Talk:Fine-tuned universe/Archive 1

Could someone flesh this out a bit? It doesn't make any sense to me

However, the observation that no other forms of life have yet been observed in our universe allows us to think that indeed our universe is incredibly fine-tuned for intelligent life, and that it is a surprising fact that needs an explanation.

Roadrunner 03:02, 10 May 2004 (UTC)

One solar system is not enough to draw such conclusions from. You could equally well argue the evidence shows our universe is not optimised for life, since it only contains one example.

How many independant constants are there really? Some of them, such as the electron-proton mass ratio are not really fundamental. They could be calculated in principle, once we have the right GUT.

Also, are all the constants narrowly tuned? Weaken gravity a few million fold, and everything will still work. There would still be stars and planets, they'd just be bigger and last longer.

Really, we can't yet state the probability that a random universe would support life like ours, or indeed any kind of life, which makes declaring the constants to be fine tuned a little premature.


 * I understand your questions and concerns. Once you realize that this is not science, but at one intersection of science, philosophy and faith, you can better understand why consensus will be hard to reach. I believe that the article explains that.  I have not read the book of John Burrow, but I would suspect it is a good read to better understand this article.  If you read it, it would be great to give us an update.  Pcarbonn 20:58, 12 May 2004 (UTC)


 * Before consensus can be reached on why the universe is fine-tuned, consensus must be reached on if it is fine-tuned and, if so, the level of explanation at which it is fine-tuned.


 * This page needs more on both issues. I.e, suppose the ultimate laws of physics turn out to be such that a life-bearing universe is almost inevitable. That just pushes the question back one step, to why those laws of physics hold, and into deeper philosophical waters.


 * On the other issue, more needs to be said about the arguments for the universe being fine-tuned. Note, the universe's flatness used be cited as such, until we found it was an inevitable result of inflation.


 * Also, the diproton argument is full of gaps. There would be some nuclear reactions that occured by the weak force, and a different initial elemental abundance, which could leave us with, say, stars slowly fusing diprotons into lithium-4, even if reaction rate controlled stellar lifespan, which it doesn't quite. After all, in our universe the lifespan of stars varies a billion fold though the strong force strength is constant.


 * The core temperature of stars is just high enough for the reactions to be fast enough to create sufficient pressure to prevent collapse. A higher reaction rate means a colder stellar core, which could potentially mean a slower net burn rate and longer lived stars. Even if not, the smallest stars -the red dwarf equivalents- are going to live a lot longer than those the size of our sun, maybe by several orders of magnitude, maybe long enough for life.


 * In short, the argument that the physical constants are fine-tuned needs a lot of strengthening.

hello, its my first post in wikipedia. i dont think this article is un-biased towards non-scientific. its quotation regarding the 2nd law of thermodynamics is fishy and almost an effort of twisting a "reason" it so it could fit into the cause of the argument.

"fine-tuned" suggests "inteligent designer" indirectly.

The "How can we choose between the competing explanations?" seems unnecessary to me. It gives very generic information which is not spacific to this article, and really it should be assumed that the reader knows this anyway. This sort of thing belongs in a philosophy of science article.

Comment
I removed the section on Ikeda-Jefferys; it is written in a way to violate the neutral POV policy of Wikipedia. If it is rewritten so as not to violate this policy (words like "deeply flawed" etc violate this policy.) then it would be OK. One can neutrally present both sides of the issue, but must eschew POV language.

As to the points made above, changing "~N" to "I" does not rescue the counterargument. The reason is that one of the interpretations of "I" is that the designer is extremely powerful ("GOD") and is capable of suspending natural law at will. This means, for example, that it would be possible for such a designer to create universes where it is impossible for carbon to be cooked up in stars (the nuclear constants could be a little off, for example), but life could still exist. In such a universe, the constants would NOT be fine tuned for life, since there would be no naturalistic way for life to arise in the absence of carbon (this is the fine tuning argument). HOWEVER, a designer of the type I mention could create such a universe and just by fiat create the necessary carbon. Therefore, P(~F|I&L)>0 since such a universe is possible (probability greater than 0) and the Ikeda-Jefferys argument goes through unscathed.

I suggest that proposals for a section discussing the Ikeda-Jefferys article be posted in the discussion, not in the main article, until a correct and neutrally POV article can be crafted. But articles like the one I removed are NOT in accordance with the Wikipedia policy and should NOT be posted.

Also, please sign your comments in the discussion. This is both courteous and expected.

Bill Jefferys 15:08, 21 Jun 2005 (UTC)

I have restored the original three paragraphs.

I have deleted the following paragraphs because they are arguing a particular POV and are not neutral, as required by Wikipedia guidelines. They also misrepresent the Ikeda-Jefferys argument and thus are arguing a strawman. Here I explain what is wrong.

The first comment is as follows


 * The counterintuitive conclusion that this attempted proof generates (that a designed universe should be haphazardly organized) results from a flaw in the proof. Their argument supposes that life can exist in a non-life friendly universe. This assumption is impossible to prove (there is no "non-life friendly" universe for us to look at). Another way of looking at this is that the third assumption in the proof is that life will arise in a purposeless universe only if random chance creates life-friendly conditions) P(N|L) while life could occur in a purpose-driven universe even if it was not geared for life ("God could break the rules and make life not supported by the universe"). Simply put, the essential flaw in the proof is that it assumes that God would create a Universe where he would not follow the rules he set down (i.e. the proof allows for universes where life could exists in contradiction to established laws of physics within that universe).

This misrepresents the Ikeda-Jefferys argument. Ikeda and Jefferys do not argue that a designed universe should be haphazardly organized, and they do not argue that God would create a universe where he would not follow the rules he set down, only that God could do so. God, being omnipotent, can certainly do so. And if there is a possibility that God could do such a thing, then the probability that God would do such a thing is greater than zero. And that being so, the probability that God would not do such a thing is less than 1, which is the fundamental inequality in the Ikeda-Jefferys argument.


 * Of course, this logic is deeply flawed and results in a flawed conclusion. Using the same proof on Windows XP (Life) and a Computer (Universe):

This is the writer's personal opinion and should not appear in a Wikipedia article. The words "Of course, this logic is deeply flawed" are the problem.


 * Windows XP runs on a computer (L),
 * The nature of the computer is coducive to having Windows XP run on it; it is "Windows XP friendly" (F), and
 * Windows XP can appear without forethought in a "Naturalist" (N) Computer (lets say a monkey jumped on the keyboard a couple times to compile the machine code) only if that Computer is "Windows XP-friendly" (N&L &rArr; F), while Windows XP might exist with forethought both in a "Windows XP friendly" Computer and in another Computer even if its design is not conducive to making Windows XP work.


 * Besides making clear the flaw in the third assumption (how could Windows XP run on a computer that couldn't support Windows XP?), as we continue with the proof we get:


 * Given our computer runs Windows XP (L), the probability that our Computer was programmed by a monkey jumping on the keyboard, P(N|L), is less than the probability that our computer was programmed by a monkey jumping on the keyboard given that it is also Windows XP friendly, P(N|L&F) = P((N|L) | F). (That is, adding the assumption of a Windows XP supporting computer increases the probability that our Computer randomly was programmed by a monkey to be conducive to the running of Windows XP, given that we know our computer runs Windows XP.)  Thus, the proof argues ironically, that people who believe Windows XP is the product of forethought should try to prove this by trying to prove that the computer running it is NOT conducive to running Windows XP (i.e. this proof expects that a computer that runs Windows XP should have its machine code be programmed more poorly by an intelligent designer than by a monkey jumping on a keyboard).  Similarly circular and nonsensical results can be achieved with a number of substitutions to this formula.

