technologically advanced space aliens, why should it not be a better explanation absent their independent attestation?
In the case of biology, as in SETI, the bare fact of detection would explain nothing. In the SETI case the bare fact of detection would not sit alone with no context. The detection of an artificial signal would immediately generate intense and prolonged interest in, and study of, the source and originators of the signal. In fact, the posing of research hypotheses began when the first detectors were deployed in the form of hardware and software decisions that were made then and later. And it is that research effort focused on the nature, properties, and identity of the signal generator(s) that would yield explanations of any signal that might be detected, not the isolated fact of its existence. An isolated fact explains nothing.
Similarly, the detection of purported design in nature, assuming it could be done, would explain nothing by itself. Rather, it would be a phenomenon to be explained, and that explanation would focus on the identity, nature, methods, and properties of the hypothesized designing agency. Detecting design in biology would not be an explanation, it would constitute that which is to be explained.
RBH
Edited for typos and one or two infelicities of expression.
[ 25. October 2002, 14:24: Message edited by: RBH ]
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 25. October 2002 14:20
Emyr asks if we detected a significant signal that passed the statistical claims for a message, would we need to know about the identity of designer?
Sober addresses the SETI example
quote:
The SETI engineers search for this signal,
not because it is “complex” or fulfills some a priori criterion that would make it a “sign of intelligence,” but simply because they think they know what sorts of mechanisms are needed to produce it.12 This strategy may not work, but it is hard to see how the scientists could do any better. Our judgments about what counts as a sign of intelligent design must be based on empirical information about what designers often do and what they rarely do. As of now, these judgments are based on our knowledge of human intelligence. The more our hypotheses about intelligent designers depart from the human case, the more in the dark we are as to what the ground rules are for inferring intelligent design. It is
imaginable that these limitations will subside as human beings learn more about the cosmos. But for
now, we are rather limited.
So while we may not know the identity of the intelligence we surely have made quite a few assumptions in order for SETI to be able to detect
IP: Logged
|
|
Jack Foster
Member
Member # 79
|
posted 25. October 2002 19:49
Hi Frances:
quote:
In the above linked Intelligent Design and Probability Reasoning Sober shows how modus tollens arguments have no probabilistic analog. See for instance his analysis of the IC claims. If it can be shown that IC is logically impossible under Darwinian evolution then finding an IC system would likely falsify Darwinian evolution. But Behe does not argue that Darwinian pathways are impossible, merely that they are improbable.
Behe argues that direct darwinian pathways are indeed impossible, and that indirect pathways are extremely improbable to the point that they become indistinguishable from impossible. I guess this would be the law of small probabilities at work.
I'm not a philosopher, so I really can't compete with Sober on that basis. But suppose I shout to you from the rooftop of a three-story building. When you ask how I got there, I respond that I jumped. Now I guess if you're Elliot Sober, you believe me. Perhaps a strong windstorm hit as I was jumping, or maybe the jolt of an earthquake accelerated me as I jumped. It's not altogether impossible that I successfully jumped three stories, given the right naturally occurring conditions. It's just so darned improbable that you doubt my story, even without providing an alternate hypothesis for yourself. (Or you should.) You would probably ask "how did you really get up on that roof, Jack?"
The same should be true for origins. If we conclude that natural selection is simply not a sufficient force to accomplish life, then design should be implicated. This seems clear to me. As likelihood of chance is reduced to nil, likelihood of alternatives increase to 1.
This really must be the case. If we assume for a moment that intelligent intervention is an absolute requirement for life (and that there is no Soberian independent evidence) then the only way to reach the correct inference is via this logic. Sober's logic would prevent us from ever being correct. He would forever cling to the mirage of a sliver of probability. He would forever be wrong.
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 25. October 2002 20:20
Behe does not really make an argument for indirect routes other than an off hand suggestion that such routes would be improbable.
But improbable does not necessarily make the impossible. After all probability is never a reliable indicator of if something could happen or not. And more importantly it does not provide us with any reference to the probability of alternative hypotheses.
I do not understand the relevance of your story wrt Sober's arguments. He may be a philosopher but his arguments are hardly philosophical in nature.
I could and would not argue what Sober would say wrt your example but I can provide you with my feedback. I wonder what you consider the relevant hypotheses to be. 1) The person is lying about jumping that high 2) the person did jump that high ? Why should the statement by the person help us determine the likelihood of his story? All the alternatives seem to be intelligent design.
You state then in an attempt at analogy, the same should be true for origins. The same what? What evidence do we have for intelligent design in nature? Why should us believing that if natural selection is not a probable force to accomplish life, ignoring for the moment that we seem to be arguing about evolution here, not accomplishing life, there should be evidence for intelligent design? Chance and design are not complimentary in the sense as you suggested, just because many chance hypotheses become improbable does not make the probability of ID increase, why should it? Nor is there the simplicity of a single chance versus a single ID hypothesis, what ID hypothesis should we consider? Front loading versus intervention? But how different if front loading from natural mechanisms? What would, even if we were to accept your claims, believe that the low probability of chance means high probability of intelligent design?
