A Scientific Intelligent Design Theory

by Chris Cogan (Copyright 2005)

Feedback, discussion, comments, questions: Chris Cogan, ccogan@ou.edu

Abstract: A scientific alternative to "Intelligent Design" (or ID) theory is sketched, showing that it is possible to formulate an intelligent design theory that is in fact scientific (though not better than current evolutionary theory). This theory is contrasted with ID to show that it is necessary to specify attributes of the designer in order to explain the same data as is explained by evolutionary theory. In particular, it needs to explain biogeographical grouping of related species, and it needs to explain the "biotemporal" grouping of species as well. ID theory has no explanation for these and similar facts, because, by refusing to specify any attributes of the claimed designer, it removes the possibility of explaining such facts. This is because such facts can only be plausibly explained in terms of constraints on the designer, constraints that ID advocates refuse to allow because they would contradict the idea that the designer is an omnipotent and omniscient God. Thus, once again, the essentially religious and unscientific nature of ID is exposed.

Introduction

Creationists and "Intelligent Design" (or ID) advocates often claim that their views are scientific. The only problem with this claim has been the complete lack of any "creation science" or ID science. Because they claim that they are interested in being scientific, and yet they insist on retaining the intelligent designer idea under it all, they should be ready to leap on any intelligent designer theory that could actually work. This should mean that, if an intelligent design theory comes along that is genuinely scientific (empirically testable and at least mostly consistent with the data), they should be eager to consider it.

Therefore, in order to help things along, I have spent some time working out a scientific intelligent design theory, one that actually has a (very slim) chance of being true, and one that even has meaningful testable predictions. People who have been debating with creationists and ID folks for a few years will notice that this theory is not exactly new, except possibly in such relatively full development. It is not intended to be as much novel as a more or less well-developed expression of a view that is normally stated in a paragraph or two, and then followed by more or less unrelated subject matter. It is possible that the complete "package" of this theory, as I formulate it below, is in fact novel, but I certainly don't know that for a fact.

Be that as it may, let's look at the theory itself.

The Theory

We start by generally summarizing the data and setting up background assumptions.

1. All of the main well-established conclusions of the sciences other than the theory of evolution are either true or close enough. For example, the age of the Earth is assumed actually to be about 4.5 billion years, just as mainstream science indicates. Another and more directly relevant example: I assume that the designer started life on Earth about 3.8 billion years ago.

2. The data of evolution is genuine. No God has been planting faked fossils, as would be required to make ordinary creationism "work." Fossils are assumed to really be about as old, usually, as radiometric and other dating techniques indicate.

3. There has been a kind of evolution of life on Earth. That is, there is a relationship between earlier and later species that connects them, but it is not one of Darwinian evolution, nor is it one of some sort of inbuilt "urge" of living organisms to evolve. This will all become clearer as we go along.

4. The designer is not omnipotent and not omniscient. He may only have the intelligence, of say, a smart human being (or there may be many designers working together).

5. The designer is definitely not supernatural. A supernaturalistic version could be constructed, but it the supernaturalism would be superfluous (it would add nothing of explanatory value), and it would be even more of an affront to Ockham's Razor than my version already is.

6. The data includes all the usual data used to support evolution. For this hypothesis, I just use it for a different purpose. This way I don't have to try to explain away the data (as creationists and ID folks normally try to do -- when they don't simply evade it altogether). The data includes all the fossils, millions of them, or even millions of tons of them (in places like the Grand Canyon).

7. I assume ordinary genetics, but there is some potential here for problems. This potential is one of the ways in which the theory is open to testing in comparison with ordinary evolutionary theory.

8. I assume ordinary biological science, including biochemistry.

9. I assume ordinary chemistry, especially those parts of it that apply to the things we find in nature now or think we would have found in nature in the past.

10. I assume ordinary geology and geophysics.

11. Biogeographical information, such as the fact that, in many cases, closely related species occur in localized geographical areas (all zebras occur in Africa, for example).

