Darwin, God and chance

One of the enduring objections to evolution of the Darwinian variety is that it is based on chance, and so for theists who believe God is interventionist, it suggests that God is subjected to chance, and hence not onmi-something (present, potent or scient). Darwin and his friend Asa Gray debated this issue in correspondence, and it ended up as the final pages of his 1868 Variation (below the fold). Effectively, Darwin argued that we cannot "reasonably maintain" that God intended for chance events that are useful to humans or to the species concerned. It is this that I want to discuss, following a line of argument made to me by Phil Dowe in conversation. Phil and I intend to make a talk out of this sometime, so this is my taking of useful notes. [Late note: Phil's views can be found here.]

First, let's look at the text in question...

... the long-continued accumulation of beneficial variations will infallibly lead to structures as diversified, as beautifully adapted for various purposes, and as excellently co-ordinated, as we see in the animals and plants all around us. Hence I have spoken of selection as the paramount power, whether applied by man to the formation of domestic breeds, or by nature to the production of species. I may recur to the metaphor given in a former chapter: if an architect were to rear a noble and commodious edifice, without the use of cut stone, by selecting from the fragments at the base of a precipice wedge-formed stones for his arches, elongated stones for his lintels, and flat stones for his roof, we should admire his skill and regard him as the paramount power. Now, the fragments of stone, though indispensable to the architect, bear to the edifice built by him the same relation which the fluctuating variations of each organic being bear to the varied and admirable structures ultimately acquired by its modified descendants.

Some authors have declared that natural selection explains nothing, unless the precise cause of each slight individual difference be made clear. Now, if it were explained to a savage utterly ignorant of the art of building, how the edifice had been raised stone upon stone, and why wedge-formed fragments were used for the arches, flat stones for the roof, &c.; and if the use of each part and of the whole building were pointed out, it would be unreasonable if he declared that nothing had been made clear to him, because the precise cause of the shape of each fragment could not be given. But this is a nearly parallel case with the objection that selection explains nothing, because we know not the cause of each individual difference in the structure of each being.

The shape of the fragments of stone at the base of our precipice may be called accidental, but this is not strictly correct; for the shape of each depends on a long sequence of events, all obeying natural laws; on the nature of the rock, on the lines of deposition or cleavage, on the form of the mountain which depends on its upheaval and subsequent denudation, and lastly on the storm or earthquake which threw down the fragments. But in regard to the use to which the fragments may be put, their shape may be strictly said to be accidental. And here we are led to face a great difficulty, in alluding to which I am aware that I am travelling beyond my proper province. An omniscient Creator must have foreseen every consequence which results from the laws imposed by Him. But can it be reasonably maintained that the Creator intentionally ordered, if we use the words in any ordinary sense, that certain fragments of rock should assume certain shapes so that the builder might erect his edifice? If the various laws which have determined the shape of each fragment were not predetermined for the builder's sake, can it with any greater probability be maintained that He specially ordained for the sake of the breeder each of the innumerable variations in our domestic animals and plants;— many of these variations being of no service to man, and not beneficial, far more often injurious, to the creatures themselves? Did He ordain that the crop and tail-feathers of the pigeon should vary in order that the fancier might make his grotesque pouter and fantail breeds? Did He cause the frame and mental qualities of the dog to vary in order that a breed might be formed of indomitable ferocity, with jaws fitted to pin down the bull for man's brutal sport? But if we give up the principle in one case,—if we do not admit that the variations of the primeval dog were intentionally guided in order that the greyhound, for instance, that perfect image of symmetry and vigour, might be formed,—no shadow of reason can be assigned for the belief that variations, alike in nature and the result of the same general laws, which have been the groundwork through natural selection of the formation of the most perfectly adapted animals in the world, man included, were intentionally and specially guided. However much we may wish it, we can hardly follow Professor Asa Gray in his belief "that variation has been led along certain beneficial lines," like a stream "along definite and useful lines of irrigation." If we assume that each particular variation was from the beginning of all time preordained, the plasticity of organisation, which leads to many injurious deviations of structure, as well as that redundant power of reproduction which inevitably leads to a struggle for existence, and, as a consequence, to the natural selection or survival of the fittest, must appear to us superfluous laws of nature. On the other hand, an omnipotent and omniscient Creator ordains everything and foresees everything. Thus we are brought face to face with a difficulty as insoluble as is that of free will and predestination. [From Darwin Online, Darwin, Charles. 1868. The variation of animals and plants under domestication. 2 vols. London: John Murray, Vol. 2, pages 430-432.]

Now Darwin's claim is that God cannot have intended, or acted by design, that an accidental or contingent outcome occur that happened to have beneficial outcomes. Elsewhere, in his letter to his niece Julia Wedgwood of 11 July 1861, he wrote:

Asa Gray and some others look at each variation, or at least at each beneficial variation (which A. Gray would compare with the rain drops which do not fall on the sea, but on to the land to fertilize it) as having been providentially designed. Yet when I ask him whether he looks at each variation in the rock-pigeon, by which man has made by accumulation a pouter or fantail pigeon, as providentially designed for man's amusement, he does not know what to answer; and if he, or any one, admits [that] these variations are accidental, as far as purpose is concerned (of course not accidental as to their cause or origin); then I can see no reason why he should rank the accumulated variations by which the beautifully adapted woodpecker has been formed, as providentially designed.