Now lets assumte that ID must be required and we have no Soberian evidence, then we are in a bind since we cannot really propose any hypothesis to compete with chance hypotheses. At most we can conclude 'we don't know', why would one expect that we have to conclude Intelligent Design? I might as well reverse the question. What if a certain hypothesis would have a very large probability for life to arise but we are unaware of such hypothesis, why should we automatically infer design just because of our ignorance? At most we can hope to find the best hypothesis with our given knowledge. Science does not guarantee omniscience. We can only deal with the information we have. Such is life but the argument you propose is not limited to Sober but equally well to a design inference.
Lets step through Sober's arguments which are quite straightforward:
quote:
This paper defends two theses about probabilistic reasoning. First, although modus
ponens has a probabilistic analog, modus tollens does not – the fact that a hypothesis says that
an observation is very improbable does not entail that the hypothesis is improbable. Second,
the evidence relation is essentially comparative; with respect to hypotheses that confer
probabilities on observation statements but do not entail them, an observation O may favor one
hypothesis H1 over another hypothesis H2 , but O cannot be said to confirm or disconfirm H1
without such relativization.
The first step is to realize that modus tollens, the argument X then Y, not X then not Y has not probabilistic equivalent unlike modus ponens arguments if X then Y, X thus Y
Thus an observation which seems improbable with respect to a certain hypothesis is not sufficient to make the hypothesis improbable.
Now the next step
quote:
The idea that evidence is essentially a comparative concept is often associated with the Law of Likelihood:
Evidence E favors hypothesis H1 over hypothesis H2 if and only if
Pr(E * H1) > Pr(E * H2).
The absolute values of the probabilities do not matter.
So when Behe makes his argument wrt IC he seems to jump from deductive to a probabilistic modus tollens.
quote:
Behe’s first sentence says that irreducible complexity cannot arise by Darwinian processes; however, the next two assert, more modestly, that irreducibly complex features are improbable on the Darwinian model and that they become more improbable the more complex they are. I hope it is clear from what I’ve said earlier why this shift is important. If evolutionary theory really did have the deductive consequence that organisms cannot have features that are irreducibly complex, then that theory would have to be false, if such features exist. But what if the theory merely entailed that irreducibly complex features are very improbable? Would the existence of such features show that the theory is improbable? Would it follow that the theory is disconfirmed by those observations? Would it follow that these features provide evidence in favor of intelligent design? The answers to all these questions are the same – no. There is no probabilistic analog of modus tollens.
The example which clearly explains this is the story about the two urns (Royall)
Another way of looking at design versus chance is to look at the Bayes theorem
P(H | O) = P (O | H) P(H) /P(O)
Probability of hypothesis H given observation I is related to the probability of the observation given the hypothesis multiplied by the probability of the hypothesis and divided by the probability of the observation.
P(H)/P(O) prior probability
P(H|O) posterior probability
quote:
Which hypothesis has the higher posterior probability depends on how their likelihoods are related, but also on how their prior probabilities are related. This explains why the likelihood version of the design argument does not show that Design is more probable than Chance. To draw this further conclusion, we’d have to say something about the prior probabilities of the two hypotheses.
Source
So the question now is, if it can be shown that P(O|H) is low should one conclude that P(H) is low?
I will unlikely respond today due to the 3 posting limit.
To add to the confusion though let me point out that the likelihood approach allows us to compare the various hypotheses being test but it does not help us establish which one is likely to be the correct one.
Jack
quote:
This is correct, and these two hypotheses are exhaustive. If we conclude that it is extremely improbable to the point of impossibility that the person could have jumped that high, then we conclude that the person is lying about jumping that high. Indeed, that is the logical conclusion.
The same is true about origins (for instance). If we conclude that the chance origin of life is extremely improbable to the point of impossibility, then we also can conclude design. Chance and design, together, . . . are exhaustive. Sober doesn't seem to recognise this.
You are confusing chance and design as logical opposites, and probabilistic opposites. First of all the hypotheses need not be exhaustive. The person may believe he jumped that high when in fact it was a drug induced situation. Furthermore there is no chance hypothesis here, merely two different intelligent design hypotheses. How is that going to help us eliminate chance? In fact in your above example you merely eliminated one of two hypotheses, exactly what Sober had in mind.
You cannot extend the analogy to design for which we have no background knowledge of likelihoods.
Perhaps it would be interesting to apply the EF to some cases and determine what its conclusion would have been and how it would have been reached? So far no real scenarios have been proposed to test the claimed efficiency of the filter.
Perhaps in working through some real life examples we can determine what the limitations are?
For one, the probabilities for known and unknown natural /chance mechanisms seems to be quite a problem for all but the most simplistic biological system.
[ 25. October 2002, 21:28: Message edited by: Frances ]
IP: Logged
|
|
|
Printer-friendly view of this topic