This argument fails. One can certainly imagine that an omnipotent God could make a Macintosh run Windows. Even Microsoft can do it (and in fact, they have done it). Look, folks, God is supposed to be omnipotent. He can do anything, even in contravention of the natural laws he has made. Get it? According to many of the supporters of ID, it is on record that God did suspend natural law ("the laws he has made") on numerous occasions. People in the Bible walk on water. They are resurrected from the dead. The Bible is replete with such suspensions of natural law ("miracles"). Anyone who supports Biblical ideas of intelligent design can not claim that a life-containing universe must be "life-friendly" (as regards physical constants). Again, I am not saying that he will do anything, only that he can. And as long as he can, the fundamental inequality of the Ikeda-Jefferys proof holds.


 * In addition to this point, supporters of intelligent design (at the cosmic, rather than the organismic, level) might counter-argue that a finely tuned, life friendly, "naturalistic" universe could have been designed and brought into being by a "supernatural" entity or entities who would not need to intervene, thereafter, in the "naturalistic" functioning of his/her/its/their creation. Thus, the Ikeda-Jefferys Theorem does not drive a stake into the heart of "supernaturalism" in so far as the latter pertains to the creation, as opposed to the functioning, of the universe.

Sure, they can argue this. Note the word could. But by the same token, they might not. And if there is a possiblity that they might not, then the probability that they might not is greater than zero, and the probability that they did is less than one. And the Ikeda-Jefferys inequality holds.


 * Supporters of organismic intelligent design might further counter-argue that the challenge posed by the Ikeda-Jefferys Theorem requires revision because the challenge is based on the false assumption that fine-tuned life-friendliness cannot consistently be claimed for a universe not governed solely by "naturalistic" law.* The universe is more complex than the computer of our earlier example:  a computer is either Windows XP-compatible or it isn't; the "naturalistic" universe (for all its fine-tuned life-friendliness) may well be necessary but not sufficient for life as we know it. To meet a revised Ikeda-Jefferys challenge, the proponents of organismic intelligent design must demonstrate that our universe is not sufficiently fine-tuned/life-friendly to produce the highly complex organic structure and functioning which we observe. [*NOTE. Revised Ikeda-Jefferys assumptions might be as follows: (1) Our universe exists and contains life (L); (2) Our universe is sufficiently fine-tuned/life-friendly (F) to create and sustain life; and, (3) Life can exist in a solely "naturalistic" (N) universe only if that universe is sufficiently fine-tuned/life-friendly. The Ikeda-Jefferys Theorem remains unchanged:  P(N|F&L)>=P(N|L) OR P(~N|F&L)<=P(~N|L) where ~N signifies "supernaturalism." The assumption that our universe is sufficiently fine-tuned/life-friendly (F) to create and sustain life strengthens the thesis that our universe is solely "naturalistic."  Just as clearly, negation of the sufficiency assumption (~F) strengthens the intelligent design thesis that the universe is not solely naturalistic but requires "supernatural" agency to create and sustain life. {Some would argue:  "this still assumes that P(~F&L) (that life can exist in a non-fine tuned universe) is in the probability space, which is impossible to prove." But, in the context of the counter-argument, P(~F&L) means "probability that life can exist in a universe not sufficiently fine-tuned/life-friendly [~F] to create and sustain life." If the universe is N (solely "naturalistic"), the probability is 0. If the universe is ~N (not solely "naturalistic," wherein "intelligent design" could create and sustain life by supplementing the universe's insufficient fine-tuned life-friendliness), the probability is > 0. What remains an unresolved question is whether F or ~F is true. Some would also argue:  "if one replaces the N with I (Intelligently Designed) in the third assumption, one obtains an equally rational pro-design conclusion."  But, in the context of the counter-argument, I ("Intelligently Designed") = ~N ("not solely "naturalistic"). The proof of the Ikeda-Jefferys Theorem relies on the Weak Anthropic Principle: P(F|N&L)=1. The proposed substitution of ~N [=I] for N renders P(F|~N&L)<1, breaking the logic of the Ikeda-Jefferys proof.}]

Since the counter-arguments offered are based on a misrepresentation of the Ikeda-Jefferys argument, they do not belong here. BTW, the last paragraph illustrates the misunderstandings of the author. The Ikeda-Jefferys inequality is precisely P(F|~N&L)<1, thus showing this to be true does not "break the logic" of the argument, it affirms it.

Bill Jefferys 21:12, 21 Jun 2005 (UTC)

RE "Ikeda/Jefferys Argument"
Ikeda/Jefferys' assumptions N, F, and L can be restated in such a way as to preserve P(N|F&L)>=P(N|L) [Ikeda-Jefferys Theorem] and P(F|N&L)=1 [Weak Anthropic Principle] without requiring intelligent design supporters to argue "that our universe is not fine-tuned." Rather, they'll be able to argue "that our universe is not sufficiently fine-tuned/life-friendly" to be solely naturalistic. Just let: N = "Life can exist in a solely naturalistic universe only if that universe is sufficiently fine-tuned/life-friendly;" F = "Our universe is sufficiently fine-tuned/life-friendly to create and sustain life;" and, L = "Our universe exists and contains life." The Ikeda/Jefferys (IJ) Argument is very clever but, ultimately, is much ado about very little. Recommend that the IJ section be deleted from Wikipedia. -- Tom Walsh


 * Look around &mdash; Wikipedia has articles on unbelievably mundane topics. An argument regarding the nature of the origin of the universe, no matter how insignificant or invalid, is eo ipso (did I use that right?) notable enough for Wikipedia. Since we have a contributor to Wikipedia who knows more about this topic than I, I will let him respond to your rebuttal. --goethean &#2384;  20:11, 23 Jun 2005 (UTC)

Above, I demonstrated how Ikeda/Jefferys assumptions could be restated in such a way as to preserve the Ikeda-Jefferys Theorem and the Weak Anthropic Principle without requiring intelligent design supporters to argue "that our universe is not fine-tuned." Instead, they'll be able to argue "that our universe is not sufficiently fine-tuned/life-friendly" to be solely naturalistic. Justification for this last statement, using symbolic logic, follows.

Let:

N = "Life can exist in a solely naturalistic universe only if that universe is sufficiently fine-tuned/life- friendly [to create and sustain life]."

A = "Life can exist in our universe."

B = "Our universe is solely naturalistic."

C = "Our universe is sufficiently fine-tuned/life-friendly to create and sustain life."

THEN:

[1]   ~(B&#8729;C) &#8594; ~(A&#8729;B)   - From N -

[2]   ~(B&#8729;C) &#8801; (B&#8594; ~C)  - Elementary logic -

[3]   ~(A&#8729;B) &#8801; ~(B&#8729;A)   - Elementary logic -

[4]   ~(B&#8729;A) &#8801; (B&#8594; ~A)  - Elementary logic -

[5]  (B&#8594; ~C) &#8594; (B&#8594; ~A)  - Substituting [2] & [4] in [1] -

[6]        B &#8594; (~C&#8594; ~A) - From [5] by elementary logic -

[7] ~(~C&#8594; ~A) &#8594; ~B      - From [6] by elementary logic -

[8] ~(~C&#8594; ~A) &#8801; ~C&#8729;A    - Elementary logic -

[9]     ~C&#8729;A &#8594; ~B       - Substituting [8] in [7] -

Reference [9]: "A" is true because you are reading this. Design proponents have the burden of showing "~C" to be true, which (with A being true) will imply that "~B" is also true. -- Tom Walsh

This proposal does not rescue the ID fine tuning argument. Let's suppose that you are right and that it is possible to distinguish universes that are "sufficiently fine-tuned/life-friendly to be solely naturalistic" from those that are not. Call this kind of universe G. Then


 * P(G|N&L)=1

while


 * P(G|~N&L)<1

since supposedly P(~G|~N&L)>0.

So if we observe G, that the universe is sufficiently fine-tuned/life-friendly to be solely naturalistic (and all indications are that it is), then N is supported against ~N. The result is the same as with the I-J argument, only with a (slightly) different observation.

I don't see an obvious way to decide when a universe is "sufficiently fine tuned/life friendly to be solely naturalistic," but apart from this technical problem you haven't made a dent in the basic idea of the I-J argument, which is simply that since there are predicates (F, or maybe G, if we can recognize it) that are entailed by N&L but not entailed by ~N&L, it follows that observing such a predicate always supports N against ~N.