Darwin's argument is roughly this: unless we want to ascribe meaning to every random event, which seems to make God a micromanager of monumental proportions and undercut the raîson d'étre for there being natural law at all (and hence the foundation of science itself), we cannot maintain that God would have chosen the random variations on which both artificial and natural selection operate. Hence, natural selection is inconsistent with the notion that God is responsible for everything.

Some theists accept this, and do not take the Leibnizian view that God has created the best of all possible worlds, so cruelly caricatured by Voltaire in Candide. But Phil thinks that Gray has legs yet to run: and not even a Leibnizian God, but for a less omni- neo-Leibnizian God. This God has chosen merely the best world of which he knows, in which events that satisfy whatever utility function such deities have when creating worlds occur.

At first I was inclined to agree with Darwin. But something struck me from my studies on species concepts, from a discussion of Augustine's. Augustine is discussing the wording of the text in Genesis 1:10-11:

Then God said, "Let the earth sprout vegetation, plants yielding seed, and fruit trees on the earth bearing fruit after their kind with seed in them"; and it was so. The earth brought forth vegetation, plants yielding seed after their kind, and trees bearing fruit with seed in them, after their kind; and God saw that it was good.

Augustine asks

Where, then, were they [plants, when they were created]? Were they in the earth in the “reasons” or causes from which they would spring, as all things already exist in their seeds before they evolve [develop—JSW] in one form or another and grow into their proper kinds in the course of time? … it appears [from Scripture—JSW] … that the seeds sprang from the crops and trees, and that the crops and trees themselves came forth not from seeds but from the earth. [De Genesi Ad Litteram, c. 390 AD, Book V, chapt 4 (Augustine, Saint, Bishop of Hippo. 1982. The literal meaning of Genesis. Translated by J. H. Taylor, Ancient Christian writers; no. 41-42. New York, N.Y.: Newman Press: 151f)]

In other words, God created the "reasons" by which these things spontaneously arose from the soil. This is not, despite a few claims to the contrary, an evolutionary view; Augustine is talking about spontaneous generation of the plant kinds, a view that survived until the 18th century. In effect, and in medieval terms, Augustine is supposing that God created the "secondary causes" by which these kinds arose.

We can translate this into an evolutionary version: did God create the laws of nature (secondary causes) by which living things, and ultimately of course we ourselves, evolved? If one accepts that selection (and modern additions to Darwin's theory like drift and neutral evolution) are the motors of evolution, however, then this gets very hard to maintain, according to Darwin's argument.

But I suggested that there is a way the neo-Leibnizian God could ensure the outcomes without micromanaging. There is a simulation program called Avida, which simulates a process of evolution by selection on "animats" or cellular automata that simulate organisms in an environment. Suppose I want a particular outcome, and I have plenty of time - not an infinite amount, but enough to run simulations of "worlds" that will approach my utility function. After running several millions of simulations, I find the "universe" that serves my needs, and build it to the specifications drawn from that. If my needs include allowing a builder to make a house from rocks off a cliff, then I ensure that the universe I choose includes that outcome.

In this way, everything is the result of secondary causes, while the Neo-Leibnizian God remains the primary cause, including of each event that satisfies the God's utility functions. So Darwin is, I think, not correct in saying that one cannot reasonably maintain that the outcome is designed by the deity.

However, Darwin is entirely correct about the secondary causes - God is not a micromanager. Instead God has chosen the best available, or most satisfactory, world to create (note that this doesn't require that God is omniscient, or choose the best of all possible worlds; in fact it may even be that to choose that world God might have to abandon some of HisHer goals, unless the two are identical - that is, unless what God wants is the best of all possible worlds). So we have a difference of levels of causes here, not unlike the distinction made by some Neo-Thomists, between creation as the instigation of each individual event, which they rejected, and creation as the subsistence of every event as they unfold according to secondary causes.

Why does this matter beyond a bit of mental gymnastics, especially since I am not a theist? Well it has one rather significant implication: it means that those who criticise theistic evolutionists (like Asa Gray) for being inconsistent or incoherent are wrong: it is entirely possible to hold that God is not interventionist, and yet hold that God desired the outcomes, or some outcomes, of the world as created. In simpler terms, there's nothing formally wrong with believing the two following things: 1, that God made the world according to a design or desired goal or set of goals; and 2, that everything that occurs, occurs according to the laws of nature (secondary causes). In other words, it suggests that natural selection is quite consistent with theism, solving a problem I discussed earlier.

So we should not, as Dawkins and his fellow ideologues do, attack those who are religious and accept and even promote evolution. The primary cause explanation is entirely distinct from the secondary causes that are the domain of science. I don't for a second think that this is the way things are, but neither do I think that someone is just playing courtier's games if they think this.

