Arguing for ID By Arguing Against Evolution: Is It Legitimate?

Arguing for ID By Arguing Against Evolution: Is It Legitimate?

Evolutionists complain that positive arguments for ID are lacking, and that all ID offers are negative arguments against evolution. Without granting all of that, I want to address whether there is something wrong with arguing for ID by arguing against evolution. I have done this before, but I’ve developed the argument further since then. This is adapted from a comment I posted on another thread earlier today.

I have argued in the past that there are only two possible explanations for biological origins on the table: either there was some intelligence guiding it, or there wasn’t. If the first is true, then some form of Intelligent Design is the right explanation. If not, then the only explanation on offer is unguided neo-Darwinian evolution (hereafter NDE).

At the end of the movie Expelled, Richard Dawkins speaks of the possibility that life on earth was designed, and opines that it might have been if the source of that design were some alien creatures. Many ID proponents found that a funny position for him to take, but in our laughter many of us missed what else he said: that those aliens, if they existed, must have come about by Darwinian processes. For Dawkins, at least, there is only one route up what he calls “Mount Improbable,” and that is the gradualistic road of natural selection acting on random variations.

So if he is right, then there are only two options on the table: Intelligent Design (in some form), or NDE. Now, these are dichotomous: if one is true, the other is false. Mainstream evolutionary scientists insist that NDE is fact, and that we know it is fact. One helpful way to express this certainty is to express it in terms of probabilities: p(NDE)=1 and p(ID)=0.

Because there are no other options on the table or even on the horizon, it would appear that the probability relationship must include only the terms stated so far here. That is, p(NDE) + p(ID)= 1. If the probability of either term is 1, then the probability of the other is 0; if the probability of either term increases or decreases, then the probability of the other term decreases or increases by like measure.

ID theorists argue that certain features of the natural world are inconsistent with NDE. The Cambrian Explosion is one of them. It is hard to explain, strictly on NDE terms, how it came about. This is an example of a negative argument against NDE. This post is not about whether that is true or not; it is about whether, if there is merit to the argument, it counts legitimately as an argument in favor of ID.

And it seems to me that given the binary relationship between ID and NDE, it must; for p(NDE) + p(ID)= 1. Suppose there is some merit to ID’s concerns about the Cambrian Explosion. The effect of that must be to reduce confidence in NDE by some non-zero amount. Supposing also that before this argument was made, the universal consensus was that  p(NDE)=1. To the extent that the Cambrian Explosion argument has merit, that confidence would be reduced, and the result would be that p(NDE) < 1, and p(ID) > 0. (How much those probabilities change depends on how successful the argument actually is.) Confidence in ID (its increase in probability) would be numerically identical to the decrease in confidence in NDE, because the sum of the two probabilities must equal 1.

Thus a negative argument against NDE is a positive argument in favor of ID.

But when I have written of this before, some have objected to my considering only two possibilities: ID (in some form) and NDE. “How do we know these are the only two possibilities?” they ask. “Science marches on, and who knows what we might discover? Why do we assume that ID is the only alternative to NDE? How could we know that?”

We can deal with that this way. Let’s grant that there could be Unknown Possibilities, and call them UP. In that case we would have to say, p(NDE) + p(ID) + p(UP) = 1. That certainly ought to cover all the possible explanations.

There must be some number between 0 and 1 that expresses the probability of Unknown Possibilities — p(UP) — explaining biological origins. How shall we assign that probability? How slippery is that number? Does it always adjust to p(NDE), such that p(NDE) + p(UP) always equals 1? It seems to me some participants in this debate or so opposed to ID, that’s what they would insist: “ID is not the answer. If NDE turns out not to be the answer either, then there must be some other unguided, naturalistic explanation for life.” That’s equivalent to saying p(NDE) + p(UP) always equals 1, and p(ID) always equals 0.

But treating unknown possibilities that way is nothing more than a “no-design of the gaps” argument. It starts with ignorance with respect to the unknown possibilities and moves to an assumption that if NDE is not the answer, then UP is. It does so with no knowledge: “unknown” means unknown, after all. That’s not very impressive reasoning, and in fact I can’t think of any scientific or logical reason to accept it.

What if we take a more balanced view, then? What if we admit the possibility of UP, and we do so in a way that avoid the “no-design of the gaps” error. In that case the most reasonable way to proceed would be to assign p(UP) some more definite value, like, perhaps, 0.2. The number we choose really doesn’t matter, in view of the point I’m trying to make here, which is that a negative argument against NDE can be a positive argument for ID. We could simply call p(UP) an unknown constant and make the same argument we made above, adding K into the equation for that constant. My central point above remains identical to what it was before:

p(NDE) + p(ID) + K = 1. If the probability of either non-K term is 1, then the probability of the other is 0; if the probability of either term increases or decreases, then the probability of the other term decreases or increases by like measure. A negative argument against NDE is still a positive argument for ID.

Series Navigation (The Evolutionist's Complaint):The Evolutionists’ Complaint: It’s Wrong to Argue For ID By Arguing Against Evolution (Part 1 of 3) >>>

59 thoughts on “Arguing for ID By Arguing Against Evolution: Is It Legitimate?

  1. Tom,
    I see it this way until the day that NDE does the work necessary to earn the right to reverse the inequality in Eqn 1.

    Eqn 1: [p(NDE) + p(UP)] < p(ID)
    Eqn 2: p(NDE) + p(UP) + p(ID) = 1

  2. Let’s test this meta-theory on a well-known problem of missing solar neutrinos.

    In a nutshell, the standard model of particle physics (coupled with our understanding of the solar interior) predicted a certain flux of electron neutrinos arising in nuclear reactions on the sun and reaching the Earth. Actual measurements indicated a neutrino flux that was three times lower, thus contradicting the theory. Naturally, creationists used the discrepancy to argue that the natural theory of the sun was wrong and thus the sun could be much younger than 4.5 billion years.

    So, along the lines of your meta-theory, there were three options: the Standard Model of particle physics + theory of solar interior (SM), God did it (GDI), and unknown possibilities (UP). Experiments prior to 2002 seemed to indicate that SM was wrong. In 2002, the SNO experiment and further experimental work showed that it was UP: neutrinos changed flavors as they flew towards the Earth, so not all of them were registered by detectors sensitive to electron neutrinos only.

    It’s clear that p(UP) was less than one prior to 2002, while afterwards it reached one because we now know with certainty that neutrino flavors oscillate. So it seems to me that your meta-theory contains at least one wrong assumption: p(UP) is not necessarily constant in time.

    At any rate, even dressed in probability theory, God of the gaps is bad theology.

  3. You are assuming there is an ID theory for e.g. the Cambrian Explosion. Please provide it. Tell us what hypothetically happened, when, how, etc., and how it has been/can be tested.

    Also, google “contrived dualism.”

  4. Thank you, Nick, for that reminder of the importance of assumptions. You are assuming that we cannot operate with a theory of intelligence acting in the world unless we know how it works in the world. Could you explain, please, how intelligence operating through your body works to produce the message you sent? If you can do that, then I will accept the premise of your question.

    You are also assuming that we cannot theorize intelligence operating to design the conditions of life unless we can describe how that intelligence did it. You’re assuming naturalism by making that assumption; for your requirement for such a description is impossible in principle unless everything that happens in the world happens naturalistically. That’s called begging the question.

    You have also failed to notice that I wrote a test for intelligence operating in the world in this post. It is not the only test, probably not even the best test. But I don’t think you realize that evolution millions of years ago has not been “tested” either. It has been inferred from evidence. We also infer the operation of intelligence from evidence. Both arguments–for evolution and ID–are arguments of the form, inference to the best explanation. Neither one can be directly tested.

    You are also assuming that if the true answer to your question should turn out to be, “God did it by means unknown,” then everything I have written in this post is false, and evidence against NDE does not provide probabilistic evidence in favor of intelligent design. Could you explain how that, too, is not begging the question?

    As for “contrived dualism,” did you notice that in this post I addressed that issue in my analysis of unknown possibilities? And are assuming, as the scientist you are, that argument-by-google is a great take-down for an argument by probabilities analysis?

  5. olegt, you are right: p(UP) is not necessarily constant in time. But to deny the main point of my post, you must assume that p(UP) will always be elastic in that way for every current and future point under discussion. I think I already addressed the weakness of making that assumption, in my original post.

