In Faith and Philosophy (January 2007), Elliott Sober once again takes up the question of intelligent design (ID) theory, arguing that the minimalist ID (mini-ID) theory–despite what its proponents claim–has theological implications. Here is how Sober describes mini-ID theory:
…mini-ID theory, says only that the irreducibly complex adaptations that organisms possess were made by one or more intelligent designers (Behe 1996, 2005; Dembski 1995, 1998b, p. 15). The identities of these designers are not specified; maybe the vertebrate eye was made by a team of Extra Terrestrials or by a God who lives outside of space and time. The mini-ID theory does not deny that human beings have common ancestors with other species, nor does it insist that the earth is young, nor does it offer an explanation of the origin of the universe. The mini-ID theory differs from some earlier versions of Creationism by virtue of its modesty. (p. 72)
After discussing individual theorists’ personal motivations for holding mini-ID theory, Sober returns to the theory itself, arguing that when taken with a set of independent premises–some of which are independently supported, all of which are held by most mini-ID theorists–mini-ID theory implies the existence of a supernatural mind. Essentially, the premises are these: (i) for any x in nature, if x is an irreducibly complex mind, then x was created by an irreducibly complex mind; (ii) Some irreducibly complex minds are found in nature; (iii) causes precede their effects; (iv) the universe is finitely old. The implicit conclusion seems to be this: Since mini-ID theory, when conjoined with these independently supported premises, implies a supernatural conclusion and since supernatural conclusions cannot be empirically tested, mini-ID theory cannot be a scientific theory.
I have been thinking about how a mini-ID theorist might respond. Now, for the record, I’m agnostic on whether mini-ID theory is true. I just don’t know, and frankly I’m unconcerned with that debate (which, however, is not to say that I believe that its turning out true would be uninteresting! But that aside…). I’m not even really concerned here whether mini-ID turns out to be a scientific theory. My concern is more general: Does a theory’s implying a supernatural (theological) proposition ipso facto make that theory unscientific? I think not, and I’ve been toying around with a modal argument to that end. I thought I would see what FSPB readers think about it.
Let’s get some assumptions and terminology out of the way. First, I shall assume that Sober’s explicit argument is sound; that is, that mini-ID when conjoined with the four independent premises above, entails the existence of a supernatural mind. Second, I shall assume that the mini-ID theorist takes these four premises to be true. Third, I assume that the proposition ‘There exists a supernatural mind’ is a theological proposition (contrary to what others have argued) and thus that ‘God’ and ‘supernatural mind’ refer to the same entity, viz., God. (Therefore, I shall use ‘God’ to refer to the supernatural mind in question.) Finally, I shall assume that the Standard Model of particle physics is true.
Regarding modal terminology, I shall follow Alvin Plantinga in taking a possible world to be a maximal state of affairs (i.e., for any state of affairs A, a possible world either includes or excludes that state of affairs). A book B on world W is a maximal set of propositions, each of which is true at W. So, for any book B on any world W and for any proposition P, either P or ~P is a member of B. Next, I shall assume the truth of S5, which says that if a state of affairs is possible, then its necessarily possible. (A quick example. Take this sentence [let’s call it ‘P’]: ‘At 10:00 am on 21 June 2007, Joe met his girlfriend for coffee at the Starbucks on the Florida State University campus.’ If P is true at some world Wx, then at every possible world it is true that P is true at some world Wx. That is, if P is possibly true, then it’s necessary that P is possibly true.) Finally, by just ‘possible worlds’ (and its cognates) I mean metaphysically possible worlds, and by ‘nomologically possible worlds’ (and its cognates) I shall mean all and only those possible worlds in which the Standard Model is true; that is, (P) is true at all and only nomologically possible worlds:
(P) The Standard Model of particle physics is true.
So much for assumptions and terminology. To the argument!
There are two conceptions of God. On one conception, God has necessary existence, God exists in all possible worlds. Trivially, if God exists in all possible worlds, then the proposition (G) is true in all possible worlds:
(G) God exists.
Thus, for any book B on any world W, (G) is a member of B on W.
Now, consider that if God exists in all possible worlds, then God exists in all nomologically possible worlds. Trivially, if for any nomologically possible world WNx God exists at WNx, then (G) is a member of the book BNx on WNx. (Hereafter, I shall just use ‘BNx’ to mean “book on a nomologically possible world.”) But of course, then, every BNx would also contain (P); and if every BNx contains both (P) and (G), then all such books contain (P&G):
(P&G) The Standard Model of elementary particle physics is true, and God exists.