More like this

199 years ago today, Charles Robert Darwin was born in Shrewsbury, Shropshire, England at the home of his family (known as the Mount). By pure coincidence, Charles would have published one of the most important books ever written 50 years later in 1859, and next year will mark not only the…
On June 30, 1860, T.H. ("Darwin's Bulldog") Huxley and Samuel ("Soapy Sam") Wilberforce met at Oxford to debate the concepts put forward in Charles Darwin's recently published On the Origin of Species by Means of Natural Selection, an encounter that is often celebrated even though the details of…
It's a season, so I am told, that has something to do with religion. We celebrate the birth of commodity capitalism, or something. So I thought I would combine my favourite issues - philosophy, religion and evolution. It's all Alex Rosenberg's fault. At a dinner before the conference, he was…
It has sometimes been said that the leaders of creationist ministries and advocates of intelligent design are charismatic, charming people who know how to play to the crowd. I don't believe it. Creationists are often just as loud, judgmental, and terse as the stereotype of evolutionary scientists…

Thanks for being . . . dare I say it . . . . rational on this topic. As a Theist, who is also an oceanographer and student of Darwin, I hope you will allow me to steal your argument from time to time.

I like your main argument. Very clever for someone with the lurgy ha ha (I really mean very clever simpliciter). But I don't get this bit:

unless we want to ascribe meaning to every random event, which seems to make God a micromanager of monumental proportions and undercut the raîson d'étre for there being natural law at all (and hence the foundation of science itself)

I don't see the conflict, in general (separately from whether there's a conflict in the specific case of evolution).

In philosophy of physics (as Phil knows, and maybe you do too), a lot of philosophers take it for granted that the block view of time is true, and also that there's no fundamental causation, and it more or less follows from that that there's no difference between managing on the large (e.g. managing the boundary conditions of the whole cosmos) and micromanaging (e.g. managing every particle interaction), modulo some room for indeterminism ... BTW, the block view is compatible with indeterminism, somethitng which is not always appreciated.

Now I'm not arguing that the above view is right; only that it's not obviously wrong. But it counts as micromanaging on your view, n'est-ce pas? (Hey, you said raison d'etre.) And yet it's not at all incompatible with natural law, is it?

The primary cause explanation is entirely distinct from the secondary causes that are the domain of science

Not only that, but it's worth pointing out that there's no way to ever know if multiple levels of causality even exist. To summarize an example I posted previously on this blog: there is no way for a population of intelligent software programs to determine the true nature of the underlying hardware they are running on (without revelations from the outside). There's a clear boundary where most types of causuality are restricted, and those that are not, are not epistemologically accessible.

Also if I'm not mistaken, the implicit assumption of universal cause-and-effect is not so universal even at our "secondary" causual level (the QM level) - not even considering block time (and block time not a view that I would take for granted).

How would an internal observer native to such a system ever distinguish the work of such a deity from the simple action of natural law, and thus come up with the notion of a Creator in the first place?

John I have a feeling that you are getting very close to Newton's god for whom Leibniz lambasted him. If god is the primary cause who sets up the secondary causes then once he has set up the system he is out of a job, redundant, surplus to requirements, superfluous, left over, in excess, unneeded, no longer of use... This as Leibniz pointed out is the start of the long slippery slope to deism or even worse atheism!

What jeff said. Doesn't your whole argument assume a completely deterministic universe, in other words, one without quantum physics?

By David Marjanović (not verified) on 05 Sep 2008 #permalink

I'm afraid I can't make heads or tails out of your argument.

First, what sort of God are you envisioning? Tell me something concrete about His attributes and His goals. I don't know what you mean by a Leibnizian God or a neo-Leibnizian God, and I definitely don't know what it means to describe God as “less omni.”

The conception of God that is said to be incompatible with evolution is the one who is omnipotent, omnibenevolent and omniscient, and who created the world specifically with humans in mind. That is also the conception of God that is typically defended by Christians, at least in my experience. If I'm understanding you correctly you are explicitly not reconciling evolution with that conception of God. Is that correct? If it is, then you really should give a more precise definition of what you mean by `theism.'

I also don't understand the point you are making in invoking the Avida simulation. I don't have a criticism to offer, it's just that after reading that paragraph several times I still can't fathom what you are trying to say. Is the idea that God had certain goals in creating the world, but was uncertain as to how to accomplish them. So he ran a whole bunch of simulations, found a set of initial conditions that would bring his goals to fruition, and then created the world according to those specifications? If it is, then once again I will need a clearer statement of the attributes of the God you are considering.

I also can not tell if you are following the example of Ken Miller and Simon Conway Morris in arguing that human-like beings were the inevitable result of the evolutionary process. If you are not arging that, then I don't see how you have laid a finger on anything Darwin said in his quote (which I agree with completely). If you are arguing that, I'd like to know the biological basis for thinking that humanity was the inevitable result of the evolutionary process.

You singled out Richard Dawkins for special criticism in your final paragraph. Can you tell me which part of your essay he has argued against? No one has ever claimed that there is no conception of God that is compatible with evolution. And few have argued that there is a logical incompatibility between evolution and traditional Christian theism. A sufficiently imaginative person can reconcile virtually anything short of X and Not X.

