|
Author
|
Topic: Elliott Sober's Independent Evidence Requirement for Design
|
Mark Szlazak
Member
Member # 391
|
posted 28. October 2002 14:39
Warren, so your will never let in the possibility of indeterminism? Well, people in sciences already are, and those that try to keep it out are setting up a dogma that makes determinism UNFALSIFIABLE.
Also it's interesting to note that the book, "The Open Universe: An Argument for Indeterminism" was written by none other than Karl Raimund Popper.
[ 28. October 2002, 20:47: Message edited by: Mark Szlazak ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 29. October 2002 08:37
Mark,
Quote: Warren, so your will never let in the possibility of indeterminism? Well, people in sciences already are, and those that try to keep it out are setting up a dogma that makes determinism UNFALSIFIABLE.
My position on ‘scientific determinism’ is fairly straight forward. 1) Scientific analysis or the scientific paradigm is most productive if theories are formulated as models of materialistic deterministic causal relationships which are testable and subject to single test falsification and 2) it is both possible and practical to express scientific theories in terms of such deterministic causal relationships.
I am not eliminating the possibility of indeterminism or non-materialistic determinism. I am simply saying that 1)science is most productive if it sticks to traditional deterministic causal relationships, and 2)there is no practical reason why needs to depart from analysis of deterministic causal relationships. This is the traditional or conventional view of scientific determinism, at least IMO.
Proposals to depart from the traditional view arise because of the technical problems associated with formulating deterministic theories or models of complex causal relationships. Because scientists couldn’t solve this technical problem, and because obviously you can’t be a real science if you don’t have theories, rationalizations were developed to justify classifying non-rigorous, non-deterministic models as scientific theories.
I am not an expert on the subject, but I have heard Popper described as an apologist for the practices of the non-rigorous ‘soft’ sciences.
IMO, the technical problem associated with formulating deterministic hard science theories of complex causal relationships is solvable. I have, in fact, developed techniques which appear to solve the problem, at least with respect to the complex ‘design generating’, ‘information generating’ processes.
The subject of the discussion here procedures for determining if ‘neo-Darwinian chance’ or ‘design by a designer’ provides a better hypotheses for explaining the origin of life or the origin of biological designs. The point I am arguing is that there are other possible materialistic/deterministic explanations of the origin of life. Specifically I suggest there are information generating logic machines using forms of selection other than Darwinian Natural Selection, and techniques for generating variations other than random variation which provide materialistic, deterministic models or hypotheses of the origin of life.
Not only do I claim there are options other than ‘chance’ and ‘design by a designer’, but I claim that the set of possible materialistic, deterministic explanations is so large and diverse that scientific analysis of life forms can always find members of this set which will fit any set of observed data. I am not rejecting the possibility of non-deterministic or non-materialistic models and explanations. I am simply stating that 1)science would not normally consider such options until the range of deterministic, materialistic explanations has been exhausted, and this can never happen.
IP: Logged
|
|
Mark Szlazak
Member
Member # 391
|
posted 29. October 2002 10:58
Hey Warren, all you've said is an article of personal faith. That's fine but in science it should be a question about the point at which enough evidence warrants a change of believe. If there is no point then the belief, in the Popperian spirit, is unfalsifiable.
quote:
My position on ‘scientific determinism’ is fairly straight forward. 1) Scientific analysis or the scientific paradigm is most productive if theories are formulated as models of materialistic deterministic causal relationships which are testable and subject to single test falsification and 2) it is both possible and practical to express scientific theories in terms of such deterministic causal relationships.
Well, the history of physics from the nearly the beginning of the twentieth century on says just the opposite. Of course you know what that can mean ... simply that materialism is just an epiphenomena or (usefull) illusion. Tough cheese for some.
quote:
I am not eliminating the possibility of indeterminism or non-materialistic determinism. I am simply saying that 1)science is most productive if it sticks to traditional deterministic causal relationships, and 2)there is no practical reason why needs to depart from analysis of deterministic causal relationships. This is the traditional or conventional view of scientific determinism, at least IMO.
Actually, this isn't the case either, since we find statistical explanations throughout the sciences all the time. There maybe a belief that later on the right type of factors maybe found to transform these into deductive-type explanations, but these aren't needed before these theories can do any usefull work.