    You’ve said that “God of the Gaps” is bad theology (I agree on that, by the way, and sometime we can talk about why that doesn’t apply here the way you probably think it does). Can you see that UP-of-the-gaps is also bad science?

  6. Tom,

    “UP of the gaps” is not good science. Why do you think I would make such a claim? If cosmology is unable to account for the bulk of matter in the Universe, it means that our current theories are incomplete and possibly wrong on some points.

    My example of the missing solar neutrinos was meant to illustrate something else: an inherent asymmetry between science and creationism understood as a purely negative argument. Science hasn’t closed all of its gaps, but it has a track record of closing them. And every time a gap closes, creationism gets discredited. That happened many, many times in history. (Just to add one more example, Newton thought that the orderly appearance of the solar system “required the divine Arm,” but now we have a natural theory of its origin.) So the problem with negative argumentation is that every time science fills a gap, creationism gets discredited.

    So here is the bottom line. Science’s promissory notes are worth something because of its track record of filling gaps. Creationism’s aren’t worth the paper they are written on.

  7. olegt,

    It seems to me that you have admitted there is a promissory note involved in your expectation that science can eventually solve problems like the ones we’re discussing. That promissory note amounts to an unknown naturalistic possibility, or UP, and it’s what you are proposing in place of ID in the probability equations I have specified above. You are saying that instead of this:

    p(NDE) + p(ID) + p(UP) = 1

    we ought to take it that

    p(NDE) + p(UP) = 1

    Or perhaps you might allow ID some scant possibility of being true, in which case we would have a set of equations that might read something like this:

    p(ID) = .01
    p(NDE) + p(UP) = .99

    (You can pick whatever figure you like for p(ID); the analysis comes out the same.)

    Now here’s what that means in context of what I have written in this post. Let’s suppose there is some evidence E that is difficult to account for in terms of NDE. In my post I used the Cambrian Explosion as that evidence, but in reality it could be any E. An honest observer should consider that E reduces confidence in NDE, meaning that p(NDE) is less than it would be apart from E.

    Your position implies that as our confidence in NDE decreases, or, as p(NDE) decreases, that should lead us to conclude that p(UP) is increased by the same amount; that we are more confident than we were before that some unknown naturalistic possibility accounts successfully for biological origins.

    In plainer language, as evidence (E) reduces our confidence in NDE, we can confidently assert that whatever probability gap is thus created can be filled by as-yet unknown naturalistic possibilities.

    That is UP-of-the-gaps, if you ask me.

    You went on to suggest that we ought to be more confident that science could fill the gaps in our knowledge, if some E causes gaps of that sort, than that non-science could fill those gaps. This is because “science hasn’t closed all its gaps, but has a good track record of closing them. And every time a gap closes, creationism gets discredited.”

    First, there is still a promissory note there, in spite of science’s track record. You acknowledged that.

    Second, we’re talking about ID, not creationism.

    Third, if ID were entirely an “of-the-gaps” argument, your argument here might have more force; but it isn’t. It is an inference-to-the-best-explanation argument, not based on unknowns but on what we do know. Our uniform experience shows us that information always comes from mind. Science shows us that there is a high level of information in DNA (and possibly elsewhere, based on recent research). It’s reasonable to infer mind as the cause of that information, and that is not a promissory-note inference. It’s based on information already at hand.

    Fourth, speaking now of biblical understandings of God, rather than of ID, God does not suffer when science demonstrates natural explanations for phenomena. There is nothing in biblical theology to suggest that God does not work through means. See my article on Lawrence Krauss at BreakPoint for more on that.

  8. “God did it by means unknown” is pretty much what you think, isn’t it? And that’s the nub of the issue. As far as actually explaining anything about the Cambrian “explosion,” it’s worthless. It’s only “useful” for e.g. apologetics, which is deep down the whole point of arguments like yours and Meyer’s. But it doesn’t actually explain anything, doesn’t predict anything, etc.

    Over here in science we like explanations that actually do some explaining rather than just inserting a big mysterious-power-that-can-do-everything-so-it-did-this-through-basically-magic “explanation.” There are series of proposed scientific explanations for the Cambrian radiation (it’s not really an “explosion” anyway), some or all of them may be true, they are all testable in numerous ways, and are being tested.

    You go on:

    “Our uniform experience shows us that information always comes from mind.”

    Wildly, horribly false claim derived from Meyer. Can be (and was in Kitzmiller) sunk with one review paper:

    Long et al. (2003), “The origin of new genes: glimpses from the young and old.” Nature Reviews Genetics.
    http://scholar.google.com/scholar?hl=en&q=evolution+of+new+genes+glimpses&btnG=Search&as_sdt=2000&as_ylo=&as_vis=0

    The paper is free online. Look at Table 1 and Table 2. New genes with new functions = new information on any meaningful definition (though creationists will often be craven and waffle to avoid admitting that their favorite argument about “information” has been falsified).

    That said, it’s not even clear that all that much “new information” was needed for the Cambrian. Meyer asserts this, but he doesn’t have a rigorous definition of information, he asserts that some things have more or less information, but he makes no attempt to provide us with a quantifiable definition so that we can check these claims.

    Coelomate animals have mostly the same sets of basic genes, the common ancestor was a wormlike critter with all those genes. So most of what happened was expressing the genes in somewhat different patterns, producing modified worms. (And most of the phyla living today are still basically just worms! Take a good look at a Cambrian “vertebrate” fossil — it looks more like a worm than a modern vertebrate! The earliest fossils, and earliest-diverging members of modern phyla, or their close sister groups, are mostly pretty wormlike.)

    It is problematic when a creationist takes scientific ignorance and infers the action of God in that gap. But that’s not the biggest problem. The biggest problem occurs when creationists take *their own personal ignorance*, conclude from that ignorance that the entirety of science is similarly at a loss to explain information or the Cambrian radiation, and then insert God to fill a gap which really isn’t even a gap to the biologists who know what they are talking about. It’s not much better than when some primitive tribe attributes volcanic activity to evil spirits.

  9. Tom,

    1. When you have a choice of lending money to two people, one of whom has some positive credit history and the other doesn’t, which one would you lend the money to? I’m sure that the answer is clear. Likewise, it is much more likely that a promissory note from science is a better investment than one from creationism.

    2. We’ve gone over that at length before and I don’t think anyone can dispute that IDers use the exact same arguments as do creationists. I have studied IDers closely for a few years and I have yet to see any new argument that would distinguish them from good ol’ creationists.

    3. I’m sorry, but these terms are not well defined. As I remarked here multiple times, creationists don’t use the standard definition of information (Shannon). They came up with some vague notion of specified, functional, etc. information that nature on its own can’t produce but mind can. The problem is that no one presented a recipe for measuring this new fancy information in any real system. (I have read Dembski’s latest, so don’t point me to it.) Wake me up when there is something to discuss.

    4. I have no problem with that.

  10. olegt:

    “When you have a choice of lending money to two people, one of whom has some positive credit history and the other doesn’t, which one would you lend the money to? I’m sure that the answer is clear.”

    It depends on rate. If the rate is high, the expected value of investment may be better in the case of negative credit history, even if the propability of getting all the money back is under the 1 (but better than 0).

    Bay the way, the return on investment (ROI) may be very good in christianity, if it is examined in eternal period.

  11. olegt:

    “I’m sorry, but these terms are not well defined. As I remarked here multiple times, creationists don’t use the standard definition of information (Shannon). They came up with some vague notion of specified, functional, etc. information that nature on its own can’t produce but mind can.”

    I’m not sure, if there are any “standard definition” of information. For example in ScienceDirect there are about 2,961,171 articles, where the word information is used. And only 36,680 articles, where are both words: information and Shannon (at least in the reference list). And many of even those articles don’t define the word information as Shannon. The most of them just mention the words Shannon and information in the same article. Shannon’s definition is not standard definition of the word information. It is popular in some areas (related for example to the topic of entropia, packaging etc.) but it is not very popular in many other areas).

  12. olegt, you write,

    I have yet to see any new argument

    You haven’t been looking as closely as you say you have. Whether ID can be distinguished from Creationism is a question that takes some extended work on definition, and we’ve done it before. I don’t think in this case it even matters. What matters is that there are new arguments out there. Good ones.

    Feel free to sleep until you think there’s something to discuss.