The fact remains that theists have to explain why a God who loves his creations and has humans specifically in mind from the start would do his creating by a mechanism as savage, brutal and unpredictable as natural selection, when presumably he could have poofed the world into existence in precisely the manner the Bible describes. I don't see how you've really addressed any of the important issues in this post.

I think this may be rather long, for which I apologize in advance.

Re Leibniz' objection as related by Thony C, if one allows for the 'multiverse' or 'landscape,' the Creator could well occupy Itself making universes ad infinitum rather than micromanaging one.

Re quantum mechanics as raised by several commenters, the probability mathematics are well worked out to a fantastic number of decimal places even by we non-omnisicient humans, so it's easily conceivable that a Creator could rig a/the Creation to tend 99.999etc% toward the desired outcomes. Granting this, there isn't necessarily any material practical difference between a clockwork and a quantum Creation, though of course there's a conceptual one.

Re causation, general and special relativity, and thus a good chunk of physics not yet explicable in terms consistent with quantum mechanics, still obeys causation, so we can't yet entirely dispense with it as a possible tool for a Creator.

Where the rubber really meets the road IMO is whether we will allow the possibility of an *uncreated* Creator at the Beginning of All Things. If yes, then ISTM we've got to throw out rationalism; if no, then ISTM rationalism reigns, but the Creator must obey natural laws and thus cannot be truly omnipotent.

Starting in on an explanation of this last bit: Cosmologists have already conceived of experiments in which one or more "pocket universes" are created. Of course we don't currently have the means to perform such experiments, but it is rationally conceivable that in time we or some other intelligent life forms could develop the means. So let's say hypothetically that *our* universe is the result of such an experiment. Then it would follow that we would have both a Creator or Creators and rationalism.

However, this doesn't solve the problem of Creation, it just changes the venue. How would the creators of our universe have come to exist, in this rationalist explanation? In accordance with natural laws, through evolution. If one is to explain the result - the universe in which we find ourselves - via rational means, then somewhere at the beginning of a chain of universes and Creators would be the quantum "blurp" that started the whole thing going along its naturalistic, rationally explicable path.

OTOH, if one allows for an un(naturally)created Creator, the Unmoved Mover, at the beginning of all things, then that's not conceivable within a naturalistic/rationalist framework, i.e., it's supernatural. Once you've allowed for that, there's no reason to say the supernatural couldn't happen again (and again and again), and thus you've done away with any requirement for the world to obey natural law and any principled objection to a micromanaging Creator.

Where the rubber really meets the road IMO is whether we will allow the possibility of an *uncreated* Creator at the Beginning of All Things. If yes, then ISTM we've got to throw out rationalism;

I'm not entirely sure about that. If there is no universe. no physical laws, no time, space, matter, etc, then philosophically (and rationally), what is there to prevent anything from "being"? For nothing to "exist" is a logical contradiction. You may argue on the basis of parsimony, which is a valid argument IMO, but the response then becomes: what does parsimony mean outside of a universe, where there are no laws that we are accustomed to?

In discussions of this type, the real question inevitably becomes, "what is absolutely fundamental"? And how can I know what is fundamental without a context to consider it in? If something is absolutely fundamental or axiomatic, then it must by definition, have no explaination whatsoever, which would mean that all of reality is dependent on unexplainable or "supernatural" things - and that is something that science can't really deal with.

A small side tangent. I think the interpretation of the Darwin passage is overlooking one element of Darwin's discussion, namely, the analogy to free will and predestination. I think the analogy is very deliberate and significant here; free will and predestination is (in Britain, at least) often taken to be the paradigm case of an insoluble dilemma in the eighteenth and much of the nineteenth century -- that is, a dilemma in which we have excellent reason to accept both horns of the dilemma, and yet find ourselves unable to reconcile the two due to the limitations of the human mind. It's an epistemological category we ourselves usually don't leave room for, namely, apparent contradictions we are in principle unable to resolve. So I don't think Darwin is coming down unambiguously on a side here; I think he is saying, or at least would have been read by most of his contemporaries as saying (as I'm sure he would have known), that we have reason to accept both horns of the dilemma, but that the solution to the problem is beyond the reach of the human mind. I think this is fairly typical of Darwin's discussions of God: non-committal and involving a deliberately and carefully balanced way of saying "I don't know."

Re quantum mechanics as raised by several commenters, the probability mathematics are well worked out to a fantastic number of decimal places even by we non-omnisicient humans, so it's easily conceivable that a Creator could rig a/the Creation to tend 99.999etc% toward the desired outcomes

Probability mathematics of what? The entire evolutionary history of life? I don't think so.

Most of the information in the universe is from random quantum events. So unless God is indeed micromanaging these events, the Divine Plan will be a rather broad outline by necessity.

I agree with some earlier commenters that statements like the above are rather useless if we don't specify what kind of "desired outcomes" we are talking about. An universe where life is likely? Intelligent life? Intelligent life that is likely to develop concepts like loving its neighbour? From what we know about evolutionary history, the latter two were by no means inevitable.