But if every BNx contains (P&G), then every BNx contains (P6G):
(P6G) If the Standard Model of elementary particle physics is true, then God exists.
In short, then, if God has necessary existence, then even the Standard Model implies his existence. But if any theory with a theological implication cannot be a scientific theory, then the Standard Model of elementary particle physics cannot be a scientific theory!
Now, let’s suppose the philosopher of science mounts a campaign against the necessary-existence conception of God. If God exists in any possible world, she says, his existence is merely contingent. In fact, let us suppose the philosopher of science convinces our mini-ID theorist that not only is God’s existence metaphysically contingent, it’s nomologically contingent: God exists at some nomologically possible worlds, but not all nomologically possible worlds. Heck, let’s even say that our mini-ID theorist assumes agnosticism as to whether or not the actual world is among those nomologically possible worlds at which (G) is true. Does (P) still imply a theological proposition? Yes.
One implied proposition is this:
(E) There exists a possible world Wγ such that (G) is true at Wγ.
If (E) is true, then (E’) true:
(E’) (G) is true at Wγ.
In fact, not only is (E’) true, it is metaphysically necessary. There is no metaphysically possible world at which (E’) is false. Clearly, (E’) is a theological proposition.
Now, suppose that Wγ is identical to the actual world. If so, then (A) is metaphysically necessary:
(A) (G) is true at the actual world.
If (A) is metaphysically necessary, then the book on the actual world contains both (P) and (A), and thus (P&A):
(P&A) The Standard Model of particle physics is true, and (G) is true at the actual world.
But, (P&A) entails (P6A):
(P6A) If the Standard Model of particle physics is true, then (G) is true at the actual world.
If (A) is true, then it’s metaphysically necessary; and if (A) is metaphysically necessary, then it is entailed by the Standard Model of elementary particle physics. But if (E’) is a theological proposition, then so is (A), in which case (P) entails a theological proposition.
If God exists in the actual world, then the Standard Model of particle physics entails that God exists in the actual world, a theological implication if ever there were one. If scientific theories cannot have theological entailments, then if God exists, the Standard Model is not a scientific theory. But surely we want to say that even if God exists, the Standard model is a scientific theory. Therefore, it must not be the case the scientific theories cannot have theological entailments.
In closing, our mini-ID theorist would likely find small comfort in forcing the philosopher of science into admitting that all scientific theories have theological implications should it turn out that God possibly exists, but doesn’t exist in the actual world. True. So, the mini-ID theorist might consider whether she can come up with a scientific methodology in order to discover whether or not the actual world is a member of the worlds at which (G) is true. The present argument is designed (if I may use that term) not to show this is possible, but only to show that the theological implications of a conclusion cannot determine whether or not her project is scientific.
- Joseph Long
My mind hurts. I read it twice and then read it slowly again. … I am still lost. As far as I could tell you eventually prove the definition that supernatural is above natural … which is proving a definition? It makes no sense but I was lost in the complexity you generated from the simplicity of the suppositions.
I am a theist by faith. I am a scientist by vocation. I am not an ID-ist. ID seems to me to be a new version of ‘the God of the gaps’ based on the assumption that complexity cannot be derived from simplicity. This assumption had been proven wrong on modestly complex systems through AI algorithms. Simple rule sets can be used to generate complex graphics, such as many unique tree patterns for virtual forests, etc. I have no trouble taking the big leap of logic and say that the Grand Unified Theory ‘rules’ can produce complex biological systems without requiring ‘external’ input into the universe.
Joseph, your argument is based on the technicality that any material implication with a true (necessary) conclusion is true (necessary).
But when people talk about how scientific claims shouldn’t have theological implications, they are not talking about material implication! That would be ridiculous, as your argument shows. But be charitable; there is a non-ridiculous interpretation of their claim that you would do better to address. (This concerns the more common-sense notion of “implication”, violated by material implication, that motivates “relevance logic” and the like.)
Vince,
Thanks for your comment. I don’t want to attempt proving a definition–that would be wrong-headed; nor am I arguing that the supernature is natural–that would be inconsistent. Essentially, I want to show that the sort of statement (theological, scientific, moral, etc.) a theory entails cannot be used as a criterion for distinguishing between theories that are scientific and those that are not. The reason is, I take it, there is one set of facts, whether natural, supernatural, moral, or whatever; every true proposition is true in virtue of corresponding to at least one of those facts; and every true proposition entails every other true proposition. Therefore, all moral truths imply scientific truths; all theological truths imply historical truths; etc. Therefore, that a theory logically entails a non-scientific (even theological) truth, cannot be used to show that that theory is not scientific.