  13. Nick, you’ve presented a general argument against ID. I’ll read the paper you linked to. But does it have anything whatsoever to do with the topic of the post?

    Every time I post on a specific point intending to advance understanding on this debate, I get the same old arguments against everything involved in ID. I credit olegt for actually responding to what I wrote about. Nick, you can spatter your anti-ID beliefs over any discussion thread you want, but when someone presents an argument from a new angle, can you respond to it? (You were pretty quiet here. So were almost all the ID opposition. Funny thing.)

  14. Tom:

    p(NDE) + p(ID) + K = 1.

    I don’t see why you use K here instead of p(UP). p(UP) seems to be the probability that life is the result of a process(es) neither intelligent nor currently known or predicted by NDE. Therefore, the correct equation should be p(NDE) + p(ID) + p(UP) = 1. But with that equation, an argument that decreases p(NDE) does not necessarily increase p(ID), and seems to work against your central point.

    What if we admit the possibility of UP, and we do so in a way that avoid the “no-design of the gaps” error. In that case the most reasonable way to proceed would be to assign p(UP) some more definite value, like, perhaps, 0.2.

    The reasonable way is to assign it unknown. Unknown means unknown, as you say. ID opponents, you say, are using this equation p(NDE) + p(UP) + 0 = 1, by insisting that p(ID) is 0. I agree with you, it should be p(ID), not 0. Likewise, your K should be p(UP). What am I missing?

  15. This “information” thing is one of the central points in the whole debate.

    I’ll put money on the point that neither Nick, nor Geoff, nor DL, and certainly not olegt can define “information” more broadly apart from mathematics or from some scientistic operational definition. Tom nailed this perfectly in pointing out Nick’s HUGE question-begging: “You [Nick] are assuming that we cannot operate with a theory of intelligence acting in the world unless we know how it works in the world.” A similar point is Nick’s misappropriation of “functionality” to serve evolution-ISM’s ends, which I noted under separate cover. Fallacious consistency does not translate to truth–it’s still fallacious.

    I’ll also put money on the point that neither can the ID theorists… and I won’t even touch the YEC-ists. I’m neither impressed nor concerned (nor are most scientists) by Dembski’s strained attempts to employ formally-understood Information Theory to allegedly claim inference to design… without first telling us what design is.

    Once these sides can understand and admit to a broader philosophical sense that considers especially final, exemplar, and formal causality, this would go a long way in avoiding the incessant speaking past each other.

    Additionally (a bit off topic) the presumed superiority of natural scientific knowledge or methodology is ideological nonsense. Olegt, for example, may throw a bone that philosophy and theology are other forms of knowledge, but the clear attempt is to elevated quite illicitly (and VERY unscientifically) MES knowledge above all others.

    The case in point is striking, and is most vociferously promulgated by DL. He often asserts that consistency in predictability is the lynch pin upon which the MESs are built. Well, of course: to a large extent that’s correct, but–as we’ve seen so often–it’s reduced to MES prediction… without addressing the blatant 800-lbs gorilla in the room: why are the MESs and their methodologies themselves consistent and “predictable.” Surely, circular reasoning that rest upon the MESs to prove their reliability, consistency, and “predictability” is a non-starter.

    What’s DL’s approach? First, that it’s axiomatic, otherwise we could not even reason. In fact, he doesn’t understand (formally) what an axiom is. Second, he buffers himself by jettisoning the Principle of Sufficient Reason–in the concrete example of his (and olegt’s) misinterpretation of what QM allegedly tells us about reality at an ontological level, but also broadly because it opens the door to following evidence beyond his scientistic capabilities. Third, even if we grant him this, it is, at base, a deep expression of… wait for it… faith in the intelligibility of the universe.

    In the absence of the faith that the universe is consistently intelligible, scientists would have neither the basis for believing that their scientific pursuits would yield fruit or the motivation for expending the considerable effort that science demands. True enough, and they might even agree with this. But is that the point? The point is that even this MUST be understood as well in order to be faithful to intellectual rigor and honesty. DL keeps on hammering over and over “just so” versions that some contingent things can’t be explained. Please… let’s stop that nonsense.

    Something else must exist to explain how man’s intellect and the intelligibility of nature (her “information” if you will) happen to conform to each other. This “thing,” because it is not present in itself but only through its effects (like a footprint), is not an object for the empirical scientist. But if empirical scientists deny its reality as a result of a method that formally excludes it before they set to work, then, in this regard, they have crossed over from science to ideology… which is pretty much par for the course these days.

    On the other side, the metaphor of the footprint (or Paley’s watch or whatever) is taken to the other extreme in the sense of imparting upon Information Theory way more than it can “see”. Information Theory doesn’t “see” information–it rests upon deep presuppositions that information’s full import is something “understood” by the MESs, and then they use it in the vain attempt to find design in nature. Information Theory is about a very rarefied notion of FORM expressed in highly abstracted mathematical formalisms. But this does not stop ID theories in incorrectly trying to “see” exemplar and final causality they term “design”… and then sell it as something to be taught in science classrooms.

  16. New genes with new functions = new information on any meaningful definition (though creationists will often be craven and waffle to avoid admitting that their favorite argument about “information” has been falsified).

    I’ll go out on a limb and say this doesn’t address Meyer’s claim at all, hence it doesn’t falsify it as Nick seems to think.

    Meyer’s is talking about the ORIGIN of information (semantics) in a materialist universe. The new functions (new genes) are ‘reading’ information, but that requires the semantics already be in place. Where does that semantic rule book ORIGINATE under materialistic NDE so that a new gene can ‘read’ and express themselves? In principle, NDE has no way to account for its origin. ID can.

  17. SteveK:

    I agree with you that Nick’s got it wrong. Evolutionism in action: the equal sign he uses is equivocation between “function” and “information” at an ontological level. Amazing.

    But, how does ID theory account for per se information, and even prior to that, how does ID theory define information? You may be in for a rude awakening in the sense that ID theory is trying to “infer design” from the MESs. It ain’t gonna happen: their own reductionism is ironic. All scientific findings must be properly interpreted, but you need something other than the MESs to form a basis for that interpretation–especially when one “infers” from one ontological kind to another. Think of it this way: when the MESs “see” pigments on a page, they can give you all sorts of interesting sensory-accessible knowledge about those pigments. What can the MESs do to “extract” or “abstract” meaning (in-FORM-ation) from those pigments… assuming there’s meaning “in” those pigments in the first place?

  18. Holo,

    All scientific findings must be properly interpreted, but you need something other than the MESs to form a basis for that interpretation–especially when one “infers” from one ontological kind to another.

    I understand what you are saying, but in my opinion science gets away with playing both sides of the fence.

    Science sees the outward effects of reasoning and consciousness in the observable physical actions of certain material objects (a.k.a. humans and animals) and infers intelligence. They ‘see’ the intelligence. If they can see it in humans/animals simply by observing the outward effects then why isn’t it possible to do the same inside the living cell?

    Nobody objects when science ‘sees’ intelligence in material objects because they got the non-empirical inference correct. Yeah, it’s not a scientific inference in that the inference can be demonstrated. What does intelligence look like and how would you describe it in physical terms? You can’t do that. Nobody says they aren’t ‘doing science’ because they inferred correctly. The problem is when they infer incorrectly as they are doing with NDE. Then they ask for the challenger to show they are wrong via the scientific method, forgetting that they can’t do that for their own inference.

  19. It’s not me that says information = functionally specific sequence — it is Stephen Meyer in Signature in the Cell:

    Thus, molecular biologists beginning with Francis Crick have equated biological information not only with improbability (or complexity), but also with “specificity,” where “specificity,” or “specified”, has meant “necessary to function.”

    Thus, in addition to a quantifiable amount of Shannon information (or complexity), DNA also conrains information in the sense of Webster’s second definition: it contains “alternative sequences or arrangements of something that produce a specific effect .” Although DNA does not convey information that is received, understood, or used by a conscious mind, it does have information that is received and used by the cell’s machinery to build the structures critical to the mainrenance of life. DNA displays a property — functional specificity — that transcends the merely mathematical formalism of Shannon’s theory. (Meyer 2009, p. 109, italics original)

  20. So let me get this straight, when you discover some thing which you believe supports the theory of evolution that counts as evidence against theism or design, but if I discover some thing which I believe contradicts the theory of evolution then that is “contrived dualism”?