Some nice replies here, thanks.

Jason G: the block universe doesn't affect the argument so far as I can see. Presuming God created the block, HeShe did so according to some utility function (I should make clear that by this I am referring to whatever it is theists mean by "providence" or "divine plan"). Natural law/secondary causes then become the regularities in that block.

On QM - suppose that God chooses the collapsed wave function that gives the outcome desired. This doesn't necessarily mean determinism, just enough determinism to achieve the outcomes.

Jason R: my point is that there can be a plausible scenario in which a God who doesn't have omniscience, etc., could desire outcomes and find a way to produce them. Phil and I are interested in what the limits of this might be. But an omni-3 God of course would not simulate these worlds but choose the best of all possible ones, relative to the utility functions. These are functions we, as philosophers and scientists, are ignorant of.

I'm not supposing this is an argument for divine creation; it clearly is not, so you might be misreading what I have said. I merely want to know if there is some formal objection to the simultaneous claims of God being the primary cause and desiring outcomes that arise by secondary causes, without God needing to intervene to produce them. Choosing between worlds is one way to do that. Another is the answer provided by Jud: God creates all possible universes (a view shared, without the God, by cosmologist Max Tegmark - the laws of mathematics define all actual universes, and we happen to be in one in which the constants are just right for us).

As to an uncreated creator - it is not incoherent in one sense: if the universe (or multiverse) is not created, then it is possible for there to be uncreated things. Hence an uncreated God is at least coherent. Not parsimonious, mind, but coherent.

And I am not using this argument to suggest that theists have evidence for their belief in a deity. This is merely a "proof of concept"; that is, a test of a possible falsifier or deprecator. I don't know how theists come by their faith (well, I know what they claim, but I don't know how that is authoritative). I'm just saying that given that faith, there are no further objections based on natural selection (and drift, etc.).

Brandon is quite right about the last sentence of Darwin's book - the free will controversy was particularly hot in the nineteenth century as physicians discovered more and more involuntary reflexes. Determinism has been increasingly creeping into that topic, but everyone at the time held to a strong voluntarism (which is why Marshall Hall could not publish about the reflex arc at first). And Darwin was very careful to say that we do not know. But I think we can say a lot more now about those aspects of the debate that are amenable to empirical and conceptual investigation.

On QM - suppose that God chooses the collapsed wave function that gives the outcome desired.

How is this not interventionist? At what point would he choose? And are you talking about the "wave function of the universe" or the wave function of something smaller (for example, a radioactive event that will trigger a desired mutation?)

I merely want to know if there is some formal objection to the simultaneous claims of God being the primary cause and desiring outcomes that arise by secondary causes, without God needing to intervene to produce them.

Theoretically not, if we are talking about some very broad type of "desired outcome". If God sets up a casino or a lottery or a tennis tournament, he can expect winners to emerge without having to intervene in the game, since he set up the rules. But as I see it, Darwin is not talking so much about the hypothetical idea, but about the practical problem of how God is supposed to have done it in the case of our history of life: how is "variation led along certain beneficial lines"?

In other words, I think you have misread Darwin's analogy a bit. It's not just about "allowing a builder to make a house from rocks off a cliff" but whether "certain fragments of rock should assume certain shapes" (to allow a house of predetermined shape). If we assume that God's simulations are so accurate that he can place a builder and rocks that have cracked into exactly the right shapes in the same time and location, it would seem to imply near complete determinism, and this is the problem Darwin points out.

I'm not sure why anybody is very impressed by the discovery that special pleading is always possible. Thing is, there are an infinite number of explanations of how the universe came to be and why, so what's the payoff for spending a lot of time establishing that a single one of them can be made coherent by a tremendous effort? Why not write apolegetical defenses of the thesis that seven bunny rabbits created heaven and earth? That's sort of argument would also be possible to make, which is why we have defense attorneys.

The reason people talk about God in the context of cosmology has nothing to do with fancy arguments about the anthropic principle or unmoved movers, or anything else--Judaism, Christianity, Islam, and the rest are normal human superstitions whose origins surely lie in the proclivity of our nervous systems to project a doer behind apparently random occurence. These errors are perfectly explicable, and grownups shouldn't make Bayseian priors out of 'em. Absent the priors, why would anybody cook up farfetched excuses for why God used so clumsy and sadistic a method as natural selection to cook up the likes of me?

Ah, but God can do anything for no reason explicable to mere mortals!

My sympathies are with Asa Gray on this, as Darwin is arguing from the sheer amount of trivial intervention needed to provide the correctly shaped stones to the builder, or to provide all the variations for natural selection to eliminate all but the desired change. If God can watch the sparrow's fall, He can also push the budgie off its perch, so to speak. Such an explanation is just as viable as the repeated and continuing poofing of new species into existence by The Designer, and has the advantage that the poofing, however miraculous, exactly matches the patterns that would be expected from natural selection.