This is in evidence everywhere, from the basic to applied sciences and is uncontraversial and fully recognized even in the philosophy of science. See "Four Decades of Scientific Explanation" by Wesley Salmon in "Scientific Explanation" ed. Philip Kitcher and Wesley C. Salmon.
As an aside, here's a recently released book that tells about scientific investigations using anything but purely materialistic, strictly "objective" scientific explanations or measures to treat patients with obsessive complusive disorders. Doing a better job with these patients than the strictly "objective" paradigm of "behaviourism" and to some degree "cognitivism."
The reason for better performance was that the "subjective" elements were taken seriously ... in ways much more than the materialists paradigm would ever wish they were. "The Mind And The Brain" by Jeffery Schwartz and Sharon Begley.
As another aside, Sober asks for independent evidence when dealing with 1) materialistic explanations and 2) Non-materialitic design explanations. If materialism is just as I've said an epiphenomena or illusion, and I take that to be the case for various reasons from the logical impossibility of implying or explaining consciousness to its apparent failure in modern physics, then we real don't have good independent evidence for 1). It's sort of unreal.
Isn't Sober then asking us to compare explanations based on design, and design is something people have done since the "dawn of time", with a hypothesis akin to saying "leprechauns did it." Besides maybe being legitimate in some small towns in Ireland, should we be really comparing design explanations with the hypothesis that say the effects are to be explained as a result of the mischievous manifestations of leprechauns?
[ 29. October 2002, 20:32: Message edited by: Mark Szlazak ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 30. October 2002 08:22
Mark,
Quote WB: I am not eliminating the possibility of indeterminism or non-materialistic determinism. I am simply saying that 1)science is most productive if it sticks to traditional deterministic causal relationships, and 2)there is no practical reason why needs to depart from analysis of deterministic causal relationships.
Quote: Actually, this isn't the case either, since we find statistical explanations throughout the sciences all the time. There maybe a belief that later on the right type of factors maybe found to transform these into deductive-type explanations, but these aren't needed before these theories can do any usefull work.
You should not there is a difference between a phenomena predictable ‘deterministic’ probability distribution and an indeterminate ‘we have no idea what will happen’ relationship. You can construct useful deterministic hypothesis involving statistical fluctuations. As far as I am aware, you can not construct useful scientific hypothesis from ‘we have no idea what will happen’ relationships. Lots of really bad science in fields like economics is based on the failure to differentiate between deterministic statistical relationships and ‘we don’t know but this is what happened in the last 5 years’ relationships.
It is obviously true that statistical theories and statistical analysis have become standard practice in many fields of scientific endeavor. Whether such analysis leads to real progress in understanding the underlying relationships is not at all clear. The saying about ‘lies, D__ lies, and statistics’ has a great deal of validity. Anyone with a real working knowledge of statistics, (and that seems to be a rather small group of individuals) can quite easily manipulate statistical analysis to produce a ‘desired result’. This manipulation can be achieved, in most instances, without engaging in what would be classified as inappropriate manipulation.
There are legitimate and useful applications for statistical analysis and statistical theories. However, as used today in academia, there do not exist sufficient controls to prevent serious abuses.
Quote: Isn't Sober then asking us to compare explanations based on design, and design is something people have done since the "dawn of time", with a hypothesis akin to saying "leprechauns did it." Besides maybe being legitimate in some small towns in Ireland, should we be really comparing design explanations with the hypothesis that say the effects are to be explained as a result of the mischievous manifestations of leprechauns?
Design, as you point out is a very real phenomena. Humans have been ‘performing design processes’ and ‘creating design’ as long as human have been human. (probably sometime after the dawn of time). Why then would anyone suggest that ‘design processes’ are not ‘materialistic processes’? Why based on the evidence wouldn’t we conclude that ‘the human design producing process appears to be a materialistic process produced by a materialistic phenomena’. Recognizing that designs by humans are the result of a materialistic process, would it not then be reasonable to speculate that similar designs produced by non-human biological systems are the result of a logically similar materialistic processes.
IMO, the concept that there are ‘materialistic human design processes’ which are likely to be similar to ‘materialistic genetic design processes’ would be widely accepted in the scientific community if it were not for Darwinian theories of evolution. Believing that Darwin can explain biological/genetic evolution, suggests that some similar processes explain human creative intelligence. Work with EA’s clearly shows that neo-Darwinian RM&NS type processes can not explain human creative intelligence.