Given the simplicity of my math above, I suspect an error in yours. Sorry, I don't have time to find it.

Bill Jefferys 02:31, 26 Jun 2005 (UTC)


 * I've never claimed "that it is possible to distinguish universes that are 'sufficiently fine-tuned/life-friendly to be solely naturalistic' from those that are not." So far as I can tell, there is only one universe and this universe appears to be "sufficiently fine-tuned/life-friendly to be solely naturalistic."  But, appearances can be deceiving.


 * Few would disagree that our universe is sufficiently fine-tuned/life-friendly for organic chemistry to flourish on at least one planet. Still, the leap to life has yet to be explained adequately (or, better yet, to be duplicated in the laboratory).  If it can't ever be explained naturalistically, can we ever know that it can't?


 * Who would have thought that Gödel (or anybody else) could demonstrate certain truths of arithmetic not to be provable within arithmetic? Similarly, the genesis of life may be inherently unexplainable within a theoretical framework for a solely naturalistic universe.  Will intelligent design (ID) proponents come up with their own Gödel?  Hard to say!  What I can say is that the Ikeda-Jefferys challenge -- that ID proponents must show our universe not to be fine-tuned (at all), which is patently absurd -- misses the point entirely.  Still, it's an attention-grabber! --Tom Walsh

Nowhere do I say that ID proponents must show our universe not to be fine-tuned at all. Please do not put words into my mouth. I suggest only that this would be one way that they could accomplish their goal, and point out that showing that the universe is fine-tuned is logically not the right way to proceed.

ID proponents need to find a predicate, say H, with the property that P(H|N&L)<P(H|~N&L), and also show that H holds for our universe. This will be difficult, because ID proponents steadfastly refuse to specify the designer sufficiently so that the requisite probability calculations could, even in principle, be carried out. I point out that one way to do this would be for them to prove that the universe is ~F (or ~G), since then this inequality follows automatically (assuming that H is possible under ~N&L). But when ID proponents assert that F (or G) is true, they can only undermine their case, since it is obvious that P(F|N&L)=1 (also true if F is replaced by your G).

BTW I would avoid bringing Gödel into it. Gödel's theorem is purely a theorem of arithmetic; there's no reason to think that anything like it can be applied to the physical universe or to the origin of life (which, also BTW, has absolutely nothing to do with the origin of the universe. For some reason, Creationists and ID-ers are constantly conflating the origin and evolution of the universe with the origin of life, and the latter with biological evolution. One could have a designed universe designed so that life would arise within it naturalistically. One could have a universe designed so that life could not arise naturalistically but once started would evolve through organic biology to the diversity we see today. Etc. These are different issues and ought to be treated separately.) I find that many people who drop Gödel into the conversation don't really understand what he did. I don't say that this applies to you.

Bill Jefferys 18:00, 26 Jun 2005 (UTC)


 * You state (above): "it is obvious that P(FN&L)=1 ..."  What is not so obvious is the best interpretation of the symbols L, F, and N.  Of course, a valid interpretation must preserve the Ikeda-Jefferys Theorem and "P(F|N&L)=1."  Moreover, the best interpretation must produce a "P(F|N&L)=1" which hews closest to the accepted plain language definition of the Weak Anthropic Principle (WAP).  The WAP definition in Wikipedia emphasizes that "observed values of all physical and cosmological quantities" are naturally constrained to those values which are compatible with the existence of carbon-based life forms.  This constraint results in a solely naturalistic universe that is sufficiently fine-tuned/life-friendly to create and sustain life.


 * Now, take the interpretation of L, F, and N offered in the accepted, "POV-neutral," Wikipedia text for the "Ikeda/Jefferys Argument:" L = "Our universe exists and contains life;" F = "Our universe is life friendly;" and N = "Our universe is naturalistic." Contrast this rather anemic interpretation with the more robust one which I offered earlier (above) [bracketed words added for clarification only]: L = "Our universe exists and contains life;" F = "Our [solely naturalistic] universe is sufficiently fine-tuned/life-friendly to create and sustain life;" and, N = "Life can exist in a solely naturalistic universe only if that universe is sufficiently fine-tuned/life-friendly [to create and sustain life]."


 * Which of the two sets of interpretations of "P(F|N&L)=1" is closest, in spirit, to the plain language version of the WAP? Perhaps I can be forgiven for believing that my interpretation, while hardly the best possible, yields the more faithful translation -- one in which "P(F|~N&L)<1" obtains, but in which ~N does not mean that there is anything supernatural about the universe. Rather, ~N means:  "Life can exist in a solely naturalistic universe which is not sufficiently fine-tuned/life-friendly to create and sustain life."  I doubt that anyone would want to assert ~N.  Instead, "supernaturalists" would accept N (as I've defined it) and then would proceed along the lines that I suggested earlier (above) in my analysis using symbolic logic. --Tom Walsh

OK, Tom, I have now read your symbolic logic carefully and find no fault with it. However, I don't think it offers anything substantively different from the original I-J proof.

You define a new predicate N. Confusingly, your new predicate has nothing to do with what I-J call N (which tells us about the state of the universe), but instead is a piece of background information that is assumed by I-J, but not precisely stated in their probabiliity formulas. I will rename your N as my K (background knowledge) so as to avoid confusion with I-J's completely unrelated N. As you say, everyone should accept K.

Your A is I-J's L; I don't see any observational difference between "Life can exist in our universe" and I-J's L "Life exists in our universe," given that we are here. (L is an observation that everyone agrees to be true).

Your B is precisely I-J's N: "Our universe is solely naturalistic."

Your C is essentially I-J's F: "Our universe is sufficiently fine tuned/life friendly to create and sustain life" is substantively the same as "Our universe is life-friendly"; at least, that is the intention of I-J. I don't see that the adjective "sufficiently" adds anything, for the negation, ~F, says that the universe is not sufficiently life-friendly to create and sustain life, which is precisely what I-J mean by the universe not being life-friendly). I would point out that when I-J write "life-friendly" they mean naturalistically. So that may be a minor quibble.

Thus, what you've done in essence is to replace what I-J wrote with predicates that seem (to me at least) to be basically identical, and adding a new one (K) as background information to be accepted by everyone. One now has the likelihood function (in my revised notation, and explicitly writing K):


 * P(F|N&L&K)=1
 * P(F|~N&L&K)<1 [Here is where F has to include the condition "naturalistically"]

These are precisely I-J's statements P(F|N&L)=1, P(F|~N&L)<1, where the K predicate is understood but not written explicitly.

In your original notation, these two statements are P(C|B&A&N)=1, P(C|~B&A&N)<1.

From the first of these statements we derive P(~F|N&L&K)=0 [P(~C|B&A&N)=0 in your original notation], which, when run through Bayes' theorem, shows that if we find that the universe is not (sufficiently) "life friendly" (~F), then we get your conclusion that N (your B) is false (since L and K are assumed true). With much less algebra BTW.

I don't see that this revision adds anything substantive to the discussion. It seems to me that you've introduced new names for what are basically the same variables that I-J used, and introduced a new predicate which you've called N but which confusingly has nothing to do with the original N (your B). You've also included what seems to me to be an unnecessarily complex proof. I'd require a lot of convincement to think that this is an improvement. And the failure to mention "naturalistically" in the definition of F might be a problem if it's not interpreted right.

However, now that I understand your original argument, I would point out that your symbolic logic says essentially (and correctly) what I-J have also been saying: To undermine the notion that the universe is solely naturalistic, one must show that F (your C) is false. Showing that it is true doesn't undermine the proposition that our universe is naturalistic, it supports it.

Bill Jefferys 28 June 2005 20:25 (UTC)


 * Thanks, Bill, for acknowledging the soundness of my symbolic logic. It has never been my intention to attempt a disproof of the Ikeda-Jefferys Theorem.  What I do object to is the characterization of the Theorem as supposedly presenting intelligent design (ID) proponents with a devastating paradox, something like:  "You ID proponents must deny that the universe is designed/fine-tuned in order to save your thesis that the universe is designed/fine-tuned."  I'm not making this up!  The "POV-neutral" Wikipedia article on the "Ikeda/Jefferys Argument" contains this gem:  "Thus, [Ikeda and Jefferys] argue ironically, supporters of intelligent design should try to prove that our universe is not fine-tuned."  To date, you have offered no objection to this passage.