Thus, as Gray indicates, both options are equal theologically. "We do not suppose that less power, or other power, is required to sustain the universe and carry on its operations, than to bring it into being. So, while conceiving no improbability of "interventions of Creative mind in Nature," if by such is meant the bringing to pass of new and fitting events at fitting times, we leave it for profounder minds to establish, if they can, a rational distinction in kind between his working in Nature carrying on operations, and in initiating those operations."

See Grays Review: The Origin of Species at the Darwin Correspondence Project, as well as other essays there.
http://www.darwinproject.ac.uk/content/view/82/67/

For one of Darwin's quotes: "What my own views may be is a question of no consequence to any one but myself." - I've sometimes wondered if that was just a generic statement about the way things should be, or whether he actually held specific theological or non-scientific views that he did not wish to discuss. The quote was from late in his life, I think.

John, one of my comments got stuck in moderation for some reason, can you release it?

The problem is that the only god for which "theistic evolution" is even remotely reasonable is a deistic god. It is one that set the wheels in motion from the start, and has since stayed well out of it. This isn't the sort of god that most religious people believe in.

To do this, you have to believe in a god that acts somewhat like this deist god, but then, every once in a while, decides he wants to tweak how things are going by putting in a miracle here, a miracle there.

The real reason to reject these sorts of ideas is simply parsimony. And the mildly-interventionist deity is quite a bit worse in terms of parsimony than the one that simple creates and leaves us alone, but even that one is much worse than merely saying, "it just exists."

By Jason Dick (not verified) on 06 Sep 2008 #permalink

Jason Dick: "To do this, you have to believe in a god that acts somewhat like this deist god, but then, every once in a while, decides he wants to tweak how things are going by putting in a miracle here, a miracle there."

That's pretty close to the God in which many Christians believe, although if one is talking about purported answers to prayers, the tweaks may not take the form of outright miracles.

By J. J. Ramsey (not verified) on 06 Sep 2008 #permalink

Jason G: the block universe doesn't affect the argument so far as I can see.

I'm confused about what you mean by "the argument". I didn't think it affected your overall argument, which I didn't have any disagreement with anyway.

I wanted you to explain to me just one little bit: why you thought that micromanaging was incompatible with natural law. And I was using the block universe as an example of why I thought micromanaging was compatible with natural law.

Now, you say:

Presuming God created the block, HeShe did so according to some utility function (I should make clear that by this I am referring to whatever it is theists mean by "providence" or "divine plan"). Natural law/secondary causes then become the regularities in that block.

OK, that helps me understand. Thanks. So then God is managing at the level of natural law. But on my understanding of the block view she is ALSO managing at the micro level, because the two are just interchangeable (unless there's some theological reason why it's forbidden to talk like that?). If I'm right about that then micromanaging and natural law are compatible, even though the rest of your argument (if I'm not confused) may well go through.

"In this way, everything is the result of secondary causes, while the Neo-Leibnizian God remains the primary cause, including of each event that satisfies the God's utility functions. So Darwin is, I think, not correct in saying that one cannot reasonably maintain that the outcome is designed by the deity."

There is one major problem I can see here: it supplants a static, global "utility function" into a natural scenario in which all utility is local and subject to various contingencies. Natural evolution, as far as I understand, doesn't allow for such things.

my point is that there can be a plausible scenario in which a God who doesn't have omniscience, etc., could desire outcomes and find a way to produce them.

Thank you for this and your other clarifications.

What still bothers me about your post is that final paragraph. You seem to think your argument constitutes some sort of refutation of something that “Dawkins and his fellow ideologues” have said. But it seems clear that you are using the words “God” and “theism” in a way that's different from how those terms are normally understood.

Virtually no one argues that even the traditional, triple-O God is logically incompatible with evolution. The issue is one of plausibility. Is it reasonable to maintain belief in the triple-O God and the authority of the Bible in the light of what evolution is telling us about our history? That's the question that generates all the heat, but it is not the one you are considering here. In my opinion it is not reasonable, and the arguments made to the contrary by people like Ken Miller, Karl Giberson, Francisco Ayala, and countless others are very unconvincing.

I'd also be curious to know whether you agree with people like Ken Miller and Simon Conway Morris that human-like creatures were effectively inevitable once evolution got going? Is this point essential to the argument you are making here?

Jason R: First of all here is what I sad: "So we should not, as Dawkins and his fellow ideologues do, attack those who are religious and accept and even promote evolution. The primary cause explanation is entirely distinct from the secondary causes that are the domain of science." By this I mean we should not attack those who are religious and want to accept and even promote religion. The argument here is a general purpose argument showing that even under constraints, a halfway decent deity has no trouble setting up a universe such that causal accidents are useful for some goal. A full deity with all the superpowers has even less trouble. But I see people like Dawkins, PZ and others attack those who are religious and pro-evolution on this line of argument, so the argument addresses that.

And no, I do not think that humans are inevitable. But if you need to think, for religious reasons, that humans are inevitable (because God planned for them) you can be accommodated without harm to the theoretical underpinnings of evolutionary biology because that is about secondary causes when God's plan is about primary causes.

I think you are importing some issues from outside the argument as presented. I am sure they matter, but not in this context.

