|
Author
|
Topic: Exhaustive Sweep of Chance Hypotheses
|
warren_bergerson
Member
Member # 262
|
posted 23. June 2002 08:08
Frances,
To some extent your comment illustrate the issues I was raising. For example your comment- "The argument that specified complexity cannot be explained by known processes does not seem to reflect Dembski's argument. Dembski argues that it is intelligence which can generate CSI." Combines/confuses the issue of ‘we don’t know’ and ‘speculation regarding what we may someday know’. Demonstrating that ‘chance plus known processes’ can’t explain complexity is separate from ‘speculation that complexity is produced by intelligence’. Failure to distinguish the two arguments leads to confusion, IMO.
The distinction between ‘what is known and scientifically verifiable’ and ‘speculation about what is not known’ can be usefully applied to the question of design. It is accurate to say that any reference to the cause of design are ‘speculation about what is not known’. Design the fact, or the existence of design is, however, ‘known and scientifically verifiable’.
It is possible to construct a verifiable operational definition of design which can clearly and unambiguously identify a whole range of biological phenomena which qualify as design. The existence of design, like the existence of evolutionary change, learning, punctuated equilibrium and all sorts of scientific concepts are verifiable using operational definitions. Such operational definitions may not be adequate to construct hard science theories, but they are sufficiently precise to verify the existence of the phenomena.
Your comments on ‘scientifically verifiable movement of the planets’ provide a useful basis for explaining the relationship between the concepts of ‘design’, ‘designer’, and ‘science’. Very early in human history it was recognized that the stars and planets moved in a predictable, non-random, ‘designed’ fashion. This led to speculation(someone must have made this speculation) that these movements were controlled by some type of designer. The discovery of verifiable scientific theories describing/explaining the movement of planets did not contradict the observation that planets moved in a systematic, designed fashion, nor did the discovery contradict the assertion that this ‘designed movement was controlled by a designer’. The discovery of verifiable scientific ‘laws’ governing the movements of planets simply moved the speculated source of design and the actions of a designer back to the beginning of the universe.
In the physical sciences, it is appropriate to assume that the role of design and designer are associated with the beginning of time. In the life sciences, it is not entirely clear if design and the speculated actions of an external designer occurred at the beginning of the universe, or at some more recent point in time. There is no a priori reason in either logic or science eliminating the possible actions of an external designer, nor is there any a priori reason that such ‘speculated actions’ must have occurred at the beginning of the universe. The rule of science, the rule of real hard science, is that if there is an explicitly formulated, verifiable, and non-falsified scientific theory describing/explaining a causal relationship, then that explanation/causal relationship takes precedence over alternative speculative explanations/descriptions.
Contrary to the peer-review practices in evolutionary biology today, science contains no a priori rules prohibiting or giving precedence to any class of speculative assertion. Science, contrary to current practice, real science does not permit granting the title of ‘valid scientific theory’ to an implicit collection of beliefs which do not meet scientific standards.
IMO, the ‘unsound corruption of professional standards’ is a simple statement of readily verifiable fact. Most obviously, professional standards in evolutionary biology do not require explicit formulations of scientific theories. Professional standards permit treating non-explicitly formulated theories as valid theories. Professional standards permit, and in some instances require, the slanting or biasing of experimental results.
The unsound corruption of professional standards is a phenomena which occurs with rather amazing regularity in essentially all areas of human endeavor. Despite impassioned pleas of ‘we would never do that’, corruption of professional standards is a widespread, predictable, and rather easily identified phenomena.
IMO, corruption of professional standards is a technical rather than moral issue. If professional standards do not permit fair treatment of an issue, then the questions to be addressed are 1)How can the corrupted standards be changed?, an/or 2)How can the corrupted standards be circumvented?