 * Let's review what you have said. In an earlier comment you wrote:  "Nowhere do I say that ID proponents must show our universe not to be fine-tuned at all.... I suggest only that this would be one way that they could accomplish their goal..."  Why would they, or anyone else (of sound mind), argue that "our universe is not... fine-tuned at all?"  Intelligent ID proponents would want to argue that "our universe is not sufficiently fine-tuned." But, in your latest comment, you go on to say:  "I don't see that the adjective 'sufficiently' adds anything..."  Apparently, you believe (with other intelligent "naturalists") that fine-tuned life-friendliness is both a necessary and a sufficient condition for life, so fine-tuned life-friendliness needs no qualification.  Intelligent ID proponents would only agree that fine-tuned life-friendliness is a necessary condition for life, and so would rely on the words "not sufficiently" to distinguish their position from yours.  By retaining the adjective "sufficiently," both sides can have their say.  "Naturalists" will claim that the universe is sufficiently life-friendly, by which they mean that fine-tuned life-friendliness is both a necessary and a sufficient condition for life. Intelligent "supernaturalists" will counter-claim that the universe is not sufficiently life-friendly, by which they mean that fine-tuned life-friendliness is only a necessary condition for life. What's wrong with that?


 * The lesson, it seems to me, is that the Theorem can only benefit by avoiding ambiguity in the definitions of symbols F, N, and L. If F, N, and L are defined wisely, there doesn't seem to be any point in introducing K (background knowledge). Here are my suggestions:  N = "Our universe is solely naturalistic;" L = "Our universe contains life;" and, F = [1] "Our solely naturalistic universe is sufficiently fine-tuned/life-friendly to create and sustain life" OR [2] "Our universe is sufficiently fine-tuned/life-friendly to create and sustain life naturalistically."  (I think that I actually prefer your F[2] to my F[1]...)  One last thought:  it would also help if you provided a clear explanation of why the Theorem matters. --Tom Walsh

Hello Tom,

I still don't understand the distinction you are drawing between "fine-tuned" and "sufficiently fine-tuned." Let me clarify by an example (which I have used before on this page).

Let us suppose that the constants of physics were found to be such that the ones that deal with chemistry, e.g., the fine structure constant, the ratio of electron to proton mass, etc., were such that life could arise and exist if the elements were all there, but that the nuclear constants, e.g., the strong force fine structure constant, were such that carbon could never be formed through natural processes in the universe (e.g., in the Big Bang or in stellar interiors). In such a universe, life could never arise unless the requisite elements were created some other way, and I presume in this example that the only way would be by fiat creation by the designer.

In your view, would this universe be fine-tuned for life, sufficiently fine-tuned for life, or not (sufficiently) fine-tuned for life?

My view is that such a universe is not fine-tuned for life, sufficiently or not, because in my view a universe would only be (sufficiently) fine-tuned for life if natural processes alone are sufficient for life to arise. But I would like your opinion, as I think that it would clarify things for me as to what you are driving at.

As to K. As I noted earlier today in a response to Timurghlu, I am actually warming to the idea of including K as an additional bit of backgrouond information. I'm now beginning to feel sorry that you want to dispense with it! Here's my reasoning. I think that a carefully stated K can actually clarify the relationships between the other predicates that we are discussing. I mean, for example, if we were to agree in the above example that the hypothetical universe that can't produce carbon through naturalistic processes (but which might be designed by a sufficiently clever designer) was "Not (sufficiently} fine-tuned for life" then we could stipulate that K includes that universes that are N&L entail that they are also F, but universes that are ~N&L might not be F. As you can see from the discussion with Timurghlu, he was worried that putting everything into F might "smuggle in" the conclusion with the hypotheses (he didn't use that term, but it's clearly what he was worried about). By separating K out as proposed here and in my response to Timurghlu, it might make it easier to see what the relationships are and to avoid the impression that the argument is "smuggling in" conclusions in an invalid way.

Why does the theorem matter? In my view, it is because ID enthusiasts (e.g., Hugh Ross, cited in the I-J paper on the web) are of the opinion that observing our universe to be "fine-tuned for life" strengthens the case for ID. This is wrong, it does not, as I-J say and Sober agrees. We go further than Sober (although I have pointed this out to Sober and he may revise his paper in the light of my comments in the future), in that we can imagine universes that are only slight variations of our own, and which might exist if a designer designed them, where the ordinary chemistry is the same but the nuclear chemistry different enough to preclude life from arising through solely naturalistic processes, thus requiring a designer. If such universes are possible, they have probability greater than zero, and so they steal some probability from universes that are completely life-friendly as I have defined it...so observing that our universe is completely life-friendly in my sense not only does not support the case for design, it undermines it.

Bill Jefferys 29 June 2005 17:49 (UTC)


 * Hello Bill,


 * As I see it, the universe you describe (which requires designer intervention to create and sustain life) is neither "solely naturalistic" nor "sufficiently fine-tuned for life." It is "fine-tuned for life," but only up to a point: it lacks the requisite nuclear constants to permit emergence of carbon-based life forms.  Some of the necessary conditions for life are present and some are not.  The hypothetical designer comes to the rescue by supplying the remaining necessary condition(s) -- perhaps, by tinkering with one or more of the nuclear constants -- so that, together, all the requisite necessary conditions are sufficient for life.


 * Now lets assume, for the sake of argument only, that this universe is our own. We observe that the nuclear constants now are consistent with the naturalistic production of life.  We believe in the WAP, so we believe (mistakenly, in our hypothetical case) that ours is a solely naturalistic universe.  The plot thickens:  one day, a theoretical physicist develops a model in which the strong coupling constant must have value X before time T (in the history of the universe) and value Y thereafter.  The change from X to Y seems to be entirely arbitrary. Before T, carbon-based life was not supported.  After T, carbon-based life was and is supported. What does our trouble-making physicist have to say?  Let's listen:  "Before T, our universe was not sufficiently life-friendly; after T, it was and is sufficiently life-friendly."  As a good scientist, he insists (mistakenly) that -- before and after T -- our universe was and is solely naturalistic, though he can't explain the seemingly arbitrary switch from X to Y. (He has nothing but scorn for "God of the Gaps" thinking. Our hypothetical designer gets away undetected with his/her/its meddling. Also hidden is the transition of our universe from "solely naturalistic" before T to "supernaturally enhanced naturalistic" after T.) In any case, the adjective "sufficiently" has served our hypothetical physicist well without compromising his commitment to a solely naturalistic universe.


 * As for K, I can take it or leave it. If its use promotes consensus, all the better! --Tom Walsh

An anonymous author added back material already deleted. This material continues to present a non-neutral POV and is thus unacceptable by wiki standards. I discuss these problems below:


 * The counterintuitive conclusion that this attempted proof generates (that a designed universe should be haphazardly organized) results from a flaw in the proof.

It is this author's opinion that there is a flaw in the proof. I disagree, there is not a flaw in the proof. Since there is a disagreement on this fundamental point, the comment is not POV-neutral, but is advocating a position, in violation of wiki policy.

This is why the author's opinion is wrong:


 * The third assumption is that life will arise in a purposeless universe only if random chance creates life-friendly conditions) P(N|L) while life could occur in a purpose-driven universe even if it was not geared for life ("God could break the rules and make life not supported by the universe"). Simply put, the essential flaw in the proof is that it assumes that God would create a Universe where he would not follow the rules he set down (i.e. the proof allows for universes where life could exists in contradiction to established laws of physics within that universe).

This is not what I-J claim. They do not claim that God would create such a universe, only that he might. Since God is by assumption omnipotent, he certainly can and might do such a thing. Therefore the probability that he would do such a thing is greater than zero. Since the entire thrust of the author's attempt to show a flaw in the I-J theorem is based on this misconception, it is wrong, and is beating a strawman.