By this I mean we should not attack those who are religious and want to accept and even promote religion But I don't like religion. Now evolution on the other hand, that's a nifty theory, and I don't mind religious people adding their extra entities to it, so long as they don't mess with the theory.....

By Brian English (not verified) on 07 Sep 2008 #permalink

I only made that comment because I thought you meant to say that "...those who are religious and want to accept and even promote evolution" but fluffed it. Obviously I was wrong. I shall slink back into the shadows. But we do agree. :)

By Brian English (not verified) on 07 Sep 2008 #permalink

Dang! I missed that. You're supposed to read what I meant, not what I wrote!

Tyler: if the secondary causal/natural law chain of events is meant, then no, it doesn't allow for a particular outcome to be selected. But the who gist of this argument is that primary cause can be something quite independent of secondary causal chains, since God is selecting the causal chains that result in the desired outcome[s].

Jason G: I suspect this is semantics, but I don't count it as micromanagement (i.e., constant intervention in the otherwise natural course of things) to select the block universe that best satisfies the deity's utility function[s].

Apropos of which, this.

John, thanks for the email. I'll try to recall my ramblings from the lost comment...

The argument here is a general purpose argument showing that even under constraints, a halfway decent deity has no trouble setting up a universe such that causal accidents are useful for some goal.

I think you are misreading Darwin in your rebuttal. I think he is talking about whether *particular* outcomes are planned (the actual organisms we see), not whether some general class of outcome is planned. Using Darwin's analogy, it's not just the matter of whether God created the laws of physics to allow rocks to crack naturally with the expectation that some of them will be useful for building houses, but whether God has provided for the right rocks in the right place to assemble into Westminster Abbey.

A "halfway decent deity" that creates natural laws with the expectation that *some* variants will be useful is of course compatible with evolutionary theory, but it's also indistinguishable from deism. I think the "guiding" in theistic evolution is a bit more than that. On the other hand you speak of a deity who has chosen some contingent features of our universe to a very exact detail (like humans) and say that it doesn't count as micromanagement. But I don't think Darwin meant his argument as a disproof of such a plan-making deity, just pointing out that it leads to ideas that are theologically hard to swallow.

And no, I do not think that humans are inevitable. But if you need to think, for religious reasons, that humans are inevitable (because God planned for them) you can be accommodated without harm to the theoretical underpinnings of evolutionary biology because that is about secondary causes when God's plan is about primary causes.

But the emergence of humans was, as far as we know, contingent on many external factors, like the impact that killed the dinosaurs, continental drift, climate change in Africa a few million years ago... If God chose the human-producing sequence of events from a simulation, that leads to the problem of determinism as Darwin recognised; or he intervened to produce them, which leads to problems with science; or they weren't part of God's plan. Can someone assume the latter, and also assume that humans were inevitable, without doing some violence to evolutionary theory? (Like asserting, without any evidence, that there is a lot of convergence towards intelligent humanoids.)

But I see people like Dawkins, PZ and others attack those who are religious and pro-evolution on this line of argument, so the argument addresses that.

Can you show me where Dawkins has ever raised this issue? As far as I know his argument is not that evolution disproves any particular notion of God, but simply that it makes belief in the Christian God unreasonable. It does this by refuting the argument form design in biology and by raising issues related to the problem of evil. It's possible that I've overlooked something that he's written, however.

And no, I do not think that humans are inevitable. But if you need to think, for religious reasons, that humans are inevitable (because God planned for them) you can be accommodated without harm to the theoretical underpinnings of evolutionary biology because that is about secondary causes when God's plan is about primary causes.

In my initial comment in this thread I said I could not make heads or tails out of your argument. I'm afraid I'm back to that. For example, in the post you said this:

In simpler terms, there's nothing formally wrong with believing the two following things: 1, that God made the world according to a design or desired goal or set of goals; and 2, that everything that occurs, occurs according to the laws of nature (secondary causes).

Again, can you show me someone who has argued that it is formally wrong to believe both of these simultaneously? The goal that is most relevant in this context is that of bringing about a human-like species. People like Ken Miller and Simon Conway Morris have defended both of these principles, based on the ubiquity of evolutionary convergences, and I am not aware of anyone who has argued against them on the grounds that they have made a logical or formal error, or that their argument is actually incompatible with the theoretical underpinnings of evolution. The argument against them is not that we can prove definitively that they are wrong, but simply that it is more reasonable to accept that a human-like species is not inevitable.

In short, it looks to me like you are working too hard to prove something that was never in doubt.

The issue is not whether God could have desired certain outcomes while still allowing everything to unfold according to contingent natural processes. Miller and Conway Morris could be right that evolution is so constrained that even though specific events are unpredictable ahead of time certain broad trends are predictable.

The issue is the one windy raised in comment 29. When we look at natural history as it is currently understood, it seems there were many places where the train to human-like creatures could have been decisively derailed. That observation is not a conclusion drawn from the theoretical underpinnings of evolutionary biology, it is instead something that is strongly implied by the facts as we currently know them. It is also a fact that is very difficult to reconcile with traditional Christian theism.