Given your second comment, I want to allay any worries that I’m trying to support ID, or show that it’s true, or show that it could even turn out true. I am not trying to do any of that. My concern is just with how domains of discourse (scientific, moral, etc.) relate to each other given that there is one set of facts and with how we can distinguish scientific theories from non-scientific theories. I’m not at all trying to support ID.
Your argument rests on one of the paradoxes of strict implication. A necessary proposition is strictly implied by any and every proposition. Using Lewis’s fish-hook symbol:
(Necessarily p) -> (q -> p)
If God either exists necessarily or fails to exist necessarily we have a necessary proposition and hence a proposition that is implied by every theory, scientific or otherwise. A much simpler argument to the same result can easily be made using the material conditional without any modal locutions. These results are well-known.
As Richard points out, charity is needed here. These obviously aren’t the senses of implication that are at issue in discussions like this. A relevant argument about these issues using modality would have to take into account epistemic space. Before I learn that Cicero is Tully it is epistemically possible for me that Cicero is not Tully, but it is not metaphysically possible that Cicero and Tully aren’t identical, given that the names are rigid designators. The current and very hot debate surrounding two-dimensional semantics addresses these issues.
I agree with Richard; your post is lacking in charity. The issues here are obviously much deeper than your post would suggest.
Richard and Jared,
Thanks for the insightful comments. As Richard alluded to in your parenthetical closing statement, the question of relevance is indeed the crucial issue: Which properties must a proposition P have in order for it to count as being relevant, in this case, to contemporary science–that is, worthy of scientific investigation? Which properties are such that if P has those, then it counts as non-scientific (i.e., not relevant to contemporary science in the right sort of way)?
Given that relevancy is the crucial issue, let me address the logical point. As you both said, I do use ‘implication’ to mean material implication, but I’m using a slightly different technical aspect than what you mention (though, a technicality it is!): Necessarily, (P&Q) implies (P→Q); The distinction is important not for the logical structure of the argument, but for the relevancy bit. P’s being true is a step towards P’s being worthy of scientific investigation. (Consider that Phlogiston theory cannot be worthy–even though it implies the Standard Model [assumed true]–for the sheer reason that it’s false.) Therefore, I want to limit the discourse to the set of true propositions. So, the next question is: Given that (a) all the propositions in question are true, and that (b) the sort of propositions a P materially implies cannot determine P’s relevancy to contemporary science, what other criteria must P meet in order to be scientifically relevant?
Now, I think Jared’s epistemological move has plausibility, but how would it go? Here’s an idea: If justification for asserting P requires appeal to only empirically observable phenomena, then P is worthy of scientific investigation. I think, though, that mini-ID theorists would embrace this. They would likely say that our ability to distinguish between nature and certain artifacts is empirically observable; we can can attempt to quantify the distinguishing features of these artificts that give rise to our ability; and then use the results of this quantification to search for signs of intelligence in empirically observable biological systems. Everything is empirical.
Now, please hear me: Again, I’m not saying I think mini-ID is right or good or that the government or universities ought to spend money on such a program. I’m just saying what I suspect mini-ID theorists would say in response to the posited epistemological crition. And what they would say, I believe, shows the epistemological criterion is insufficient for determining relevancy. Therefore, I can only conclude that the criterion for saying that mini-ID is not a science has not yet been presented.
Thank you both for your comments!
Joseph,
There are two points at issue that your last reply runs together:
i) The idea that a theory having theological implications means that the theory is non-scientific
ii) The notion of implication that is at issue in (i)
Richard’s point wasn’t about relevance to science, it was about relevance logics i.e. it concerned (ii) rather than (i) [except indirectly]. Relevance logics give a different analysis of implication in an attempt to avoid the paradoxes and problems of implication as traditionally conceived by logicians. The examination of (i) in your post relies on a notion of implication that is notoriously problematic. It is clear that those who make the claim in (i) are not using material or strict implication, hence the claim that you weren’t being charitable to them.
