|
Author
|
Topic: Principle of Causal Adequacy and Specified Complexity
|
William A. Dembski
Member
Member # 7
|
posted 23. March 2002 13:07
A basic regulative principle for science is the Principle of Causal Adequacy. Briefly, satisfactory scientific explanations must invoke causes are that adequate to account for their effects. If a cause is not adequate to account for an effect, the effect is not scientifically explained. The transformation of lead into gold by alchemical means violates the principle of causal adequacy because it proposes causal powers like heat and chemical forces that, for known theoretical reasons, are incapable of effecting the desired transformation.
Consider now the Principle of Causal Adequacy in relation to specified complexity. For something to exhibit specified complexity it must exhibit a detachable pattern that maps onto an event of small probability (the "complexity" in "specified complexity" is a measure of probability). Now, the determination of small probability is made with respect to all relevant probability distributions that might account for the phenomenon in question. In practice this means limiting oneself to probability distributions induced by all "known naturalistic mechanisms operating in known ways" (I'm grateful to Steven J. on talk.origins for this apt phrase -- though the criticism he intended severely misses the mark). Evolutionary naturalists often cite this as a defect of specified complexity claiming that it merely provides a sophisticated cloak for ignorance.
Two comments to this objection: (i) the great strength of Darwinian and other naturalistic accounts of evolution was precisely to show that known naturalistic mechanisms operating in known ways could produce all of biological complexity, so at the very least specified complexity is showing that the problems claimed to be solved by naturalistic means have not been solved. (ii) The argument from ignorance objection can be raised for any design inference that utilizes specified complexity, including those where humans are implicated in constructing artifacts. There may be unknown naturalistic mechanisms that lead to identical essays being written by independent agents even though now we routinely refer such coincidences to design (i.e., plagiarism).
Specified complexity, by being defined relative to the relevant probability distributions that might account for the occurence of an event, or alternatively by being defined relative to the known naturalistic mechanisms operating in known ways, might always be defeated by showing that some relevant probability distribution was omitted. That's always a possibility (though as with the plagiarism example and with many other cases, we don't take it seriously). As William James put it, there are live possibilities and then again there are bare possibilities. There are many design inferences which, to question or doubt, requires invoking a bare possibility. Such bare possibilities, if realized, would defeat specified complexity. But how would they defeat specified complexity? Not by rendering the concept incoherent but by dissolving it (or as I put it in NFL, by rendering the specified complexity only apparent).
In fact, that is how Darwinists, complexity theorists, and anyone intent on defeating specified complexity usually attempts it, namely, by showing that the probability or complexity that was thought to be so extreme really wasn't all that extreme after all (cf. Dawkins's _Climbing Mount Improbable_). Those who want to defeat specified complexity therefore try to show that it isn't real -- that the notion dissolves once we have a better understanding of the underlying causal mechanisms that render the object in question reasonably probable. By contrast, the design theorist argues that the specified complexity is real: that any attempt to palliate the complexities/improbabilities is destined to fail. This can in some cases be argued conclusively, as when the geometry and chemistry of some biological structure allows complete freedom in possible arrangements of parts (cf. the sequencing of nucleotide bases). Michael Polanyi made such an argument in the 1960s.
To bring this discussion back to the Principle of Causal Adequacy, it is known that intelligence has the causal power to produce specified complexity. In other words, intelligence is causally adequate to produce specified complexity. Moreover, in certain cases like those Polanyi, Michael Behe, James Shapiro, and others have considered, there is good reason to think that no natural mechanism has the causal power to produce such systems and thereby to dissolve the specified complexity attributed to them. It seems, then, that the Principle of Causal Adequacy requires, at the very least, that design be permitted as a live option within biology.
[Note, this post draws heavily from my response to Darel Finley's article on the ISCID archives]
IP: Logged
|
|
nobody
Member
Member # 145
|
posted 23. March 2002 21:49
quote:
the Principle of Causal Adequacy requires, at the very least, that design be permitted as a live option within biology.
I agree completely. I thank you for pointing that out to so many people. You are having a very positive impact.
I assume under the design umbrella you would include programming?
A GRN is essentially very advanced programming. The following paragraph says it better than I ever could. Pleae notice the the reference to "multiple tiers of regulation". Emphasis added is mine.