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 23. June 2002 15:12
Warren,
You raise some interesting issues. It seems to me that your views are in disagreement with the ID inference ala Dembski. When you claim that excluding natural law and chance is separate from infering design. In the design inference ala Dembski, ID is infered through elimination of chance and regularity. As you seem to agree, the problem with this is that incomplete knowledge can lead to assigning intelligence to phenomenon which are explainable by natural law. You claim that once such laws are identified the designer is moved to the beginning of times. The latter conclusion is however not obviously and certainly makes the formulation of a scientific theory of design problematic since a designer seems to be subject to the 'moving goalpost' approach. In fact, this seems to be very similar to the "God of the gaps" fallacy.
Then you continue to make an interesting statement that "in the physical sciences, it is appropriate to assume that the role of design and designer are associated with the beginning of time". I find this to be a fascinating assertion which I would love to see supported with references to the relevant literature.
You also claim "that science does not contain a priori rules prohibiting or giving precedence to any class of speculative assertions". I would argue that science indeed does prohibit certain classes of speculative assertions within its realm of scientific inquiry.
You then continue to make accusations of "unsound corruption of professional standards" as if it were a simple statement of readily verifiable fact. May I invite you to show us that your premise is correct and ask you to show some of these 'readily verifiable facts'? Especially the claim about "requiring the slanting or biasing of experimental results".
Until then I propose that we are a bit more careful in making such accusations since they do not further the conversation.
I am also still curious to hear why you claimed in the previous posting that "you can't eliminate the current theory as a valid explanation unless all possible material mechanisms have been eliminated" since this seems to go contrary to scientific standards. I suggested that you may be confusing the ID approach which does rely on elimination with the reigning scientific paradigms. You refer to this as an "unsound corruption of professional standards" and yet it seems that this 'corruption' is a fallacy in that it does not seem to describe scientific inquiry accurately. Again I certainly invite you to show us references to the literature that your assertion about the elimination of theories is correct.
This would imho surely fascilitate a discussion on the merrits of the arguments based on relevance to the actual situation.
I would certainly be interested in exploring with you the ide that "specified complexity or the sweep of chance hypothesis shows that observed complexity cannot be explained by known processes and /or chance". In fact I would argue that this is not obvious and in fact I would argue that there are experiments and data suggesting the opposite.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 24. June 2002 06:13
Frances,
I believe I said that excluding ‘known processes and chance’ was different than ‘defining design’. ‘known process and chance’ is an expression referring to an existing scientific theory or ‘testable/verifiable expressions of causal relationships or natural laws’. This is not the same as natural laws. I don’t think that noting the importance of the distinction between ‘excluding or falsifying an existing theory’ and ‘defining a concept’ is inconsistent with Dembski’s concept of ID. My earlier comments were to points out that some confusion might be eliminated if components of Dembski’s argument were considered separately. IMO, separating the components strengthens, not weakens Dembski’s argument.
IMO, ‘what are real scientific standards’ and ‘corruption of professional standards’ are both interesting and legitimate topics for discussion, but beyond the scope of this thread.
QUOTE: I would certainly be interested in exploring with you the ide that "specified complexity or the sweep of chance hypothesis shows that observed complexity cannot be explained by known processes and /or chance".
IMO, a claim that ‘complexity can be explained by known processes and chance’ requires the development and presentation of an explicit testable, reviewable model which can show how the observed complexity realistically could have evolved. To my knowledge, no one has proposed/presented such a model. In the absence of a verifiable counter argument, Dembski has shown the ‘observed complexity cannot be explained’. Speculation about what someone may someday be able to explain is a separate topic.
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 24. June 2002 12:26
Your argument that "In absence of a verifiable counter, Dembski has shown" suggests that Dembski has proposed a tentative "4th law" . In fact in NFL Dembski addresses several 'verifiable counter arguments' but claims, largely based on the NFL theorem that algorithms cannot generate complexity. As far as the validity of this argument is concerned, it is important to recognize that Toussaint et al have shown that the NFL theorem likely does not apply to RM&NS.
And I disagree that speculation what someone, someday may do are a separate topic. This topic is about an exhaustive sweep of chance hypotheses, which suggests that there will not be such future discoveries, or otherwise we need to accept that the ID inference will suffer from false positives. But if we accept this, then we also need to accept that the ID inference requires not only eliminative evidence but also positive evidence.