In other words, the argument between those who do and do not believe that evolution and Christianity are compatible based on this issue does not take place at the level of theoretical underpinnings or formal compatibility. It takes place at the level of the facts and our understanding of natural history. People like Ken Miller, Simon Conway Morris, Keith Ward and John Huaght argue not simply that these ideas are compatible, but that the most reasonable understanding of natural history suggests that both are true (or at least that humans were inevitable, thereby making it viable to think that humans are in some sense the point of creation). Michael Ruse also makes this assertion central to his defense of compatibility.

I still don't entirely understand your argument, so it is certainly possible that I remain confused about your intentions. But it looks to me like you are not really addressing the concerns of those ideologues you mentioned in your final paragraph.

issue is the one windy raised in comment 29. When we look at natural history as it is currently understood, it seems there were many places where the train to human-like creatures could have been decisively derailed. That observation is not a conclusion drawn from the theoretical underpinnings of evolutionary biology, it is instead something that is strongly implied by the facts as we currently know them. It is also a fact that is very difficult to reconcile with traditional Christian theism.

Assuming you're a Christian with a good enough imagination (which I am not), there are probably an infinite number of models one can devise to circumvent this problem. For example, assuming block time is real (time as structural feature of this universe and others), a sufficiently powerful deity - who can view many universes from the outside at any point in their history - could have chosen among a googeplex ^ googleplex number of existing universes, and this one just happened to be the one he chose to imbue these little ape-like creatures with a "soul". Their evolutionary past, contingencies, etc in block time would be irrelevant. Parsimonious? Hell no, but that's not really an issue for a determined Christian. And a multitude of even stranger models are possible, if one were to embrace an idealistic, rather than a realist view of the unvierse.

jeff: does it matter whether God chooses the universe containing humans just by omniscience, based on a simulation, or from infinite possibilities of block time? Each way, God has "microchosen" this universe down to very small detail. Darwin doesn't deny that this is possible, but it still leads to the predestination problem.

windy: The difference is that in my example, God's choice occurs outside our universe, where it is not detectable or subject to our laws.

You can also just as easily argue that he didn't choose the universe from infinite possibilities, but from a limited subset that he could deal with, and we were his best choice. In that case he would not be truly omnipotent or omniscient, but still a transcendent God. The point is, the number of models you can come with are endless...

windy: The difference is that in my example, God's choice occurs outside our universe, where it is not detectable or subject to our laws.

That is the case in all the examples discussed so far.

The point is, the number of models you can come with are endless...

Sure, natural selection is consistent with an endless number of models where the universe is a fully deterministic puppet show that was set in motion by God. It is also consistent with a more flexible kind of deism where God simply creates the natural laws but does not predetermine the details. I think Darwin, and many commenters here, were asking for a model in between, where God "guides" evolution toward *special* goals like humans and woodpeckers without having predetermined the path of the Chicxulub meteor.

windy: The difference is that in my example, God's choice occurs outside our universe, where it is not detectable or subject to our laws.

That is the case in all the examples discussed so far.

Not that I can see. In those examples, God intervenes at innumerable points in the universe's history to make it what it is today. That will potentially leave a much larger footprint than simply choosing a universe from large number of alternatives. Each of these two methods of "choice" would require immense processing resources, but in my example, they would not be visible in this universe, at least not until very recently. Perhaps a sufficiently powerful God would set up a separate universe dedicated to making that choice.

I think Darwin, and many commenters here, were asking for a model in between, where God "guides" evolution toward *special* goals like humans and woodpeckers without having predetermined the path of the Chicxulub meteor..

Actually, I think what most Christians are looking for is a way to reconcile their faith with what is observed. Whether or not God began to guide things before or after Chixulub is not important, as long as he's guiding them now.

John S. Wilkins wrote: But I think we can say a lot more now about those aspects of the debate that are amenable to empirical and conceptual investigation.
-and-
As to an uncreated creator - it is not incoherent in one sense: if the universe (or multiverse) is not created, then it is possible for there to be uncreated things. Hence an uncreated God is at least coherent.

COBE appears to have shown pretty definitively a "Big Bang" style creation of our particular universe between 13 and 14 bya, so an uncreated (which I take to mean eternal) universe is on the wrong side of the data, at least regarding the only universe we know.

To the extent one allows for the intrusion of this empirical data into the conceptual argument regarding an uncreated creator, I'd say it tends to militate against the coherency of that argument.

Jason G: I suspect this is semantics, but I don't count it as micromanagement (i.e., constant intervention in the otherwise natural course of things) to select the block universe that best satisfies the deity's utility function[s].

Good. We're pretty much sorted. Of course it's semantics. It's philosophy. And now I think this is not such a minor point after all.

Just try saying "I suspect this is semantics" when Alison is in the room. Sheesh.

Seems to me that on at least some versions of the block view it IS constant intervention but, OK, it isn't constant intervention in the OTHERWISE natural course of things. That means you're OK, but that your case rests on a pretty fine counterfactual distinction, so it's no wonder I was confused. Pointy Haired Boss suggests you spell this out in your paper.