  21. woodchuck64, in regard to your 9:21 pm question about assigning a value to p(UP), did you miss these paragraphs in the OP?

    Does it always adjust to p(NDE), such that p(NDE) + p(UP) always equals 1? It seems to me some participants in this debate or so opposed to ID, that’s what they would insist: “ID is not the answer. If NDE turns out not to be the answer either, then there must be some other unguided, naturalistic explanation for life.” That’s equivalent to saying p(NDE) + p(UP) always equals 1, and p(ID) always equals 0.

    But treating unknown possibilities that way is nothing more than a “no-design of the gaps” argument. It starts with ignorance with respect to the unknown possibilities and moves to an assumption that if NDE is not the answer, then UP is. It does so with no knowledge: “unknown” means unknown, after all. That’s not very impressive reasoning, and in fact I can’t think of any scientific or logical reason to accept it.

    It was after I wrote that, that I suggested treating p(UP) as a constant. I took two extreme possibilities: that p(UP) is completely and continuously adjustable to whatever p(NDE) might be, or that it is a constant; and showed that in either case my main point was supported: evidence against NDE is evidence in favor of ID, unless one moves into question-begging with some unknown possibilities or resorts to unknown-possibilities-of-the-gaps. Now, if you want to suggest a way that some intermediate treatment of p(UP) avoids those problems, I’ll be quite happy to hear you do so.

  22. Tom,

    One difference between “UP of the gaps” and God of the gaps is that the former’s probability is able to reach 1 (we’re now certain that the puzzle of missing solar neutrinos is explained by neutrino flavor oscillations), whereas the latter always forever remains a promissory note. The hypothesis of direct divine involvement in this or that aspect of nature is always in danger of being disproved by a not-yet-found purely natural explanation.

  23. Nick: My apologies for the incorrect attribution.

    Olegt: Belated apologies for the earlier snarkiness. The point still holds, though. By the way, what possible new “scientific” evidence/finding do you think would undermine the explanation of human blood circulation by means of the human heart as a four-chambered biological pump “by a not-yet-found purely natural explanation”? Have not the MESs achieved 100% certitude in that case? What about the scientific method itself? Do you think there is any possible new scientific finding that would undermine or even alter the scientific method, and if so on what basis do you say that? Are per se findings in the MESs or natural science as the human endeavor it is even able to say anything in this regard? Is the knowledge provided by the MESs fundamental or is it derivative knowledge? If the former, then on what basis is that claim made? If the latter, then surely some other form of knowledge is prior to the MESs? I don’t think there’s a single thing you’ve said in all these exchanges that indicates you have a sound understanding of this point.

    SteveK: … which is the same trap into which the ID theorists have partially fallen.

  24. Trap or not, Holo, the standard for what constitutes ‘science’ should apply equally and clearly that is not being done. That is perhaps the most frustrating part of the shell game being played here.

  25. olegt, you say,

    The hypothesis of direct divine involvement in this or that aspect of nature is always in danger of being disproved by a not-yet-found purely natural explanation.

    Well, of course that’s true. Does it discount the main point of my post here, I’m wondering? Does it mean that p(NDE) + p(UP) = k, where k is 1 or nearly 1, regardless of how some evidence E might affect p(NDE)? (I’ve already used “K” to designate a different constant above.)

  26. Tom Gilson says:
    January 25, 2010 at 12:12 pm

    olegt, you say,

    The hypothesis of direct divine involvement in this or that aspect of nature is always in danger of being disproved by a not-yet-found purely natural explanation.

    Well, of course that’s true. Does it discount the main point of my post here, I’m wondering? Does it mean that p(NDE) + p(UP) = k, where k is 1 or nearly 1, regardless of how some evidence E might affect p(NDE)? (I’ve already used “K” to designate a different constant above.)

    Actually, I will disagree with both of you here. Showing that there is a natural explanation for phenomenon X does not disprove a miraculous explanation. The supernatural could do anything, including imitate a natural process, suspend the natural process which would otherwise have worked, etc.

    All having a natural explanation does is render a supernatural explanation unparsimonious. But this is different than disproving it. This is part of why so many scientists (who are not New Atheists) take a strong line against considering as science projects that either try and prove or disprove supernatural explanations.

  27. Here is a short report on the IPCC and its version of “settled science”.

    The scientist behind the bogus claim in a Nobel Prize-winning UN report that Himalayan glaciers will have melted by 2035 last night admitted it was included purely to put political pressure on world leaders.

    Dr Murari Lal also said he was well aware the statement, in the 2007 report by the Intergovernmental Panel on Climate Change (IPCC), did not rest on peer-reviewed scientific research.
    In an interview with The Mail on Sunday, Dr Lal, the co-ordinating lead author of the report’s chapter on Asia, said: ‘It related to several countries in this region and their water sources. We thought that if we can highlight it, it will impact policy-makers and politicians and encourage them to take some concrete action.

    Read more: http://www.dailymail.co.uk/news/article-1245636/Glacier-scientists-says-knew-data-verified.html#ixzz0dem6fwec

    In another thread I linked to a speech by late author Michael Crighton in which he expresses concern about the direction “science” has taken in the last few decades. My own opinion is this has been happening since the early 19th C and Darwinism is a symptom as much as SETI and AGW.

  28. … which in turn give rise to disordered movements, myths and gurus such as transhumanism, secular humanism, progressivism, “spiritual” (!?!) naturalism, environmentalism not ordered to the truth of the human person, VHEMT, human “overpopulation,” Peter Singer, Steven “dignity is a stupid idea” Pinker, John Brockman, futurism, Ray Kurzweil, etc., etc.

  29. Nick,

    All having a natural explanation does is render a supernatural explanation unparsimonious.

    Scientists have a theory about what it means for something to be natural. BTW, that is a metaphysical inference from observed fact much like the inference to intellegence from observed fact. But I digress…

    If, in principle, ‘what is natural’ can’t fully explain something under the terms of its own theory, then that theory is incomplete. If the theory is incomplete then there must be more to nature than just nature. What scientists (and those who suffer from scientism) do is retool their metaphysical theory of what it means to be natural, and they file it under the heading of Scientific Fact.

    Naturalism is a theory that cannot be falsified under the rules of science – yet for some strange reason people think ‘natural’ is a scientific conclusion. It’s the reason why the naturalism of today looks more like theism than the naturalism of yesteryear. Someday I expect ‘natural’ to take on the qualities of necessity, agency, sovereignty, omniscience and omnibenevolence.

    Fortunately we all have minds and we can gain knowledge about reality from sources other than the methods of science. Fortunately again, most everyone knows this, which is why most accept the reality of supernatural explanations.

  30. Tom,

    I’m assuming that p(UP) is the probability that life is the result of a process(es) neither intelligent nor currently known or predicted by NDE.

    I wrote that I couldn’t understand why you use K instead of p(UP) in your formula. There is no reason to prefer some subset of values between 0 and 1 for the value of K, therefore p(UP) is a better term in the equation.

    p(NDE) + p(UP) + p(ID) = 1

    did you miss these paragraphs in the OP?

    That’s equivalent to saying p(NDE) + p(UP) always equals 1, and p(ID) always equals 0. It starts with ignorance with respect to the unknown possibilities and moves to an assumption that if NDE is not the answer, then UP is.

    Yes, the equation p(NDE) + p(UP) = 1 leaves out p(ID) without a valid reason (at least no valid reason was given in your post). Therefore, it isn’t correct.

    It was after I wrote that, that I suggested treating p(UP) as a constant. I took two extreme possibilities: that p(UP) is completely and continuously adjustable to whatever p(NDE) might be, or that it is a constant;

    Why would either of these be warranted? p(UP) is not inversely proportional to p(NDE) if we don’t know what p(ID) is, so “continuously adjustable” is out.

    But neither can p(UP) be considered a constant if we simply don’t know what value or range it can be beyond the end points 0-1.

    Now, if you want to suggest a way that some intermediate treatment of p(UP) avoids those problems, I’ll be quite happy to hear you do so.

    If we don’t know what values p(UP) holds beyond the extreme values of 0 and 1, it is best represented simply as p(UP). And the equation becomes

    p(NDE) + p(UP) + p(ID) = 1

    evidence against NDE is evidence in favor of ID, unless one moves into question-begging with some unknown possibilities or resorts to unknown-possibilities-of-the-gaps

    Is UP question begging? It seems to be that it is simply logical completeness. p(UP) is the probability that life is the result of a process(es) neither intelligent nor currently known or predicted by NDE. Is this an improper definition?