Tom Schneider, Evolution of biological information, Nucleic Acids Res, 28(14), 2794-2799, 2000
C. Igel, M. Toussaint (2001): On Classes of Functions for which No Free Lunch Results Hold. Submitted to IEEE Transactions on Evolutionary Computation.
quote:
In a recent paper it was shown that No Free Lunch results hold for any subset F of the set of all possible functions from a finite set X to a finite set Y iff F is closed under permutation of X. In this article, we prove that the number of those subsets can be neglected compared to the overall number of possible subsets. Further, we present some arguments why problem classes relevant in practice are not likely to be closed under permutation.
Adami, Evolution of Biological Complexity
Proc. Nat. Acad. Sci. 97 (2000) 4463-4468
quote:
In order to make a case for or against a trend in the evolution of complexity in biological evolution, complexity needs to be both rigorously defined and measurable. A recent information-theoretic (but intuitively evident) definition identifies genomic complexity with the amount of information a sequence stores about its environment. We investigate the evolution of genomic complexity in populations of digital organisms and monitor in detail the evolutionary transitions that increase complexity. We show that because natural selection forces genomes to behave as a natural ``Maxwell Demon'', within a fixed environment genomic complexity is forced to increase.
Machine learns tones
Explorations in Design Space: Unconventional electronics design through artificial evolution (FINAL DRAFT)
Adrian Thompson, Paul Layzell, Ricardo Salem Zebulum
Creatures from primordial silicon
Dembski suggests more examples. Wesley Elsberry has suggested the problem of the traveling salesman as an example of algorithms generating complexity.
[ 24 June 2002, 12:46: Message edited by: Frances ]
IP: Logged
|
|
complex
Member
Member # 259
|
posted 24. June 2002 23:07
Bill,
Sorry for this late reply.
It seems to me the more concrete the cases in which we examine probability, the more clearly it probability calculated and the more easily we can eliminate a chance hypothesis.
So for example, if all I have is a set of cards, I can quite clearly figure out probabilities, even on the assumption of an even probablity distribution for each card.
The more I get away from these concrete examples, the more difficult it becomes to be convincing that the probablities that I am calculating are sufficient epistemological grounds for eliminating explanations.
For an example in this area, what is the probability of the formation of a black hole? Given everything we know, I could probably come up with some "reasonable" number. But how would this number have looked 10 years ago, or 20 years ago, or 50 years from today? Especially given our understanding of sub-atomic particles, their interaction and the fundamental forces and matter at action in the universe.
Similarly, your example of a copy of data is distinct and easy to see. Any process that takes a different direction with a different environment and different conditions is expected to have so many variables in difference with the original replicated data, that a replication is clearly out of the question by random means. We don't even have to calculate probabilities given the fact that we assume that all of the variables are radically different (both by design and environment).
However, the statement which you are trying to refute (by elimnating chance hypotheses) consists of the statement that "the laws of nature are sufficient grounds for a medium to low probability generation of life on earth". Now by "hyothesis elimination" you are covering the fact that you are question begging of the orignal claim. You are saying that probabilities are low for a particular natural phenomena A and then for another B and then for another C and so on and go on to eliminate them. The question begging is that you have not fought your argument in the probability calculation domain where those with the probability calculator come our against your probability calculator and the numbers come out in force.
In such an instance you would have to be convincing why your probability calculations for each known natural phenomena and all of their correlations and all known and unknown self-organizing principles are of such low probability against those that claim that they are medium-low probabability.
It is the probability calculations that are crucial to your argument, not the hypothesis elimination given the probabilities.
And given your knowledge distance (and that of those who claim different probabilities) from the mechanism at play in the natural world, it is difficult to take these probability arguments too seriously unless the probabilities are well mapped out to some computer simulation or mathematical equation. The natural world is simply to complex to be able to come up with reasonable probability numbers that don't have a huge error bound.
best regards,
IP: Logged
|
|
|
Printer-friendly view of this topic