Rather than accepting the logical conclusion that ‘there is likely some as yet unknown materialistic processes which will explain both human and genomic design processes’, we have the academic community accepting the conclusion ‘Since RM&NS/EA processes are the only possible materialistic explanations of evolution, they are the only possible materialistic explanations of human design processes. It therefore follows logically that both human design and genetic design may involve non-materialistic design processes’.
IP: Logged
|
|
Mark Szlazak
Member
Member # 391
|
posted 30. October 2002 11:02
quote:
You should not there is a difference between a phenomena predictable ‘deterministic’ probability distribution and an indeterminate ‘we have no idea what will happen’ relationship. You can construct useful deterministic hypothesis involving statistical fluctuations. As far as I am aware, you can not construct useful scientific hypothesis from ‘we have no idea what will happen’ relationships. Lots of really bad science in fields like economics is based on the failure to differentiate between deterministic statistical relationships and ‘we don’t know but this is what happened in the last 5 years’ relationships.
So you consider Quantum Theory really bad science? Strange, since it's probably the most powerful, useful and successful theory ever developed in physics. It's also considered to be indeterministic.
INTERESTING ASIDE: Within the determinists framework, high probabilities are useful because they increase confidence in being on the right track of a deterministic basis. However, if there is no deterministic basis to be found, then in this regard, probability values are irrelevant. Isn't indeterminism fun! ![[Big Grin]](biggrin.gif)
quote:
Design, as you point out is a very real phenomena. Humans have been ‘performing design processes’ and ‘creating design’ as long as human have been human. (probably sometime after the dawn of time). Why then would anyone suggest that ‘design processes’ are not ‘materialistic processes’? Why based on the evidence wouldn’t we conclude that ‘the human design producing process appears to be a materialistic process produced by a materialistic phenomena’. Recognizing that designs by humans are the result of a materialistic process, would it not then be reasonable to speculate that similar designs produced by non-human biological systems are the result of a logically similar materialistic processes.
IMO, the concept that there are ‘materialistic human design processes’ which are likely to be similar to ‘materialistic genetic design processes’ would be widely accepted in the scientific community if it were not for Darwinian theories of evolution. Believing that Darwin can explain biological/genetic evolution, suggests that some similar processes explain human creative intelligence. Work with EA’s clearly shows that neo-Darwinian RM&NS type processes can not explain human creative intelligence.
Rather than accepting the logical conclusion that ‘there is likely some as yet unknown materialistic processes which will explain both human and genomic design processes’, we have the academic community accepting the conclusion ‘Since RM&NS/EA processes are the only possible materialistic explanations of evolution, they are the only possible materialistic explanations of human design processes. It therefore follows logically that both human design and genetic design may involve non-materialistic design processes’.
I said it before and I'll say again. Materialism is at LEAST empirically false and if taken as an overall view which "isms" typically are then it's also logically false. In other words a fantasy, kind of like leprechauns, expect a more useful one than these little folk.
I'll close with words of wisdom that may sort of characterize our situation.
"The Tao that can be spoken of is not the true Tao" -- Lao Tzu
[ 02. November 2002, 18:24: Message edited by: Mark Szlazak ]
IP: Logged
|
|
Jack
Member
Member # 265
|
posted 30. October 2002 11:08
Frances>> let me also try to return to the original purpose of this thread namely the suggestion by Sober that we need ID hypotheses in order to determine the validity of the claims.<<
Can you provide an example of what you would consider an ID hypothesis?
Frances>>Behe's argument however is not an argument for ID as much as an argument against Darwinian pathways. One should not confuse low probability of such pathways with evidence FOR ID.<<
What would you consider evidence for ID?
Frances>>Dawkins may not like pure chance alone as an explanation but in order for ID to be infered ID has also to deal with such hypotheses.<<
Are you saying that in order for one to infer ID was involved in the origin of some biological thing that one must prove that the biological thing couldn't have originated through chance?