 * As discussed however, this line of reasoning still centers on the assumption (played out with ciruclar results in the above proof) that God would not author a universe with working set of physical laws (or conversely, that God created a multiverse with innumerable "magical" worlds without working physics and ours, with working physics).

This has the same flaw. I-J do not claim that God would not author such a universe, or that God created a multiverse with innumerable "magical" worlds, etc. I-J claim that an omnipotent God could do such a thing (remember what omnipotence means), and that therefore the probability is greater than zero that he would do such a thing.

Dear anonymous correspondent: Let us work this out in the talk page. As long as you continue to misrepresent the I-J argument in the main page, as you have been doing, it will be removed. We can discuss the particulars of the arguments here on the discussion page, but let's keep the main page accurate as to what the I-J argument actually is, and not misrepresent it as you have done now at least twice.

BTW, why do you remain anonymous? You know who I am, and I don't hide my identity behind an IP number. Courtesy demands that you come out of hiding.

Bill Jefferys 23:02, 26 Jun 2005 (UTC)

An anonymous author wrote:


 * Becuase it attempts to separate God and Law, some have seen the Ikeda/Jeffrey argument as a classic case of Affirming_a_disjunct.

The I-J argument is a Bayesian argument. Classical fallacies (fallacies of deductive argument) such as affirming a disjunct do not necessarily apply to inductive arguments. For example, if one has A&#8594;B, and observes B to be true, this does not show that A is true (a classical fallacy of deductive argument). However, if one has A&#8594;B, then (as Jaynes points out in his classic book: Probability Theory: The Logic of Science) observing that B is true makes it more probable that A is true. This is classic scientific induction, as used every day by scientists of every kind and across the board. We predict that B will be observed to be true, if A is true. We observe that B is true. Therefore, we conclude that it is more probable that A is true than we believed before. Our confidence that A is true is increased (but in general does not achieve certainty).

BTW, the I-J argument is of exactly this type. Let A=N&L, and B=F. Then observing F makes N&L more probable, and since we know L to be true, it makes N more probable.

I would be very interested in seeing how you get this fallacy out of the I-J argument. I looked at the wiki page you pointed to, and I don't see it.

And keep this to the discussion page. This comment isn't yet ripe for the main page.

Bill Jefferys 23:36, 26 Jun 2005 (UTC)

I-J entry
Hello, I was the original author of the changes to the section Ikeda/Jeffries critique (though it appears there have been additions, subtractions, erasures, reposts since then). What form of critique would satisfy all parties for posting on this article? Would a simple removal of verbs ("circular" "deeply flawed") be enough?

You are right that you have to remove all language that expresses a POV about the I-J argument, and the above would be a good start, but it's not enough.

You also have to represent the argument you are criticizing accurately, and you have not done so. If you say that I-J is flawed because their argument "assumes that God would create a Universe where he would not follow the rules he set down", then this is wrong, since that is not what they assume. If you destroy an argument on that basis, you have not destroyed the I-J argument, but another argument. This is called knocking down a straw man, and it is a blatant fallacy of argument (one of the worst IMO).

I-J assume only that it is possible that God could do this. This means that God might do it, but does not have to do it. Thus, they presume that the probability that God might do it is greater than zero (not that it is equal to one). For example, we do not see people walking on water, and we do not see people being resurrected from the dead, and according to what we know of natural law, these things (and many other Biblical miracles) are impossible in our world. But it is claimed by Christians, many of whom are ID-ers, that these things, clear violations of "the rules that [God] has set down" (at least as we understand them from physics) actually happened, because that is what God wanted. According to Christian theology, God can (but does not have to) suspend the rules any time he wants to, even at whim for no discernable reason (look at the Hebrew Bible for numerous examples of God's unpredictable and arbitrary suspension of natural law). God is omnipotent, according to this theology, and he can do anything, even violate the laws he has set down.

Now, if your point is that God would never create such a universe (where natural law is sometimes violated), then I think you have an issue with the Bible. The Bible clearly says God has done such a thing. So if this is your position, I think it is incumbent upon you to show that no possible God would ever do such things.

I suspect that the problem we are having is that you don't actually understand the argument I-J actually made, but have a mistaken view of it and are arguing against that.

My suggestion to you is that you post your proposed additions, in small pieces (no huge edits please) so that they can be assessed easily, and not on the main page, but here on the discussion page, where they can be properly critiqued. If the proposed additions are right, I am intellectually honest enough to say so. If they are wrong, I will point out where I think they are wrong. My purpose here is to end up with an accurate article that presents all sides of the issue in a neutral, POV-free way. I am not interested in "winning" an argument. I want the truth to be presented honestly and fairly. That's the purpose of WikiPedia, and I hope that you can join with me in this effort.

Once we have a fair expression of what you're trying to present, then I'll be happy to support your posting it on the main page. I think this is much better than having an edit war, don't you?

Bill Jefferys 01:50, 27 Jun 2005 (UTC)

Hi Bill,

Just want to set some commong ground:

-The article in question is concerning cosmological fine-tuning and whether the structure of constants in the universe indicate a designer (i.e. this article is not about evidence of intelligent design through supernatural intervention but through the structure of natural laws).

-The I-J conclusion being proposed (correct me if I am wrong) is that a fine-tuned universe has a higher probability of being made randomly than on purpose.

In the interim, I will repost the Windows XP proof (with minimal commentary) and remove the paragraph on supernaturalism.

Point 1: The question of a supernatural designer cannot be divorced from the intelligent design question in general. This is first and foremost because the actual purpose of the ID movement has nothing to do with science and everything to do with religion and politics. The ID movement tries to pretend that religion has nothing to do with it, so that they can sneak this past the Supreme Court, but this is a deception and a lie. They say that they make no assumptions about the nature of the designer, but in reality, they think that the intelligent designer is (wink wink nudge nudge) the Christian God. So let's cut the crap. Intelligent design has everything to do with supernatural intervention.

Now I will point out that some in the ID movement (e.g., Michael Behe) have said that the intelligent designer might well be space aliens. Of course, Behe doesn't believe this, and it is for him just a posturing point to pretend that religion has nothing to do with ID. Remarkably, there are folks who think that the intelligent designers are space aliens, and who have enthusiastically, and much to the embarrassment of the ID movement in general, embraced the idea. These are the Raelians. If the sincere Christians who follow the ID movement in hopes that through it they might get God back into the science curriculum were to learn of the Raelians, or were to consider that the intelligent designers might be plural (thus polytheism), they might think twice. Indeed, a polytheistic group of multiple designers is on the evidence more likely than a single designer...since this way one could more easily explain the apparent contradictions in the world (e.g., the Problem of Evil).

So you can't divorce supernaturalism from ID, and when you do you run into (Christian) heresy. Not a pleasant situation, eh?

Point 2: You still don't understand the I-J argument. It is purely an argument that asks, how should your view of N versus ~N change if you were to learn certain information, namely F, that you did not know before? The answer is that, since P(F|N&L)=1, N cannot be undermined, and since there are good arguments (which I have detailed) that P(F|~N&L)<1, ~N will be undermined.

Point 3: I have removed the Windows XP analogy, as it is not at all analogous to the I-J theorem, for reasons I will explain below. I have restored the section on supernaturalism since it is relevant (see above). This section does need editing for improvement, but it is basically an accurate description of the Occam's Razor aspect that underlies this entire issue.

The reason why your Windows analogy is not apropos can be easily seen. You define


 * 1) Windows XP runs on a computer (L)
 * 2) The nature of the computer is coducive to having WindowsXP run on it; it is "WindowsXP friendly" (F)

Item 3 is irrelevant.

It is obvious that with the definitions above, L&#8594;F. If XP runs on a computer, the nature of the computer is conducive to having XP run on it (obvious, eh?) If L is true, then F must be true. This means that P(F|L&anything)=1. This in turn means that learning that F is true cannot change your opinion about anything, regardless of what anything happens to be. For example, if anything=N (your N, my N, any N) then from the above P(F|L&N)/P(F|L&~N)=1 and the posteror probability of N, whatever it is, remains unchanged when you learn that F is true.