  31. woodchuck64, I think I’ve addressed all these questions already, and explained why I’ve taken the approaches I have.

    If I’ve missed something here, would someone please help me out with it?

  32. Regarding “Parsimony”

    No group is as guilty of misusing Occam’s Razor as atheists when applying it to the existence of God. In fact, in the face of the last century of scientific discovery, once would expect atheists to be downplaying the importance of O.R., rather than harping on it. Given all we’ve learned about the history of the universe, the arrangement of the cosmos, and the sophistication of matter, it would seem that whatever grasp atheism has on O.R. certainly isn’t on the handle. Taking God completely out of the picture requires the postulation of everything from fantastic coincidences to multiple universes to unobservable, one-time alterations in the laws of physics, and so forth. That’s all well and good, but that kind of unproven, un-testable conjecture smacks of the “faith” that atheists are so dismissive of. Old habits die hard, though, and O.R. is still a popular line of attack from critics of religion. It’s also worth mentioning that Occam wouldn’t have supported an atheistic interpretation of his own philosophy. He was a theist – actually, a Franciscan friar.

    The general application of O.R. by atheism is to say, “we can craft an explanation for such-and-such without mentioning God, therefore disbelief is more appropriate, as per Occam’s Razor.” However, this is not in keeping with the purpose of the principle. Remember, the guideline does not say, “fewer beings is better,” or, “any explanation without a God is better.” It says, “Plurality ought never be posited without necessity”. It’s the “necessity” part where interpretation comes into play. Just because I can explain how something happened without postulating a certain being’s involvement doesn’t make my explanation more likely by default. This is especially true when removing said entity from every explanation requires a mind-boggling number of assumed replacements. It’s inescapably true when the removal of said entity makes the end result impossible.

    Firmly By The Blade

  33. Tom:

    woodchuck64, I think I’ve addressed all these questions already, and explained why I’ve taken the approaches I have.

    If I’ve missed something here, would someone please help me out with it?

    You mean someone other than me? 🙂 Let me try again.

    p(NDE) + p(ID) + K = 1. If the probability of either non-K term is 1, then the probability of the other is 0; if the probability of either term increases or decreases, then the probability of the other term decreases or increases by like measure. A negative argument against NDE is still a positive argument for ID.

    According to my understanding, you have no basis for setting p(UP) to be a constant. The correct equation is p(NDE) + p(ID) + p(UP) = 1 and decreasing p(NDE) may increase either p(ID) or p(UP). Therefore, a negative argument against NDE is not necessarily a positive argument for ID.

    Your argument for using a constant seems to be this:

    What if we take a more balanced view, then? What if we admit the possibility of UP, and we do so in a way that avoid the “no-design of the gaps” error

    But the “no-design of the gaps” error was only in dropping p(ID) from the equation. As long as p(ID) is left in, the equation should be still valid. Therefore, I see no explanation or justification for replacing p(UP) with a constant.

    You seem to recognize this in the second to last paragraph:

    The number we choose really doesn’t matter, in view of the point I’m trying to make here, which is that a negative argument against NDE can be a positive argument for ID.

    I agree that a negative argument against NDE can be a positive argument for ID. But then in the final paragraph, you state it too strongly:

    a negative argument against NDE is still a positive argument for ID.

    As shown above, a negative argument against NDE may benefit either p(ID) or p(UP) but is not automatically a positive argument for ID once we use the correct equation.

    As always, I invite clarification if I’ve missed something obvious.

  34. Aha. We’re more in agreement than either of us realized, I think. The point of my discussing p(UP) was to show that unknown possibilities should not be taken as an escape from the conclusion that a negative argument against evolution can be a positive argument for design. Since you have agreed with that main point, we’re not really in disagreement here.

    But once again, why did I set up p(UP) as a constant? Often in dealing with unknown probabilities in philosophy, one has to recognize that one cannot reasonably set any probability as more accurate or supportable than any other. In light of that, one does the best one can. I did that by analyzing two extreme possibilities, for those who want to admit unknown possibilities into the equation. One of those was p(NDE) + p(UP) = 1, or nearly 1. That was the case where, whatever might happen to the uncertainty of NDE through the introduction of some evidence E, UP was always there to the rescue, and the probability of UP was infinitely fluid in response to the probability of NDE. I showed that this was question-begging and therefore illegitimate.

    But that was obviously not the only way UP could be treated. UP’s probabilities don’t have to be regarded by everyone as so being fluid. I didn’t have a way of assigning it some more viscous probability measure, because, well, there just isn’t a way to do that. Instead I examined the opposite extreme, where p(UP) was fully solidified as some constant K, and showed that in that case my main point was affirmed: evidence decreasing the probability of NDE increases the probability of ID.

    So I hope you understand now why I took the approach I did. As I have already said, if you think there is a better way to analyze the situation in light of unknowns, I’m certainly open to hearing it.

  35. Hi Tom

    I think you might want to adjust your terms a little. Division between NDE and ID and UP isn’t necessarily exhaustive and there may be considerable overlap – some ID proponents accept premises from the NDE hypothesis and some NDE proponents accept (at least implicitly) some ID premises (theistic evolution). \Evolution doesn’t exclude God\ etc.

    Something like \naturalism\ vs. \agency\ would dived teh premises more completely if we define \naturalism\ as purely electro/chemical/phyical forces acting according to the \laws of nature\; and \agency\ as the creative act of an intelligent agent.

    There would still remain a great deal of evidential overlap due to different interpretations of the various \facts\ but the causal end is pretty much divided between agent/not-agent and avoids multiplication of terms. But theistic evolution would (protests to the contrary) invoke agency and all forms of \naturalism\ would be classified as atheistic.

    Makes it a little more difficult to hide behind ambiguous terms.

  36. Nick (Matzke) wrote:

    Actually, I will disagree with both of you here. Showing that there is a natural explanation for phenomenon X does not disprove a miraculous explanation. The supernatural could do anything, including imitate a natural process, suspend the natural process which would otherwise have worked, etc.

    Nick, I don’t think we disagree on that point. I don’t think that science can disprove God. I said so previously.

  37. Nick, I don’t think we disagree on that point. I don’t think that science can disprove God. I said so previously.

    Fair enough, I also agree that science can’t disprove God.

    The specific question I was talking about though is whether or not the existence of a plausible natural explanation for a particular phenomenon “disproves”/”falsifies” a supernatural miraculous explanation. Many on all sides assume it does, or at least talk that way, but really it doesn’t, it just renders a supernatural explanation superfluous, which is different from falsification. It looks like you agree with this according to the post you linked to.

    Just to continue a bit…Notably, even direct observation can’t falsify a miracle, we all know of miracle stories where some people saw something miraculous, and other people saw nothing extraordinary, because God or whomever was making a special revelation to just certain people. The supernatural is an unconstrained explanatory hypothesis, but we have to assume numerous constraints (e.g. two people in the same place will observe the same thing) which miracles, even according to their proponents’ accounts, do not necessarily follow.

  38. SYNOPSIS
    Science, we are told, studies natural causes. To introduce design into science is therefore to invoke a supernatural cause and abandon science. Science deals with natural causes — mechanisms. Religion deals with supernatural causes — magic. Because they are hopelessly irreconcilable, science must keep design outside its purview.

    This distinction between mechanism and magic is flawed. The proper contrast is not between mechanism and magic or, alternatively, between natural and supernatural causes. The proper contrast is between unintelligent causes and intelligent causes.

    Intelligent causes can do things that unintelligent causes cannot do. Unintelligent causes can throw Scrabble pieces on a board but cannot arrange the pieces to form meaningful words or sentences. To obtain a meaningful arrangement requires an intelligent cause. Whether an intelligent cause operates within or outside nature (i.e., is natural or supernatural) is a question separate from whether an intelligent cause has acted. Intelligent causes are detectable. In fact, we have reliable methods for detecting them, and their detection involves no recourse to the supernatural. Affirming intelligent design is common, rational, and objectifiable and no magic is required.

    http://www.equip.org/articles/mechanism-magic-and-design?msource=EC100121WKLY&tr=y&auid=5835680

  39. An ID explanation of the Cambrian explosion is not ID. ID is not a theory about how historical changes like that took place. To even pose the question is to tell everyone that you’re misconstruing ID.