Frances>>Does not present much support for the ID scenario either but merely points out our ignorance. But should our ignorance perse be evidence of support of ID?<<
Mike Gene says:
"No, ID is not an argument from ignorance. It would be if I thought I had proved the flagellum was designed. But I am not trying to prove things about the world. I am engaged in an investigation that is looking for the best explanation. One looks for clues and features that cohere to an explanation. Now, I have not concluded the flagellum was designed; I have tentatively inferred design for the origin of the flagellum."
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 30. October 2002 12:58
Jack wrote quote:
Are you saying that in order for one to infer ID was involved in the origin of some biological thing that one must prove that the biological thing couldn't have originated through chance?
Um. Correct me if I'm wrong, but I thought that was in fact the whole object of the Explanatory Filter and of the notion of irreducible complexity: to show that at least some biological structures (e.g., the E. coli flagellum) or biological processes (e.g., the blood clotting cascade) could not have occurred by either chance (coincidence) or via natural processes governed by known or knowable natural laws. Every ID inference I've seen that invokes CSI, the EF, and/or irreducible complexity rests on some form of the argument that 'it couldn't have occurred without the intervention of an intelligent designer or agency,' meaning it couldn't have occurred either through natural processes operating according to natural laws or by random chance in the absence of natural causal operators. That is the logic of the EF.
RBH
IP: Logged
|
|
Jack
Member
Member # 265
|
posted 31. October 2002 11:57
RBH>>Every ID inference I've seen that invokes CSI, the EF, and/or irreducible complexity rests on some form of the argument that 'it couldn't have occurred without the intervention of an intelligent designer or agency,' meaning it couldn't have occurred either through natural processes operating according to natural laws or by random chance in the absence of natural causal operators. That is the logic of the EF.<<
I disagree. The most the EF can do is show that something is highly improbable. It can't prove that something is impossible. ID critics seem to think that a blind watchmaking inference will always trump an ID inference unless ID advocates can prove blind watchmaking is impossible. I think that's a very weak position. How does one prove a negative? Can the ID critic prove ID is impossible?
[ 31. October 2002, 12:06: Message edited by: Jack ]
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 31. October 2002 12:14
Jack:
quote:
I disagree. The most the EF can do is show that something is highly improbable. It can't prove that something is impossible. ID critics seem to think that inferences to blind watchmaking will always trump inferences to ID unless ID advocates can prove blind watchmaking impossible. I think that's a very weak position.
There seem to be two major problems with your argument. RBH is correct and your disagreement is also correct insofar as its capabilities are concerned. First in order to understand the EF, one need not look further but Dembski. You will notice that the EF claims to be able to INFER design, _WITH NO FALSE POSITIVES_, by elimination of chance and regularity. The logic of the EF is to show that a certain event E could not have happened through chance/regularity and thus ID should be concluded.
Once you realize the extend of the claims of ID, not only can it show that a particular hypothesis is unlikely but it can also show without false positives that ID should be infered. RBH's description of EF seems correct.
ID critics do not seem to be arguing as much that 'blind watchmaking' will always trump ID but rather that since ID does not present us with any hypotheses it cannot be compared with chance/regularity hypotheses. When we cannot describe the detailed processes involved (flagellum) how can we infer design? So in fact the ID inference is based on showing that 'blind watch making' is impossible but the approach used suffers from a variety of what I consider to be show stoppers. First of all the EF requires probabilities to be calculated. But how does one calculate the probability of the flagellum arising through various proposed pathways? Secondly, the EF is claimed to be reliable (no false positives) which makes its claims even more unlikely. No False positives seems to require either omniscience on us (so that we can test and eliminate all hypotheses) or places some restrictions on the mechanisms of ID, namely only those which could not possibly have chance/regularity origins are allowed. Thus my argument, which I will present more fully at a later time, is that ID needs to deal with 'incomplete knowledge' and the only way to do this at this moment seems through a Bayesian approach. This however requires ID to propose testable hypotheses.
There may be other ways of infering design. Direct observation would not require one to eliminate chance/regularity for instance.
It may be helpful to compare ID with areas in which ID can be sucessfully applied and determine what makes them succesful (while also realizing that they are not infallible either). Criminology comes to mind as a great example. Motive, means, opportunity, eye witnesses, indirect evidence, are all used to propose and reject various hypotheses. It seems inevitable to me that one has to propose hypotheses.
IP: Logged
|
|
|
Printer-friendly view of this topic