Thus, your example is not analogous to the I-J theorem.

For an example of a well-written neutral POV article, look at the article on Intelligent design. Instead of saying "this is right and that is wrong," it presents an objective statement of what the ID folks say and an objective statement of the criticisms it has received. It does not present an opinion that one side is right and the other wrong.

Again, I urge that before you make changes to the main page, you run them past this talk page. Since the arguments you've posted have either misrepresented the theorem you criticize, or have presented counter-arguments that aren't analogous or relevant, what you've done so far is simply to cause a lot of editing activity. It would be much more efficient to discuss it on the talk page, find something everyone can agree is a neutral POV and accurate statement, and post that.

Bill Jefferys 28 June 2005 12:35 (UTC)

Hi Bill,

I didn't mean to offend you be reposting the XP example, though I still hold that the argument doesn’t survive a challenge by substitution.

I still want to start this discussion on common ground so:

-First off, I want to make clear what your conclusion is. In the argument you say: “the presence of fine-tuning actually argues against such design”. So, you are saying that a designed universe should be less organized than a randomly created universe? I don’t ask this with incredulity, I just want to get this fact on the table.

Second, I noticed that you changed your argument over the last month. Your first postulation was:

1.	our universe exists and contains life (L), 2.	our universe is "life friendly" (F), and 3.	life can exist in a "naturalistic" (N) universe only if that universe is "life-friendly" (N&L ⇒ F), while life might exist in a "designed" universe even if it is not "life friendly"

Your new postulation is: 1.	our universe exists and contains life (L), 2.	our universe is "life friendly" (F), in that its conditions are compatible with life existing naturalistically, and 3.	life can exist in a "naturalistic" (N) universe only if that universe is "life-friendly" (N&L ⇒ F: the Weak Anthropic Principle).

So, the big change is in the second step, which states: our universe is "life friendly" (F), in that its conditions are compatible with life existing naturalistically. In the argument, you also mention that “In this context, "naturalistic" is taken to be synonymous with "not intelligently designed”. So, to make clear, your second step says:

“Our universe is “life friendly” (F) in that conditions are compatible with life existing without design.

This not only says nothing about life-friendliness as it relates to physical laws but it also seems to be stating your desired metaphysical conclusion in the proof. Remember, the argument is talking about cosmological fine-tuning (or it should be if it’s in this article), so second step should be about the connection between cosmological fine-tuning and the existence of life. A more neutral second step should say:

“Our universe is “life friendly” (F) in that the laws of the universe exist in such a way as to support life.

This is not my main concern with the argument, but it’s an issue.


 * NOTE: Bill didn't change the statement of the postulates; I did, to define (F), which had previously gone undefined.  This is not circular (it says "compatible with", which does not imply naturalism); in fact, it was supposed to prevent the argument from being circular:  if you define "life friendly" to mean "supports life", then (L) automatically implies (F), by definition.  (i.e., a universe with life must "support" life in SOME way, or else the life wouldn't be there!)  It seems to me that what "life-friendly" means is that life is in some way compatible with "natural law" --- or, if you prefer, "physical law" (which of course does not exclude it also being compatible with supernatural law). Nurban 30 June 2005 14:33 (UTC)

I also noticed that you removed the links I had added (one with Nobel Laureate Charles Townes (http://www.berkeley.edu/news/media/releases/2005/06/17_townes.shtml), another to a fine-tuning article (http://www.firstthings.com/ftissues/ft0106/articles/barr.html), and a third to a bilbiography on design and fine-tuning (http://www.behacker.at/CHF). What was objectionable about these links? Their pro-design content?

Timurghlu

Hello Timurghlu,

I-J do not say that a designed universe should be less organized than a randomly created universe. They say that a designed universe could be such that life could arise in it despite the fact that "the constants are not right" for life to have arisen in it solely through naturalistic law. For example, the "constants might be wrong" for carbon to cook up in stars, so that carbon-based life could arise. If we observed that the universe contained carbon (because we, being carbon-based, exist in it), but found that there was no way that carbon could have been formed in our universe through naturalistic processes, either in the big bang or subsequently in the interior of stars, this would be a huge clue that the carbon in our universe could not have been formed naturally...which would point to a designer.

Please note: When you talk about "randomness", "less organized," etc., you aren't dealing with the issues raised. In the above example, the universe might well be perfectly orderly, it would just be that we would not be able to explain the carbon without resorting to extra-physical (not fully naturalistic) explanations. The issue isn't randomness, chaos or order, it is design versus the "apparent" design that would exist in an undesigned universe that contained life. The constraints on naturalistic processes that would produce life in a universe are so stringent, in other words, that it would inevitably appear designed (that is, highly orderly) even though no designer was involved. So I would suggest leaving aside considerations of "randomness" and "organization," as they are not germaine to the subject at hand and only make it harder to understand the actual argument.

Also note: I-J never say in connection with the argument that a designed universe "should," "must," or "will" be a certain way. It is certainly possible for a designer to design a universe with life that is life-friendly, etc., etc. But their point is that this is the only kind of universe with life in it that you could have if there were no designer, whereas it is possible for designed universes to exist with life in them that do not meet these criteria (for example, universes of the type I mention above). The words "might," "it is possible that" and so forth mean that there is some nonzero probability that we would observe features about our universe that would be incompatible with a purely naturalistic origin, and would thus lead us to think it was designed. This is the point of conditioning the probability statements on the one thing we know to be the case, namely, that we exist in the universe (L).

As to your second point, the fact that the universe is "life-friendly" in that life could exist in it without design does not in any way "smuggle in" our conclusion. The reason is that if life could exist in our universe without design, it could also exist in our universe with design. The thing is, it is possible to imagine universes where life could exist with design, but which are such that life could not exist in them without design. See the example above.

A way to make this clearer is to adopt some version of what Tom Walsh has suggested recently (and Tom, if you are listening to this, I have suddenly become warmer to the notion of introducing the predicate you introduced, which I called K into the formulas, just to clarify this point). Tom suggested something like this:


 * 1) The universe contains life [L]; everyone agrees this is true
 * 2) The universe is life-friendly [F] or not [~F] (Tom says "sufficiently" life-friendly, but I still think this is superfluous)
 * 3) The universe is purely naturalistic [N] or not [~N]
 * 4) A universe that is N&L must be F [K] (Tom asserts, and I agree, that everyone should assent to this)

Then the probability statements become


 * P(F|N&L&K)=1
 * P(F|~N&L&K)<1 and P(~F|~N&L&K)>0 (because, per the discussion above, non-naturalistic life-friendly universes do not have to be F)

The links you added disappeared by accident because I reverted to a previous version that did not have your Windows example but which coincidentally also did not include the links you added. I have no particular objection to your adding back those links (though I haven't yet had a chance to read the articles to see if they are appropriate...perhaps the thing to do is to separate and clearly label such links as being on one or the other side of the debate). But their disappearance was purely an unintended side-effect of the way I reverted to the earlier version.

I hope this helps.

Bill Jefferys 29 June 2005 15:25 (UTC)

Hi Bill,

I think we are getting closer to that common ground. Hopefully this hammers out the two points:

1) I see the point you are trying to get at (that by allowing God to “break” the laws of physics over an infinite possible range of universes, while allowing random chance to only follow those laws, the observable universe should not be designed as we see chance designing it). Another way of saying this is:

“Given the existence of an omnipotent designer, a universe that has life and finely-tuned cosmological laws is more likely to have been created by random chance than by design.”

Is this an adequate verbal summation of your argument?

2) Do you have a problem with changing the second step of the argument to:

“Our universe is “life friendly” (F) in that the laws of the universe exist in such a way as to support life (L).

If you do have a problem with changing the argument, what is the definition of Life-Friendliness (F) to be used? My proposal is above (that a universe is life-friendly if the laws and faculties of that universe are such that they support the existence of life we see in our universe (L)).