12. Biotemporal information, such as the fact that all members of any species appear grouped in time so that, for example, the first zebra does not go extinct before the next zebra species comes along, and such as the fact that fish came before reptiles, and reptiles before mammals, and so on.

13. Methodological naturalism. I assume that no scientific theory ever needs to posit a supernatural whatsit, because it can just as well posit something natural that has the same powers that the supernatural whatsit supposedly has. This doesn't exclude designers and such, but it does require that they be naturalistic designers, not supernatural ones.

Basically, this simply includes all of the data now used to support the modern neo-Darwinian theory of evolution, and all the normal assumptions required for science to be done at all.

Now for the theory itself. Current design theory assumes an omnipotent and omniscient God did the designing (maybe only one in a million ID advocates honestly thinks the designer may be something other than God; the rest of those who say that their designer may be something else than God are lying; it will become clear why this is so later).

The Necessity for Constraints on the Designer

This assumption of the designer being God or at least of having God-like powers relative to the needs of designing life on Earth is the source of all of the main problems of current design ID theory.

Why? Because, if we assume that the designer is omnipotent and omniscient, then we can't explain many of the oddities and infelicities of nearly all living things that we have ever studied in detail, such as the placement of the human eye's light-receptors behind the neurons that carry their signals to the brain. Another example: The fact mentioned above, that, often closely related species occur in a geographically localized area. Further, they often occur in a temporally localized way (all within a relatively short period of time after the occurrence of the first species in a group of related species). These various oddities make no sense given the designer usually posited.

That is, in order to make the designer theory work, we have to impose some severe constraints on the designer. One of them is that, while the designer is intelligent, he (or it or they) is not so intelligent as to be able to create the more sophisticated life forms from scratch -- or, if he is, he is unable to do so because of limitations in power, such as not being able to construct a suitably large genome from scratch.

This restriction means that, except for the very first (and, we may guess, simplest) instance of a living organism, all of the following species are constructed strictly by a process of manipulating molecules (almost always only very slightly) during or before reproduction, so that, one day, a proto-hominid primate gives birth to a true hominid offspring that becomes the genetic basis for a new, truly hominid species (on the way, perhaps, to some future species).

This process is always a matter of "tinkering" because the designer doesn't know enough or have good enough equipment to reliably perform much larger changes in a single reproductive cycle, and, in any case, the offspring has to be enough like the parent organism (for sexually reproducing organisms) that it will be able to find a mate. In principle, larger changes could be made to non-sexual organisms, but there would still be the issue of viability. Any really large single-generation change will almost always kill the organism that carries it. Since the designer can't make more than a few changes at once that it can be sure will work, it tends to make only one change per parent organism or pare of parent organisms per generation (though the designer obviously has to make multiple changes to a species over short periods of time, it doesn't have to make many changes per individual organism).

Minimal Physical Force Available to the Designer

One major constraint has already been hinted at: The designer cannot bring much physical force to bear, or at least he cannot do so without killing things that it this force is applied to. For this reason, he can't transport organisms over any significant distances (if he can move them at all). This is why all the zebras occur in Africa: The designer started with a proto-zebra that existed in Africa (but not elsewhere), tinkered with it to get the first zebra, and then tinkered with that one (or a sub-population of it) to get the next zebra species, and so on.

Because of the constraints on the designer, there is no way he could easily create a zebra elsewhere, because he couldn't move a zebra and he had to start with a species that was already very close to being a zebra. This is why related species are related: Except for the first one of each group, all of the rest are created from that first (perhaps indirectly -- and it is created from a closely related species that is not quite part of the new group). This is also why we see temporal groupings of species: The first species is created, and all the subsequent ones have to derive directly or indirectly from that one, and when that group is no longer viable and goes extinct, it will not ever be precisely re-created again unless the designer find a genome that is "close enough" to be (eventually) tweaked into being the genome of that species.