    ID is a theory about whether or not life was designed. It’s not even an argument against evolution.

  40. Tom:

    The point of my discussing p(UP) was to show that unknown possibilities should not be taken as an escape from the conclusion that a negative argument against evolution can be a positive argument for design. Since you have agreed with that main point, we’re not really in disagreement here.

    Okay, thanks for the clarification. The uncertainty, though, cuts both ways; a negative argument against evolution might be a positive argument for design but it could also turn out to be a positive argument for unknown processes. All things being equally unknown, it’s a 50% chance. Looking historically at arguments that relied on gaps in our knowledge and how well they fared, we’d be even more tempted to avoid such arguments. So it seems a better investment and approach all around to work on positive arguments for ID.

  41. I’m glad I finally succeeded in making sense. It takes me more than one try sometimes. I agree that positive arguments for ID are extremely important direction to go. And they do exist. I don’t know of any arguments in favor of these unknown possibilities we’ve been talking about, so I’m going to disagree with your line, “All things being equally unknown.” An unknown that is absolutely unknown is more unknown than a concept with an argument and a history behind it.

    At least we’re making progress, though.

  42. “A negative argument against NDE is still a positive argument for ID.”

    Respectfully, no. You have spent a long time trying to justify excluding the middle (that neither ID nor NDE are correct i.e. P[UP]=1). The reason that evidence against NDE always increase P(UP) rather than P(ID) is not because a person assumes that P(ID)=0, but rather because compelling logic dictates that it is so.

    Rather than being a single theory, there are an infinite number of unknown theories that are competing against our known explanations. Unless we can explicitly state properties these unknown theories have and show that they cannot possibly be a correct explanation for some phenomenon, than we cannot say that evidence against NDE counts towards ID. The logically honest thing to do is say that no current theory explains the information.

    Imagine that we apply your reasoning to a criminal trial. Does evidence that you did not commit a crime immediately implicate me? The equation that we can use to represent the probabilities that each of us commited the crime is exactly the same as the one you used in your post.

    P(TOM)+P(MATT)+P(Unknown)=1.

    Now, if I assert that evidence that you did not commit the crime is in no way evidence that I did, does that mean that I am asserting that P(TOM)+P(Unknown)=1 (i.e. “saying p(NDE) + p(UP) always equals 1, and p(ID) always equals 0.”)?

    Obviously not. Instead what I am saying is that, because there are so many people who could have committed the crime (roughly 6 billion), that evidence that any given person did not commit the crime has very little probative value. However, if it is clear that P(Unknown)=0. (Perhaps you and I were the only two people locked in a well monitored building.) Then, yes evidence that you did not commit the crime does implicate me. However, these are exceptional circumstances.

    If we look to see whether the situation with scientific theories is closer to that of an average criminal case
    or that of the exceptional case where we can be sure that one of two people committed a crime, then I think you will agree it is closer to the former.

    How do you know that probability that some other theory explains the facts is very small? How do you know that the list of other potential theories if very short? Unless you can justify these assumptions, then I think you will have to grant that evidence that one theory is wrong is not evidence that another theory is correct under most circumstances.

  43. Emanon,

    I’m not sure what you’re getting at, because this is just not at all what I was saying:

    You have spent a long time trying to justify excluding the middle (that neither ID nor NDE are correct i.e. P[UP]=1)

    Not even close. Maybe what you meant was this:

    You have spent a long time trying to justify excluding the middle (that P[UP] can equal approximately 1 under circumstances where we reject ID and NDE)

    But I’m not sure. Anyway, you go on to say,’

    Rather than being a single theory, there are an infinite number of unknown theories that are competing against our known explanations. Unless we can explicitly state properties these unknown theories have and show that they cannot possibly be a correct explanation for some phenomenon, than we cannot say that evidence against NDE counts towards ID.

    Sure. There are infinite theories to account for just about anything. How many good ones are there, though?

    Our theories tend to assume, for example, that where points of observation lie on a straight line, there is something linear at the base of reality that’s causing them to line up that way. But maybe we just caught the points that happen to lie on that line. Maybe the deep underlying reality is a sine curve, or something almost random (by the way: why is the universe not random, anyway?).

    Imagine that we apply your reasoning to a criminal trial. Does evidence that you did not commit a crime immediately implicate me?

    Yes. If you were the only other person who conceivably could have been on the scene, then evidence that I did not commit the crime is evidence that you did. You go on to agree with that.

    But then you ask,

    How do you know that probability that some other theory explains the facts is very small?

    First, note that Richard Dawkins has said that there is no naturalistic possibility other than the Darwinian one. That was the whole thrust of The Blind Watchmaker, and he said it again (as I noted in the OP) in Expelled. So I think he would say p(UP) = 0. If I were taking him as representative of the evolutionary position, I wouldn’t have had to introduce the UP idea at all.

    But I’m not taking Dawkins as representative of the evolutionary position, because I don’t think he’s a very good representative in this case, and I don’t want to try to get off that easy.

    But did I say p(UP) is small? No. I said that it is unknown. And then I analyzed two possible extremes with respect to that probability: one where it is continuous and infinitely adjustable with respect to p(NDE), and one where (on the other extreme) it is constant. I assigned no value to that constant. The first case, where p(UP) is continuously, infinitely adjustable with respect to p(NDE), begs the question, as I have shown above. It is a fallacious argument with respect to ID.

    The second case looks like this mathematically (I’m repeating myself, but apparently it’s necessary).

    p(NDE) + p(ID) + p(UP) = 1

    p(UP) = K, a constant.

    Thus
    p(NDE) + p(ID) = 1 – K

    1-K is also a constant. Let’s call it K1.

    Therefore

    p(NDE) + p(ID) = K1, or

    p(ID) = K1 – p(NDE)

    Which in English means that if some evidence E decreases the probability of NDE, the probability of ID increases.

    In sum, these two extreme treatments of p(UP) lead either to a fallacious argument or else to support for the position I have taken in this post. And they do so without assigning any numeric value to p(UP).

    You closed with,

    Unless you can justify these assumptions, then I think you will have to grant that evidence that one theory is wrong is not evidence that another theory is correct under most circumstances.

    I haven’t made the assumption that you think I made. If you can show that there is another way to analyze p(UP) in an intermediate (I used the term “more viscous” above) condition, and if you can show that this analysis invalidates mine, then you will have undermined my argument. You have not done that in this comment. I’m genuinely open, even quite curious, to see if someone will figure out a way to do that intermediate kind of analysis. You’re more than welcome to try.

  44. You alledge that I say:

    Imagine that we apply your reasoning to a criminal trial. Does evidence that you did not commit a crime immediately implicate me?

    Yes. If you were the only other person who conceivably could have been on the scene, then evidence that I did not commit the crime is evidence that you did. You go on to agree with that.

    Now, notice how you immediately have to add the caveat

    If you were the only other person who conceivably could have been on the scene

    Which is of course, precisely the question I wish to look at. You are saying that evidence of your guilt implicates me if and only if there are no other people who could have committed the crime (p(UP)=0, where p(UP) is the probability that an unknown person committed the crime). However, in your mathematical analysis you conclude that evidence of your innocence is evidence of my guilt when p(UP)>>0.

    Common sense should tell you that you have gone awry somewhere in your analysis. There is no particular reason that evidence of innocence should implicate me any more than any one else who may have committed the crime (unless you assume that P(U.P.)=0, which you explicitly do not do).

    Now let’s suppose that some investigator is fairly certain that Nathan Darwin Edwards committed a crime, so that P(N.D.E.) approximately 1. It follows that the probability that Isak Dinesen (I.D.) committed that crime is very low as is the probability that anyone else committed that crime. You would have us believe that if evidence came forward that completely exonerated N.D.E., then we would be nearly certain that I.D. committed the crime, since:

    1-K is also a constant. Let’s call it K1.

    And K1 is very small. If K1 is very small, and P(N.D.E.) is very small, then P(I.D.) must be very large, right?

    Which in English means that if some evidence E decreases the probability of N.D.E., the probability of I.D. increases.