You seem to have misunderstood what I said about epistemic space. My epistemological point was that the relevant notion of modality in such arguments is epistemic rather than metaphysical or nomological. My comment was not about a criterion for scientific justification. The idea is that epistemic possibilities may not collapse into metaphysical possibilities. There is no metaphysically possible world where Batman is not Bruce Wayne, but epistemic space is different. If someone knows mini-ID theory to be true, they have closed off all epistemic possibilities in which God doesn’t exist. But if someone knows the standard model to be true their epistemic space is still open as to the existence of God regardless of the metaphysical mode of God’s existence or non-existence. Issues relating to this are at the forefront of philosophical discussion thanks to the debate over the two-dimensionalism in the philosophy of language and mind. The issues involved here are subtle and hotly contested.
Here’s a quick follow up regarding the charge of uncharitability. (This is more directly relevant to Sober’s actual article in Faith and Philosophy, and therefore, to my mind, where the focus of the present discussion should be.)
I certainly do not want to be uncharitable, but I’m not sure in what way, other than through logical implication, Sober can move from the conjoined premises of mini-ID theory and the listed Four Premises to the conclusion that mini-ID implies a theological claim,viz., that God exists. As I understand everyone, all agree that there are no theological propositions within mini-ID theory, but while mini-ID theorists claim that their theory does not imply the existence of God, Sober is arguing that it does. So, assuming my understanding is right, it seems that (with regard to Sober’s argument here) it’s the mini-ID theorists who are using the less strict notion of implication, not Sober.
Perhaps, then, I can boil my point down to a dilemma: By the term ‘implication’, either Sober means strict implication, or he does not. If, on the one hand, he does mean strict implication, then mini-ID theory implies all true theological claims, but then again so do all true scientific theories; thus, mini-ID’s strictly implying a theological claim cannot be used to separate mini-ID from (the rest of?) science. If, on the other hand, Sober does not mean strict implication, then it is not clear in what sense mini-ID theory implies a theological claim–it is not clear that Sober’s conclusion follows from his premises.
So, the (non-rhetorical) question I put forward is this: What sense of ‘implication’ could be used to show that mini-ID theory implies a theological claim while, say, the Standard Model does not?
Again, thanks for the comments.
Thanks for the clarifications, Jared. Points well taken and interesting, to boot!
I wonder if relevance logics, rather than strict implication, will help Sober out if Sober–rather than the mini-ID theorists–is the one using the less strict notion of implication (as per my last comment). Perhaps so, if mere strictness of implication is not at issue: Sober’s ‘implication’ could be true in virtue of more worlds than is mini-ID’s ‘implication’, but Sober worlds could be more relevant (in the proper sense, whatever that is) than mini-ID worlds. Then, the question would turn on what is the criteria by which one world is judged to be more relevant than another. Obviously, in speaking of relevance logics, I’m speaking of things of which I’m ignorant.
Nonetheless, appealing to relevance logic sounds like a very interesting move indeed. It also sounds like Sober’s argument really rests on his use of ‘implication’. But Sober hasn’t told us how he is using the term, and it’s not obvious to me how is using the term. Must we have relevance logics to tell us? If so, then the sense of ‘implication’ at issue is a large not so obvious point, in which case it seems that until we are told what the relevent sense of ‘implication’ is, I wonder if Sober has made his case.
Yes, I think we’re making progress now. Sober effectively points out that, with mini-ID as a premise, we can (epistemically) infer theological conclusions that we would not be able to infer without it.
This differs from your Standard Model argument, because adding SM as a premise does not allow us to infer any *new* theological conclusions, beyond those that could just as well be established without it.
It is in this sense that mini-ID has theological “implications” whilst SM does not. The former, but not the latter, has epistemic import for theological claims.
This has been a very useful discussion, indeed!
Now, it seems to me that if the implication relation in question is in epistemic space, then a two-valued semantics for the implication relation will be understood in terms of known and unknown (agnosticism). (I’m going to assume a two-valued semantics even though it might be argued that knowledge comes in degrees. As far as I can see, nothing really hangs on this assumption.) Therefore, an implication relation between P and Q fails if (and only if?) it’s possible that one knows P and is agnostic about Q. And, as Richard and Jared have pointed out, this is precisely what we see with the SM example: Knowing SM allows for agnosticism about the theological claim (G), while mini-ID seems not to.
If all of this is right, it seems to me that the mini-ID theorist has three moves available to her: (i) deny my assumption that the mind outside spacetime is God; (ii) deny Sober’s premise that all minds in nature are irreducibly complex; (iii) try another tack in showing that mini-ID’s implying a theological claim is insufficient to make mini-ID a non-science.