I will repost the links tomorrow, unless there is an objection.

Sincerely,

Timurghlu


 * “Given the existence of an omnipotent designer, a universe that has life and finely-tuned cosmological laws is more likely to have been created by random chance than by design.”


 * Is this an adequate verbal summation of your argument?

No, there's a problem here. If one assumes (given) the existence of an omnipotent designer, then (presumably) the universe couldn't have been created by "random chance" (but I don't like this term, it isn't what I mean by a naturalistically arising universe).

I think you need to avoid vocabulary that talks about randomness, etc. The problem is that "random" means things to a lot of people that it doesn't necessarily mean to a scientists. I would argue that natural processes have rules, and that they are therefore not just "random chance". This presumably could extend to the creation of universes (e.g., in inflationary cosmologies), as Smolin has argued. The fact that we don't have these rules worked out doesn't mean that everything is willy-nilly "random chance."

What I am claiming is rather different. I am claiming that, given that we exist (or that there is life in our universe), our observation that the universe has laws of physics that are life-friendly tends to support the hypothesis that it is naturalistic against the hypothesis that it is designed. By "support" I mean that whatever we thought before about the probable truth of these two alternatives, our belief that it is naturalistic should increase, and our belief that it is designed should decrease.

Note that if you already had a very strong belief that the universe was designed, you might still (on balance) believe that it was designed after learning this new information...but your belief would not be as strong.

Does this help?


 * “Our universe is “life friendly” (F) in that the laws of the universe exist in such a way as to support life (L).

This isn't quite what I mean. A universe could be life-friendly in this sense and still be of a kind that would not allow life to arise through natural processes. An example is my carbon-challenged universe. Such a universe can support life, but it could never get started without some help from the outside to get the carbon there in the first place.

Does this help?

OK for putting your links back. Do you want to think about separating the links into pro and anti subheadings?

Bill Jefferys 29 June 2005 18:13 (UTC)

---

A quick question: you state that ""naturalistic" is taken to be synonymous with "not intelligently designed.""

Does this mean (assuming God's omnipotence) that you believe that N is a subset of Intelligently Designed (i.e. Intelligently Designed laws that we observe acting in our universe), or that the two are mutually exclusive? I assume the latter (that N is "naturalistic" and ~N is Intelligently Designed). If so, why?

I will reformulate my "common ground" points tomorrow. Pro-and-anti subheadings are an excellent idea for the links.

Sincerely,

Timurghlu

Timurghlu,

Someone deleted your latest comments, but I have restored them because you raise a crucial point.

When comparing hypotheses, it is essential to enumerate only alternatives that are mutually exclusive and exhaustive. This means that one and only one of the alternatives listed is true. After all, we are trying to decide which of two (or more) alternatives is the actual state of the world. For example, if someone takes a medical test to see if they have a certain disease, say a test for fecal occult blood that is used to detect colon cancer, there are four possibilities:


 * 1) The test is positive (+) and you have the cancer (D)
 * 2) The test is negative(-, or ~+) and you have the cancer (D)
 * 3) The test is positive (+) and you do not have the cancer (~D)
 * 4) The test is negative(-, or ~+) and you do not have the cancer (~D)

The states of the world are D and ~D, either you have the disease or you do not. They have to be mutually exclusive and exhaustive, that is, one and only only one of them is true.

The observation is the result of the test (+ or -). You know what it is, and again, one and only one of these is true. Given this information, you have to infer whether D or ~D is more likely to be true. This is the essence of inference. One and only one of these alternatives is true, and we have to decide. My preference is for Bayesian inference, which Ikeda and Jefferys use in their article.

So, to go to our present problem, the states of the world are N or ~N. By N I mean that only natural processes are involved in our universe, from start to finish ("no design"). By ~N I mean anything that is not N, including any kind of designer, whether a Raelian type designer (space alien), a Hindu designer, a designer of the Vulcan religion (whatever that is), or the Christian God. It also includes the possibility of a universe that was "wound up" in the beginning by a Deistic sort of entity, and allowed to run since by its own power. It also includes a universe that was designed by multiple designers who had conflicting agendas (see for example Zoroastrianism). Any possibility that is not N is included.

You asked,


 * A quick question: you state that ""naturalistic" is taken to be synonymous with "not intelligently designed.""

That is pretty much my definition, since it seems to me that any possibility that is not "natualistic" must have some "intelligent design" involved (see above).

The observation is F. We observe (I presume, and people like [Hugh Ross] affirm) that our universe is constructed in such a way that the prerequisites for life (e.g., carbon) and the conditions of the universe (e.g., ordinary chemistry) are such that life could arise and flourish in it, without intervention by any outside agent ("designer"). This does not mean that a designer didn't do it, only that the designer would not be necessary for us to see what we see.

Under these conditions, I claim that if you observe F to be true, then whatever you thought about the relative merits of N and ~N as defined above, your opinion should be changed to think that N is more likely to be true than you thought before, and ~N less likely.

Bill Jefferys 30 June 2005 00:23 (UTC)

---

Hi Bill,

A quick overview of the common ground points:

Point One: The argument’s conclusion is: “Under the three listed conditions, if you observe F to be true, then whatever you thought about the relative merits of N and ~N as defined above, your opinion should be changed to think that N is more likely to be true than you thought before, and ~N less likely.”

Point Two: Definitions of our variables –

L = Life that we see in our Universe

F = A specific summation of the lawful interaction of the constants of the Universe which allows for the development of the Life we see in our Universe.

N = A Universe not Intelligently Designed.

-We could say that L is the result which develops when the lawful relationships between the constants of the Universe equals F. Yes?

I also want to set the “terms” of the debate. If I can show that Prob(~N&L|~F) = 0, will you agree that the argument does not reach the “point one” conclusion? I believe I can do this with only one very reasonable assumption. If I can do this, I believe it is fair to post the assumption (acknowledging why the assumption may not be true) and a discussion of the ensuing consequences for the argument (all on this page first, of course).

Sincerely,

Timurghlu

No, P(~N&L|~F) is not relevant to point 1. L is known to be true, so it can ultimately lie only on the right side of the conditioning bar in our calculations. One has by the conditional probablity formula P(~N&L|~F)=P(~N|~F&L)P(L). Since L is known to be true [so P(L)=1], what you've written is equivalent to the statement P(~N|~F&L)=0, which says that if out universe is not fine-tuned, and contains life, then it is not naturalistic. While this is true, and I have pointed out repeatedly that it is true (remember, this is what I have been telling ID fans that they ought to do...show that our universe is ~F and you prove that it is ~N), it has no bearing on what we should think if we find that F is true (the universe is fine-tuned). So no, I do not agree that if you show that P(~N&L|~F) then one does not get to point 1. This formula has no bearing on point 1.

BTW I tied myself in knots for a year trying to manipulate the formulas. Manipulating the formulas (such as putting L on the left of the bar) won't get you anywhere. Believe me, I have tried. [I'm just trying to save you from wasting your time]. Basically, probability theory is probability theory, and no amount of manipulation can destroy point 1 provided that the assumptions of point 1 are correct. If you wish to show that point 1 doesn't hold, you have to work on the assumptions, not on the probability formulas.

One can of course redefine F more carefully. In principle, there is some region in the space of physical parameters that allows for life, and some region that does not allow life. Presumably these two regions are mutually exclusive and together comprise all possible combinations of parameters. One corresponds to universes that are F and one to those that are ~F. Of course, we don't know what the precise boundary of this region is, but in principle one could calculate it (if we knew enough about "life, the universe, and everything" (TM).

You are almost correct in your question about how L arises. L can arise in any sort of universe if F is true; but L can arise in a solely naturalistic universe only if F is true. It's a subtle point, but it is crucial to the discussion.

Bill Jefferys 30 June 2005 12:36 (UTC)

Hi Bill,

P(~F|~N&L)= 0 is the point in question then. The question is whether life can exist outside of fine-tuning (i.e. life can exist only with fine-tuning (irregardless of Chance or God).