"Gross anatomy of a minimal gene regulatory network (GRN) embedded in a regulatory network. A regulatory network can be viewed as a cellular input-output device. At minimum, a gene regulatory network typically contains the following components: (1) an input signal reception and transduction system that mediates intra and extracellular cues (left box; often, more than one signal impinges on a given target gene); (2) a "core component" complex composed of transacting regulatory proteins and cognate cis-acting DNA sequences (circle; functionally similar components may be associated with multiple target genes, resulting in similar gene-expression patterns); and (3) primary molecular outputs from target genes, which are RNA and protein (box to right of circle). The net effects are changes in cell phenotype and function (right box). Direct and indirect feedbacks typically are important. More realistic networks often feature multiple tiers of regulation, with first-tier gene products regulating expression of another group of genes, and so on. Beyond GRN boundaries are signaling responses and feedbacks, such as those that drive bacterial chemotaxis, which do not involve regulation of gene expression but instead act directly on proteins and protein machine assemblies (dashed arrows). Some regulatory networks have no embedded GRN component."
Link here: http://doegenomestolife.org/roadmap/grn.html
This is a US Dept of Energy site, not an IDist site. I believe is is overwhelmingly obvious that advanced programming such as this requires a highly intelligent programmer. Programming of this type is currently beyond human capability.
How advanced is this type of programming? The DOE's goal is to: "Develop the capability within the next 10 years to comprehensively map microbial and metazoan regulatory circuitries."
Obviously we still have a long way to go.
Link for the quote is here: http://doegenomestolife.org/roadmap/goal2.html
IP: Logged
|
|
James A. Barham
Member
Member # 50
|
posted 24. March 2002 12:43
Bill:
I agree with you---partly.
I certainly agree that present-day science does not have the conceptual resources to explain the teleological organization of life.
However, the question is, What does this imply for the future progress of science? Can we confidently predict, on the grounds of a priori reasoning, that any particular phenomenon will forever remain outside the scope of future science?
What is to prevent us from saying that biology today is in the same position as physics at the end of the nineteeth century? It is beginning to dawn on many people (in no small part thanks to your efforts) that there is a gaping hole in our knowledge of how life works, just as it dawned on Planck, Bohr, and the others that classical mechanics did not have the conceptual resources with which to explain the stability of the atom. I suppose someone could have argued then (maybe someone did?) that this means that a supernatural force must be supporting the atom externally. But in the event, a way forward was found, and a whole new formalism was discovered that was more adequate to the phenomena. What is there in principle to make us think that a similar way forward may not be found eventually in biology?
IP: Logged
|
|
Jules
Member
Member # 181
|
posted 24. March 2002 13:44
James,
Perhaps I should save these remarks for the future thread on Extrinsic/Intrinsic Teleology, but it seems to fit here. It seems that you are willing to grant what appears to be some sort of intelligence to organisms, that has had some influence in directing their evolutionary development. And that seems to be based on some sort of principle of causal adequacy. But I may be misunderstanding you here, so what I say next is said cautiously. If I understand you so far, doesn't this leave a problem for the first appearance of life? You appear to deny any intelligent agency apart from organisms, so we have a missing teleological agent for the first appearance of organisms, therefore causal inadequacy. But then I may have misuderstood you to begin with.
[ 24 March 2002, 13:45: Message edited by: Jules ]
IP: Logged
|
|
James A. Barham
Member
Member # 50
|
posted 24. March 2002 22:42
Jules:
Oh, yes indeed, not only would I acknowledge "a" problem with the origin of life, I would say that it is "the" problem---that is, the fundamental problem at the heart of all our discussions, from a scientific point of view.
Well, I don't claim to have a theory of the origin of life. Nor, I think, does anyone else, at least not one that holds waterd. Not yet. So what can I say about it?
I spoke a little bit about the concept of "emergence" within the context of quantum field theory and the theory of condensed matter over on the GA thread, so I won't repeat all that here. (And, anyway, it is all readily accessible in much more coherent form in Robert Batterman's new book, The Devil in the Details.) I will just say that I do not see why in principle the same set of ideas may not be extended to encompass the origin of life eventually. That is to say, while every one of the levels in the emergent hierarchy is distinctive or special in that it has its own particular feature, and so life is special too in this same sense, I do not see the level of life as being "specially special," so to speak. Only special in the generic sense that it has qualitative features unique to it, which in this case happen to be those of teleological organization or intelligenty agency.
As for the evolutionary process, I would want to invoke the distinction discussed at greater length over on the Evolution and Design thread between goal-directed systems or agents, on the one hand, and dynamically stable ensembles of such agents, on the other. That is, the distinction between organisms and ecosystems, or between firms and markets. Now, while I believe that organisms strive intelligently to maintain themselves in existence, and can apparently undergo transformations over time in ways we do not really understand, I am in no way claiming that the process as a whole is intelligently directed, as though by an external intelligence. Rather, just as the individual strivings of various organisms result in various dynamically stable ecosystems synchronically, so too do organismal transformations result in various evolutionary trajectories and ecosystem successions, diachronically.