Regarding (i), mini-ID theorists would have to show that it is possible to know mini-ID and be agnostic about what kind of mind the inferred mind is. Mini-ID, the theorist would say, is agnostic about whether the mind is God’s or the mind of some great computer programmer and that our spacetime is just the logical space of a computer program. Then, the mini-ID theorist who wants to argue that the mind is God’s, would need a separate argument; but that argument would fall outside of mini-ID on this move.
Regarding (ii), the mini-ID theorist would argue that it’s possible to know mini-ID and be agnostic about whether all minds in spacetime are irreducibly complex. Mini-ID, the claim would be, allows for the possibility that there are immaterial minds in spacetime that do not exhibit irreducible complexity; mini-ID is agnostic about whether it was ghosts who created the first irreducibly complex systems.
Moves (i) and (ii) deny assumptions I’ve been holding, and so fall outside the scope of this discussion. Therefore, I’ll not say any more about those.
Regarding (iii), the mini-ID theorist might accept my assumptions and Sober’s argument and argue that implying a theological proposition is insufficient to make a theory non-scientific. One way to do this is to posit a thought experiment: Assume for the moment that within the pineal gland of each person, God wrote in Hebrew, say, the statement, “Your name shall be [x]. Love, God,” where substituted for ‘x’ is the person’s name. Now, suppose that a group of neuroscientists discover this writing. Talk about making headlines! A new neuroscience lab is opened for the explicit study this phenomenon; bodies are exhumed, recent cadavers are searched, and the results are replicated. The inference is made: everybody has in his or her pineal gland the above statement with his or her name substituted for ‘x’. Then, the neuroscientists get together and publish a paper in Nature entitled, “The pineal gland proves God’s existence!” The mini-ID theorist then would ask, does the research standing behind the article fail to count as science? If not, the mini-ID theorist would continue, then while mini-ID might turn out to be false and ultimately fail, by anology it too counts as science.
The third move seems, to my mind, the best move available to the mini-ID theorist. The fear is the dubiousness surrounding thought experiments these days. Perhaps the mini-ID theorist could try to formalize the argument, how. I don’t know how it would go, but it might be an interesting project.
Anyway, again, this has been a really good discussion. Thanks for the thoughtful comments.
Yes, I agree with your suggestion that route (iii) is the way to go.
I think that ID is, as it happens, misguided (bad science). But this is a merely contingent fact, at most. If it could be shown that living organisms really did exhibit “irreducible complexity” that couldn’t be explained by unguided evolution, then wouldn’t that be solid empirical evidence for a cosmic designer? There happens to be no such evidence in reality; but it doesn’t seem like the kind of thing one could deny in principle.
This is interesting. It seems then that we have arrived at the conclusion for which I was arguing in the post, albeit by a more cirtuitous (and enlightening) route.
I wanted to deny the often stated–in Sober, implicit, I believe–claim that since supernatural conclusions cannot be empirically tested, mini-ID theory cannot be a scientific theory. We agree that there is an in-principle possibility of the right sort of irreducible complexity, the investigation of which could be scientific. If we’re right, then the italicized claim is false. This is exactly what I wanted to argue. (The “in-principle possibility” bit was the motivation for my describing things in terms of metaphysical and nomological possibility.)
Showing the falsity of the italicized claim is important, I believe. For, it shows two things: First, it shows that mini-ID is not just “Creationism light,” the latter of which, as I understand it, utilizes scriptural creation stories as empirical evidence. Second, and much more importantly, a great swath of the Intelligent Design debate is misguided: the debate should not be about whether mini-ID is a science. Yet a great deal of the debate–in academic journals, in popular books, in courts of law, in university auditoriums–just is about whether it is a science. Rather, the debate should be about whether mini-ID is true. Whether mini-ID is true is (contrary to what many philosophers of science and scientists writing popular books tell us) a scientific debate, not a philosophical one. And if our conclusion has shown this, then I think we’ve shown something of considerable importance.
[...] case with the piece submitted on behalf of Joseph Long on Florida Student Philosophy Blog titled God’s Existence and the Standard Model of Particle Physics: An Entailment Relation? Though obtrusive, Mr. Long deals with those self-same principles of the Timaeus: God and Particles. [...]
Higg’s field seems to me the equivalent of God’s breathing…
Nothing can exist in the absence of it.