Here is the assumption that I will make, which you can feel free to disagree with:

Given an omnipotent God, God is a God of Law. He designs the laws that then result in the reality, and he governs the reality through the laws he has created.

The contra is: God is not a God of law. He does not design laws, but rather rules reality directly.

Why is this so important? Because “Life” as described in the argument is a resultant reality and Fine-Tuning as described in the proof is the set of causal laws that creates the resultant reality.

Using a very simplified example, we can say that a given Universe is a collection of constants, each with a unique value. So in our universe the constants would include Gravity and Water Density, and in our Universe, lets say that Gravity = 1 and Water Density = 1.5 with a simple linear relationship connecting the two.

Life is the resultant reality that occurs when the relationships of the constants in the universe equal F, lets say that Life exists when all the constants equal 10. So, F = 10.

Because Life only arises when F = 10, Life can only exist in a Universe where F = 10. The only way that it could exist outside of this is if God either introduced a new law (discussed next) or was not a God of Law.

Cosmological Fine-Tuning is not the fact that F = G(1) + WD(1.5) + etc… = 10, but rather that life can exist only when F = 10. This also means that Life does not arise solely from when the values of each constant of our Universe are at a unique point, but rather when the totality of their interactions results in a certain equilibrium (in our case 10).

Interesting question: could Life exist in places where Gravity = 2 (i.e. could a Cow live a non-Naturalistic world as we have defined it)? Yes, but this non-Naturalistic Universe would still be Fine-Tuned, and interestingly enough a scientist making the I-J argument would come to a erroneous conclusion about his Universe. Lets say that there exists a God-Created Universal variable “J”, which in our Universe is set at Zero (J = 0). Life is a result when the Universe’s constants are fine tuned, equaling 10. By setting J = -1 (assuming a strictly linear relationship for simplicity) Life would emerge. If we assume God is a God of Law, creating the causal reality which unfolds into the resultant reality, then Life could not emerge unless it was Fine-Tuned regardless what the specific makeup of the fine-tuning was (L will not occur when Gravity = 2 but when universal constants are at F (10 here) ). In this example, God is not acting outside of the law (or even revealing a heretofore unknown law though J could be exactly that), rather he is changing the causal variables which result in the resultant reality L.

Realize that making God a God of Law does not negate intervention as was argued in a prior post (hidden laws, analogous to cheat codes in a computer game provide a good example of interventionism by a God of Law) or the ability to modify constants (as shown above, it’s the resultant reality that counts).

But our result is that by definition, in order to arrive at the desired conclusion of L, the Universe’s causal reality must be set at F; While we see one set of relationships in our Universe, and God can create differing sets of relationships in other Universes, the result of those relationships must always be at F for L to occur. ( P(~F|~N&L)= 0  )

This will always be true unless (making the entire notion of laws moot since it is based on the belief on a law-governed universe) we say that God doesn’t work through laws, but rather can control resultant reality directly (life exists “just because”). I would argue however that it would be incumbent on you to explain “just because” in a bit more logical detail if you are going to use it as the underlying assumption in a logical argument.

The computer/XP example worked because it made this distinction. It makes sense to us because we ourselves, when we create, do so as lawmakers. As any programmer can tell you, you create the parameters and language that runs a program; you do not run the results of a program except through the underlying laws. Now, you can believe that God is not this way (contrary to our own experience with creating, and how we perceive the universe). I would point out however, that if you truly believe that God is not a God of Law (or at least at a minimum that the universe is governed by law which is comprehendible to the human mind), then science is nothing but an illusion.

This of course leads us back to what cosmic-fine tuning has to say about Intelligent Design in the first place. The conclusion is that Life can only exist in a fine-tuned universe. Period. The second step (not discussed in the I-J argument), is that the variables/constants/faculties in a Universe must have a set of relationships and values leading to (F) in order to create Life and assuming that each constant could have any number/relationship from 1 to infinity with equal probability, the probability that this universe (assuming no multiverse) got to F is very small (even with the addition of zero parameters such as J, they still have to be calibrated properly to get to F), and since this is the reality we have around us (looking at this evidence alone) it is very reasonable to assume an Intelligent Designer.

Of course, in my experience, most people who do not believe in God have other concerns/experiences in which they place greater weight in than cosmological fine-tuning (bad experiences with people of faith, academic pressure, failures of religion to describe causal reality, etc...) but perhaps fine-tuning is a key that can set people searching for future answers to those questions instead of simply being content deriding current, inadequate explanations.

Sincerely,

-Timurghlu

Timurghlu,

It seems my most recent changes were deleted, or maybe I didn't do them correctly. So I will try again.

First, I made a mistake in my earlier posting. I read too hastily and based my conclusions on the notion that you were claiming that P(~N&L|~F)=1, whereas you were actually claiming that P(~N&L|~F)=0.

It is easy to show that your claim is incorrect.

For P(~N&L|~F)=P(~N|~F&L)P(L) by the conditional probability formula.

And P(L)=1 so you are claiming that P(~N|~F&L)=0. But then you are claiming that P(N|~F&L)=1 since N and ~N are mutually exclusive and exhaustive. But this says that if there is life in our universe, and if it is not "fine-tuned" then it is solely naturalistic (no design component). But this is wrong! If our universe contains life and is not fine-tuned, then it cannot be solely naturalistic. There must be a design component.

So your claim cannot be correct.

As for your claim


 * Given an omnipotent God, God is a God of Law. He designs the laws that then result in the reality, and he governs the reality through the laws he has created.

Once a God (or Gods, don't forget there could be more than one) is omnipotent, there is nothing that he (they) cannot do, even govern reality by breaking the laws. That's what omnipotent means. See the grue paradox for some insight into this. If there is more than one God, they can even decree contradictory laws. The God of the Bible breaks the laws...obviously. I have not seen you deal with this, and until you do, the rest of what you wrote in this long addition isn't convincing. It is just philosophical blather of the sort that people used in former times to "prove" that we would never know what stars were made of, what caused disease, etc. Useless.

I have already shown why your Windows XP example is wrong, why don't you address what I wrote instead of bringing up other arguments that ignore what I wrote? If you think that I am wrong, address what I wrote directly and simply.

BTW, I am intrigued by your "handle." On the surface, it appears to be "Son of Timur" (Timur-oghlu, Timuroğlu) in a Turkic or Mongolian language. What's the story? [Timur "the Lame" was of course a famous Mongol warrior and leader.]

Bill Bill Jefferys 1 July 2005 00:41 (UTC)

Alright people....
While the above discussion is all very intresting, its edging on a private conversation ala message boards. So, distilling it down:

Is the Ikeda/Jefferys argument pertinent and notable? Is the argument presented fairly and accurately? Is there an obvious counter argument that does not count as orginal research? Is there a notable counter argument? Is that counter argument presented fairly?

Leave debating the argument itself to usenet and e-mail.--Tznkai 1 July 2005 04:51 (UTC)

---

Hi Bill,

You are missing the point. Essentially I am saying that Life (as we have defined it) cannot exist outside of a Fine-Tuned state (as we have defined it), even for God. You state:

...[For God there is] nothing that he (they) cannot do, even govern reality by breaking the laws. That's what omnipotent means. ... The God of the Bible breaks the laws...obviously. I have not seen you deal with this.

Read what I wrote again. Summarize the argument I made if it helps.

In addition I would posit this challege to make my point clearer:

Describe an intervention into the universe that cannot be explained or described in terms of the creation, modification, or existence of an underlying law or faculty within that universe.

Try it; it is harder than it looks. Actually it is impossible (I forget what this fact is called in philosophy / philosophy of science terms).

Remember, omnipotence requires a framework over which to be omnipotent.

Sincerely

Timurghlu [Yes, Timurghlu, Turckic for Son of Timur. Timur is "steel" or more literally "the unbendable stuff underneath". When steel was introduced to Turks, their men could not bend it and they believed it was due to the Timur inside of it]. Timur-i-Lenk is the most famous historical personage with this name.]