This does not mean I do not think that there is directionality to the evolutionary process. I think it is obvious that there is (see my "On the Objectivity of the Scala Naturae," Evolution and Cognition, 1999, 5: 2--11.) It just means that I view this directionality as a resultant of the individual goal-directed strivings of organisms, not as goal-directed in and of itself.
Perhaps I should also stress that I do not think we understand how the intelligent striving of the organism impacts evolutionary change. Apparently the genome is under cellular control, but that does not mean there is a direct Lamarckian mechanism of instruction of the environment upon the genome. Rather, it is most likely some sort of highly constrained stochastic search mechanism, just like other forms of trial and error that organisms use intelligently.
We should be careful not to confuse intelligence with clairvoyance. Intelligence always implies a means of direct dynamical interaction between agent and world. How this works in detail in evolution remains to be worked out.
[ 24 March 2002, 22:48: Message edited by: James A. Barham ]
IP: Logged
|
|
Mister Pamboli
Member
Member # 175
|
posted 25. March 2002 12:02
I may be missing something, but it appears that in essence this topic boils down to one sentence ... quote:
Moreover, in certain cases like those Polanyi, Michael Behe, James Shapiro, and others have considered, there is good reason to think that no natural mechanism has the causal power to produce such systems
But if this is all that is intended, the debate does not advance, except in so far as it has a new label to use.
If we regard specified complexity in the abstract, without regard to how an intelligence could implement its designs, then intelligence is clearly causally adequate. But then again, is there any conceivable phenomenon for which intelligence is not causally adequate? Design could be "permitted as a live option" in explanation of anything. If intelligent design is always available as a default explanation, then when do we reject it? If, as I suspect, the answer is only by applying Occam's razor when a naturalistic explanation is causally adequate, then we are back to debating not what intelligence can explain, but what naturalistic mechanisms can not. We are back to "design in the gaps."
Prompted by this chain of thought, and turning it on its head, I wonder if the ability to produce specified complexity may be a diagnostic feature of intelligence?
IP: Logged
|
|
Jesse
Member
Member # 112
|
posted 25. March 2002 15:34
William A. Dembski:
To bring this discussion back to the Principle of Causal Adequacy, it is known that intelligence has the causal power to produce specified complexity. In other words, intelligence is causally adequate to produce specified complexity.
If we're still talking about specified complexity relative to "all known naturalistic mechanisms operating in known ways" then it is certainly not "known" that intelligence can produce specified complexity--in fact this claim is equivalent to the claim that there can be no naturalistic explanation for the behaviour of intelligent beings. Most cognitive scientists and neuroscientists suspect that intelligence can, at least in principle, be explained naturalistically, which would imply that things like art, technology, architecture, discussion board posts, etc., do not really exhibit specified complexity in the first place. Of course our understanding of the brain/mind relation is still pretty limited and this issue is up for grabs, but it is at least true that we should not assume that intelligent agents can produce specified complexity.
IP: Logged
|
|
Jeremy Alder
Member
Member # 32
|
posted 25. March 2002 19:54
quote:
Mister Pamboli said:
If we regard specified complexity in the abstract, without regard to how an intelligence could implement its designs, then intelligence is clearly causally adequate. But then again, is there any conceivable phenomenon for which intelligence is not causally adequate? Design could be "permitted as a live option" in explanation of anything. If intelligent design is always available as a default explanation, then when do we reject it? If, as I suspect, the answer is only by applying Occam's razor when a naturalistic explanation is causally adequate, then we are back to debating not what intelligence can explain, but what naturalistic mechanisms can not. We are back to "design in the gaps."
I think it is true that in principle, design cannot be completely ruled out as an explanation for any event. But the same goes for chance. The danger of "design in the gaps" has its counterpart in "chance of the gaps". To make either one of these notions (design or chance) useful to science we need to have a scientific way of ruling them out. According to Dembski, design can be ruled out if the event/object in question is not specified and sufficiently complex. In other words, if it is explainable as a result of either chance, law, or a combination of the two then for scientific purposes design should not be inferred. As for chance, Dembski calculates a "universal probability bound" which allows us, for scientific purposes, to rule out chance for events/objects whose complexity/improbability exceeds those bounds. I would recommend Dembski's book "No Free Lunch" for more details.