    However, there is a problem. The exact same logic applies not just to I.D., but to every other person in the world. So if we look at the identical condition where

    P(N.D.E)+P(A.L.)+P(U.P.)=1

    by your logic we should conclude that if you did not commit the crime, then we are almost certain that Isak Dinesen committed the crime AND we are almost certain that Abraham Lincoln committed the crime AND …

    This is obviously wrong, so you must have made a mistake.

    Now, while it should be 100% apparent that your reasoning is mistaken at this point, still it would be nice to know what mistake you made.

    It should be fairly obvious to most people that P(UP) is not a constant. That is, when there is evidence that decreases P(N.D.E.) then it might increase P(UP). The reason that we prefer increasing P(UP) to increasing P(I.D)is because there are simply so many people who could have committed the crime, and unless we have some reason to prefer Isak over anyone else, then there is no reason to assume that Isak is implicated by Nathan’s innocence.

    Now let’s switch back from the land of carefully constructed analogies. You and I both believe in objective truth, so we know that one, and only one theory is true. Now granted, there are some shades of truth, so Newtonian mechanics can be wrong without being completely wrong, but ultimately there is only one truth.

    While we can test a known theory to ascertain whether it is true, it is impossible to test an unknown theory to ascertain it’s truth. Just try it: I’m thinking of a theory, is it true?

    What is worse than that, we cannot produce an exhaustive list of theories. if we could, then the work of science would be almost done; we could just systematically examine each theory in the list until we come to the true one.

    So we cant assign a value to p(U.P.) and we can’t assign a dimensionality to it either.

    So, because we don’t know what we don’t know, finding out that what we thought we knew was wrong does not force us to conclude that something we thought was wrong is actually right.

    You assert that the when the p(U.P.) is a variable then we ‘beg the question’ (i.e. assume the answer). However this is manifestly not the case. We do not have to assume that p(I.D.)=0 to assert that a decrease in p(N.D.E.) results in an increase in p(U.P.).

    Try this on for size: Assume that p(I.D.)=0.5, and that p(N.D.E.)=0.5, so that p(U.P)=0. This satisfies your equation. Then, evidence which decreases p(N.D.E.) arises. Does that imply that p(I.D.)=1. Clearly not, setting p(N.D.E.)=0, and p(I.D.)>0.5, there are a range of solutions to the problem where 0<=p(U.P.)<=0.5. So you see, quite clearly treating p(U.P.) as a variable does not beg the question.

    Far from it, not doing so begs the question.

  45. Tom Gilson wrote:

    But did I say p(UP) is small? No. I said that it is unknown. And then I analyzed two possible extremes with respect to that probability: one where it is continuous and infinitely adjustable with respect to p(NDE), and one where (on the other extreme) it is constant. I assigned no value to that constant. The first case, where p(UP) is continuously, infinitely adjustable with respect to p(NDE), begs the question, as I have shown above.

    Give it a rest, Tom. I gave you an example for which your assumption of a constant p(UP) is wrong. However much you dislike the idea of that probability varying in time, it happened in that particular case. You agreed to that.

    So, because your construction relied on p(UP) being constant in time, the “proof” failed. Back to the drawing board.

  46. olegt,

    What do I need to give up? I don’t need to “give up” the admission that p(UP) could be variable with time. I have acknowledged that from the very start.

    You say my construction “relied on p(UP) being constant in time,” but my construction examined the case where it was not constant in time. So I don’t need to give up on my construction relying on p(UP)’s constancy in time. I never depended on it. I did make a case for what would happen if it was constant, but since I also examined what would happen if it wasn’t, my case was not dependent on that constancy.

    So I don’t have to give up what you say I need to give up. What do I need to give up instead?

    I am also, by the way, perfectly willing to believe that UP’s probability is more viscous (see above) than either of the two extremes I’ve analyzed. I’ve been open to that from the start. What I have invited you and others here to do (more than once) is to present an analysis of how that would affect the case I have made.

    That’s what nobody has done here yet. You have pointed out that my assumptions regarding the probability of UP could be wrong, but in so doing, you misread the assumptions I was making. I wasn’t assuming what you thought I was assuming; I was running scenarios on “what if assumption x is correct? What if assumption y is correct?” You can point out that x or y is incorrect. You can say there is another possible scenario z. I have acknowledged that from the star.

    What you need to do to overcome my analysis is to counter the analysis. Show that there is something affecting the analysis based on some other scenario z that shows my scenarios do not do the job I have tried to make them do.

    And honestly, as I have said more than once, I would openly welcome that. I don’t mind if you prove me wrong on this. But just pointing out that there is another scenario z doesn’t prove me wrong—not unless you show how z invalidates the reasoning I’ve presented.

  47. olegt:

    I agree that Tom’s assumption that a non-constant p(UP) begs the question is the key error he makes.

    Tom: don’t take that the wrong way. I still like you, I just don’t know how to say someone is wrong without saying they are wrong. Perhaps I should post occasional ‘Ataboys’ instead of constant ‘Youarewrongs’. *sigh* oh well.

    On a mostly irrelevant note: I think that almost anytime that you publicly disagree with someone they will likely take umbrage to your points unless you bend over backwards to pad their egos. Almost every time I post on Panda’s thumb someone goes into a long rant about me be a creationist hack (though a number of people usually agree with my point). This is particularly hilarious since I am in fact a population geneticist. And I don’t mean I like population genetics. I have favorite enzymes (Pol IV, RAD14, UDG) for Pete’s sake.

    I think I just come across as abrasive on the internet; I will try to work on it.

    Peace be with you Tom. (Am I allowed to say that? You can’t possibly take offense as long as I mean it, right?)

  48. Emanon, in response to your 5:20 comment

    There are numerous logical flaws in your analysis, and several ways in which you have completely misread what I wrote. First, you write,

    However, in your mathematical analysis you conclude that evidence of your innocence is evidence of my guilt when p(UP)>>0.

    No, I didn’t state anything about what happens when p(UP) >> 0. My analysis is not based on values for p(UP) but on the variability of values for p(UP). (Is p(UP) constant, or does it vary continuously and fluidly in an inverse relationship with p(NDE)?) I hope you will do me the courtesy of going back and reading that over again.

    You say,

    ommon sense should tell you that you have gone awry somewhere in your analysis. There is no particular reason that evidence of innocence should implicate me any more than any one else who may have committed the crime (unless you assume that P(U.P.)=0, which you explicitly do not do).

    Common sense will fool you when the analogy you draw does not apply to the argument you are trying to analyze.

    For example:

    You would have us believe that if evidence came forward that completely exonerated N.D.E., then we would be nearly certain that I.D. committed the crime, since:

    1-K is also a constant. Let’s call it K1.

    And K1 is very small. If K1 is very small, and P(N.D.E.) is very small, then P(I.D.) must be very large, right?

    Can you find any place where I said K1 is very small? I didn’t. This is an example of your misreading the text.

    Bearing in mind that this is only half of my argument (I have also treated the case where p(UP) is not at all constant), let me show you what I mean. Let’s run a scenario where K1 is large. These numbers are purely for illustration, but where variables don’t get through, sometimes an example with numbers will. In this scenario let us set these probabilities:

    p(UP) = K1 = 0.9 (I’m exaggerating this for effect, just to show that my argument does not depend on K1 being a small number)
    p(NDE-initial) = 0.09
    p(ID-initial) =0.01

    The sum of their probabilities is 1.

    Now suppose some evidence E comes in that reduces our confidence in NDE. Mathematically that is expressed as a reduction in the probability of NDE. Suppose our confidence in NDE is reduced by 1/2 of one percentage point. The result would be:

    p(UP) = K1 = 0.9
    p(NDE-second stage) = 0.085
    p(ID-second stage) =0.015

    The evidence against NDE has increased the probability of ID, which means that it has functioned as evidence in ID’s favor.

    That is all I have tried to show in this post: evidence against NDE is evidence in favor of ID. (And please don’t forget that this is just one of the scenarios I have run.)

    Now, when you extend your analogy to “every other person in the world,” you are in effect taking some set of possibilities (a, b, c, …) and making them by sheer force of enumeration seem greater than UP was in the original post; but UP was already defined as the totality of all possibilities (a, b, c, …) You’ve just re-named the variable. That won’t get you anywhere.

    And then you get even farther off base:

    by your logic we should conclude that if you did not commit the crime, then we are almost certain that Isak Dinesen committed the crime AND we are almost certain that Abraham Lincoln committed the crime AND …

    First, I know enough to know that the sum of probabilities here cannot exceed 1. Second, I know that a zero probability (Abraham Lincoln committed the crime) is a zero probability.