IP: Logged
|
|
Art
Member
Member # 179
|
posted 25. March 2002 22:25
quote:
I think it is true that in principle, design cannot be completely ruled out as an explanation for any event. But the same goes for chance. The danger of "design in the gaps" has its counterpart in "chance of the gaps". To make either one of these notions (design or chance) useful to science we need to have a scientific way of ruling them out.
I think this statement reveals too great a dependence on negative results or inferences. In the normal course of scientific exploration, things are not explained by ruling out other possibilities, but by ruling in a specific hypothesis, mechanism, or whatever.
IP: Logged
|
|
Jeremy Alder
Member
Member # 32
|
posted 25. March 2002 23:29
quote:
Art said:
I think this statement reveals too great a dependence on negative results or inferences. In the normal course of scientific exploration, things are not explained by ruling out other possibilities, but by ruling in a specific hypothesis, mechanism, or whatever.
The purpose of my statement was simply to point out that ID is no more a default explanation for everything than bare chance. That which explains everything explains nothing. So in order for either chance or design to be scientifically useful explanations there must be some way to rule them out. The same goes for natural selection or any other possible explanation.
IP: Logged
|
|
Mister Pamboli
Member
Member # 175
|
posted 26. March 2002 00:00
Jeremy, thanks for responding. I have read Dembski's work including NFL, but my argument here is not with the design inference itself. (Not that I have no arguments with it, he hastily adds.)
My point was simply that invoking causal adequacy adds nothing to the debate. Art is right, in that it provides no way to "rule in" ID, except in the facile use of ID as a default mechanism.
Your later suggestion that "ID is no more a default explanation for everything than bare chance" addresses that comfortably, but my orginal point still remains: that the principle of causal adequacy, borrowed into science from Descartes' ontology adds nothing but a label to the same old problem of delineating what naturalistic and intelligent causes can effect.
Jesse spies an important weakness, I think, which will give me (and others, I hope) much food for thought.
IP: Logged
|
|
Micah Sparacio
Member
Member # 6
|
posted 26. March 2002 00:39
Jesse:
I think that your question is a valid and important one. My point of contention is with the idea that if we have a natural explanation for intelligence, then the concept of specified complexity is null and void. I don't think that this follows. Regardless of its origin, there is still something unique about the causal force of intelligence. Dembski is arguing that the unique feature left by this intelligent force acting on matter is specified complexity. Perhaps intelligence is some sort of special category of naturalistic explanation. In this case, it is distinct from all other known naturalistic explanations. Therefore I think a constructive empiricist approach to design is still valid, even if we grant that intelligence can be explained naturalistically.
The theoretical notion of a designing intelligence need not suppose that there is an actual designer. However, the one thing we can be sure of is that human intelligence is capable of leaving unique identifiers of its activity in the physical world. If intelligence has a naturalistic explanation, then we would simply have to identify those features of this special naturalistic explanation(intelligence) and Dembski contends that these features would be specified complexity.
[ 26 March 2002, 09:21: Message edited by: Micah Sparacio ]
IP: Logged
|
|
Jesse
Member
Member # 112
|
posted 26. March 2002 01:22
Micah Sparacio:
If intelligence has a naturalistic explanation, then we would simply have to identify those features of this special naturalistic explanation(intelligence) and Dembski contends that these features would be specified complexity.
Well, if we had a naturalistic explanation in hand for the behavior of intelligent beings, then presumably the "chance hypothesis" induced by this theory would yield a reasonably high probability that such beings would produce specified objects like art, novels, buildings, technology, and so forth. Since the amount of "complex specified information" is defined by Dembski as the probability of getting a specified event under the appropriate probability distribution (the chance hypothesis), this would mean that such things don't actually exhibit specified complexity. It is still possible there could be some unique thumbprint of intelligent action, but it wouldn't be specified complexity.
IP: Logged
|
|
Jules
Member
Member # 181
|
posted 26. March 2002 01:27
Jesse: "Since the amount of "complex specified information" is defined by Dembski as the probability of getting a specified event under the appropriate probability distribution (the chance hypothesis), this would mean that such things don't actually exhibit specified complexity. It is still possible there could be some unique thumbprint of intelligent action, but it wouldn't be specified complexity."
But the "appropriate probability distribution" in this case would be ones where there are intelligent agents.
IP: Logged
|
|
Jesse
Member
Member # 112
|
posted 26. March 2002 01:47
Jules:
But the "appropriate probability distribution" in this case would be ones where there are intelligent agents.
Right, that was exactly my point. Things like art and literature and technology would presumably have a fairly high probability under such a distribution, and high probability = low "complexity" according to Dembski's definition.
IP: Logged
|
|
|
Printer-friendly view of this topic