    It should be fairly obvious to most people that P(UP) is not a constant.

    That’s why I addressed that in the original post.

    So, because we don’t know what we don’t know, finding out that what we thought we knew was wrong does not force us to conclude that something we thought was wrong is actually right.

    Of course. I didn’t say evidence against NDE proved (“forced us to conclude”) that ID was actually right. I said that it incrementally increased the probability that ID was right. Which addresses this further misreading you committed:

    Try this on for size: Assume that p(I.D.)=0.5, and that p(N.D.E.)=0.5, so that p(U.P)=0. This satisfies your equation. Then, evidence which decreases p(N.D.E.) arises. Does that imply that p(I.D.)=1

    No, it doesn’t. I didn’t say it did, and I didn’t hint that it did, and I didn’t imply it did.

    I’m about to look at your most recent comment to me. If your objection there proves to relate to something I have actually written, Emanon, it will be a significant step forward.

  49. Emanon, thanks for trying to keep the peace. I’m not taking umbrage, and I hope you aren’t. But I am trying to make my points clear. If I say M implies N, and you say, No, Tom, you’re wrong, L does not imply N, I’ll say (as I just did) I never claimed that L implied N. And I will insist that your reference to L does not have much at all to do with whether M implies N.

    Now as to this:

    I agree that Tom’s assumption that a non-constant p(UP) begs the question is the key error he makes.

    You misread again. I was really hoping for better.

    Let me try again. If you use your find-on-this-page search function you’ll surely be able to find where I’ve said this before.

    I ran two scenarios. One analyzes the case where p(UP) is a constant. I just went over that again with you in my last comment.

    The second scenario I analyzed was the one where p(UP) is infinitely adjustable with respect to p(NDE). In non-mathematical terms, it is the scenario where every decrease in the probability of NDE’s being true is considered an increase in the probability of some other, currently unknown materialistic explanation’s being true. It is the position that, no matter what, there is nothing that can increase the probability that ID is true; the only probability that can increase if there is any evidence against NDE is the probability that there’s some other materialistic explanation out there we haven’t thought of.

    That approach assumes that no matter what evidence can ever be adduced, it can only be evidence in favor of materialistic, non-design origins. It’s saying I don’t care what the evidence is or someday might be, we’ll get the same basic answer anyway: ID is wrong!

    If you can get to the same conclusion no matter what evidence exists now or might ever exist in the future, that’s certainly not basing your belief on evidence; it’s begging the question.

    So now in these last two comments I have re-analyzed my two scenarios. I have (shall I go back and count the times?) several times acknowledged that they are not the only two possible scenarios. p(UP) might be something in between the two extremes of absolutely fluid and absolutely solid. In fact, I’m sure it is.

    Maybe a better philosopher than I could run an analysis of what happens in the case where it’s moderately viscouse with respect to p(NDE). I will more than gladly welcome that analysis and learn from it.

    But this is not what either you or olegt have offered in your supposed refutations. You have only found various creative ways to say, “No, Tom, L does not imply N.” You have kept me busy replying repeatedly, “I never said L implied N; I said M implied N.”

    This is the last time I intend to reply to an “L implies N” attempt at refutation. I’d be glad to entertain an analysis of the viscous version of p(UP), but I’m done with re-explaining over and over again what I’ve already said. I hope you can understand why I would think enough of that is enough.

  50. Tom: Alright, you say “[If] p(UP) is infinitely adjustable with respect to p(NDE).” then you are begging the question, and if p(UP) is constant then a decrease in p(NDE) is an increase in p(ID). Since scenario one is clearly false and since scenario two directly leads to your conclusion, therefor your conclusion is correct (That “A negative argument against NDE is still a positive argument for ID”.)

    Does that seem to be a fair representation of your logic?

  51. Tom Gilson wrote:

    Maybe a better philosopher than I could run an analysis of what happens in the case where it’s moderately viscouse with respect to p(NDE). I will more than gladly welcome that analysis and learn from it.

    But this is not what either you or olegt have offered in your supposed refutations. You have only found various creative ways to say, “No, Tom, L does not imply N.” You have kept me busy replying repeatedly, “I never said L implied N; I said M implied N.”

    Tom, it was you who came up with this meta-theory. I pointed out a loophole in your reasoning, and you agreed that it exists. Now you are asking me to step in and present my own version of how this meta-theory should be fixed. I have no idea, Tom, and no interest in developing it. This kind of reasoning is completely unreliable for a simple reason pointed out many times: we can’t assign probabilities to such complex events as the development of a currently unknown scientific theory. This meta-theory of yours is fubar. Just drop it.

  52. Yes, that’s it in summary; bearing in mind that I’m open to being refuted in the case of a viscous version of p(UP). I would genuinely like to see someone run that analysis. The fact that my conclusion is supported in the case of the two extreme scenarios is suggestive of what might be true in the intermediate (viscous) scenario, but it’s not demonstrative, which I freely acknowledge.

  53. Well, Tom, why don’t you run the analysis in the “simple” case where the outcome is known. I’m curious to see how you would assign probabilities even in this case. I don’t think you can do it.

  54. olegt, I’ll answer that this way. The specific theoretic question you were speaking of ended up with a decrease in ID’s overall probability. It’s evidence against ID. I hope you weren’t thinking that one question’s being answered settles the whole question, or determines the future of all other evidentiary results—thought that’s what it seems like you were implying.

    But let me add this, more importantly. I woke up after an hour and a half of sleep, thinking again about the intermediate/more viscous version of p(UP), and I realized I really do have the whole analysis of p(UP) wrong. I’ve been treating it as if it were its own independent reality floating out somewhere away from human beliefs and opinions. It isn’t, and that changes everything. Some time tomorrow I’ll post a corrected version of the whole argument.

  55. One more thing: I should have recognized the real flaw in my argument sooner than I did, but olegt, your ordering me to “give it up” and to “just drop it” because it was “fubar,” besides being terribly imperious, was also not very logically impressive. That’s because the problem with the theory up until now has never been what you have said it was. When you told me to give it up (thank you for that order, sir!) I asked you what I should give up, and you didn’t answer. Except to tell me to just drop it.

    Your concern that “we can’t assign probabilities to such complex events as the development of a currently unknown scientific theory” turns out to be true (obviously) but also irrelevant, for reasons I will develop in a post yet to come. Hopefully this morning, but it may take until Saturday to get it written.

    In the meantime, I’ll ask you to cease the acronymic use of profanity here, please, and to bear in mind that telling someone else to quit arguing is no way to prove their argument fails. For someone who pointed fingers recently at someone else regarding the Starbucks Standard, it looks a lot like a Double Standard.

  56. Tom:

    There is no objective sense of a “double standard” for moral relativists and atheists–especially those who loath or intentionally relativize philosophy and theology as sources of knowledge. At a deep level they sense how morally and intellectually repugnant such a subjectivist position is, so they rely on all manner of deflection–trying to anchor or “locate” moral principles in society, individual preference, “science” (read: scientism), yadda-yadda to appease their conscience. This permits them the convenience of pursuing their own disordered (meaning: not ordered to truth) personal, subjective and quite proximate vision of morality without feeling any compulsion to “play by the rules” if it doesn’t suit them.

  57. Either an intelligent designer created life or not is fine, but evidence for or against evolution in no way influences whether or not God created life. Yes, p(ID) + p(!ID) =1 but there are innumerable theories. Let’s say there are 5 theories- aliens, wizards, God, evolution, and unicorns are 5 possible explanations for life. Any evidence for or against evolution in no way bolsters any of the remaining 4 theories. There is simply no evidence that wizards created life. And to cling to this false implication is just a mask for bolstering the belief in Jesus Christ as a magical superhuman. It’s religious thinking not Socratic thinking. It’s just not rational logic. P(ID) = P(Wizards) = P(Aliens) = P(unicorns). There’s equal proof of wizards, aliens unicorns, and God having created us.

  58. Karl, you’ve refuted an argument I did not make, nor would I; it’s a classic straw man. You have also made a rather surprising and unsupported (unsupportable, too) assertion that “there is equal proof” that we were created by God or alien unicorns. And you also did not read the comment guidelines….

Comments are closed.

Comments are closed.
Real Time Web Analytics