My assumption here is that, once he has created a species or a group of species and it goes extinct, he have generally have no desire (or not enough resources to bother) to re-create it.

Who is the designer? I don't know. I've been speaking of "a" designer, but it may be that there are millions of them, all working, perhaps intermittently, on "Project Earth-Life." The designer may be some species on a planet very far away, who have managed to use their presumably advanced technology to reach out to Earth (without actually coming here) and that, advanced though it may be, their technology is able to exert only enough force to perform small molecular modifications to the genes of existing organisms (or, if they are lucky, perhaps to create some minimally complex life form that they can then manipulate to produce the rest through time).

Why did life change so slowly over time for the first three billion years? Again, I can't say, but perhaps the designer had other fish to fry, such as defending against enemies from neighboring star systems, or he just lost interest for the most part. For whatever reason, the progress of his many changes has been irregular in speed and direction. One constraint may also be that he had to make new life forms be adapted to their respective environmental conditions, and he may not really have known how to do this for more complex organisms for the first few billion years. Another possibility is that the designer is another species and that most of them have moved into sophisticated virtual reality systems and didn't have much interest in exploring the universe for a long time, and it may only have been recently that more of them decided to get involved in the directed-modification project on Earth again, thereby increasing the labor pool enough to get more work done.

Does the Theory Work?

Now, let's consider whether my theory works well, given the data. First, let me admit that, in some respects, it's not a great theory, because, for example, it makes many more -- and more radical -- assumptions than does the competing neo-Darwinian theory. Of course, theistic design theory adopts the most radical of all assumptions, and is wildly supernatural to boot (and, it doesn't work scientifically!).

To begin the examination: First, this theory accords with the general development of life on Earth from the simplest to the sophisticatedly complex (such as humans). Evolution says this is because life started out simply and really had nowhere to spread out except into the niches suitable for more-complex organisms as the niches for the simplest ones were used up. My theory says somewhat the same thing except that it assumes that the first life-form was simple because that was all the designer could directly produce, given his knowledge and technology constraints. Some organisms were gradually "tinkered" by the designer into progressively more of the same niches that evolution would fill. The designer had to follow the same kind of path that evolution would have because he could only make small changes and because each new organism, to be successful, had to be suited to its environment (because the designer couldn't keep an organism alive if it was too unsuited to live in its environment -- again because of technology limitations).

At each step along the way, each new species was achieved generally by subtle steps of genetic manipulation, done according to the designer's hopes or plans for the future species. Humans were created from pre-humans because that pre-human species was what was available to work from, and no really large changes could easily be made to any other species in a reasonable time that would also be compatible with whatever environments the various prospective organisms would find themselves in.

I think my theory even allows us to preserve all of genetics. Our designer interferes in genetic processes, but not terribly often, and only enough to get the results he wants, which means that most genetic observations are of genetic processes or events that are not being interfered with at the time. Further, because he can generally only tweak, any individual change he makes is likely to be so small as not to significantly upset genetic statistics. It's only in the long haul that his interventions are noticeable -- or would be if we had a good enough record of exactly what would have happened on Earth without the designer's interference (presumably nothing much at all, biologically, if the current theory of evolution is really false in a basic way).

Predictions for Testing the Theory

Just one, for now: I predict, on the basis of my theory that, if a new group of related species is found in some geographical area that has been "enclosed" in some way since the first such species appeared (by mountain ranges and/or water, for example, such that there has been no opportunity for any of them to escape on the back of another species, or any thing of that sort), then all members of this related group of species will be in that very same geographical area. Why will they all be in the same area? Because the designer can't move a species to another area, and so all species that he derives from an existing species will have to be derived by manipulating genetic material in individual organisms or germ cells.

I note that this same prediction can be based on the conventional theory of evolution as well, so the results of this prediction (assuming we find such a new group of related species) will not show that my theory is better than the conventional theory. It will, however, show that it is better, by far, than the current ID theory touted by the likes of the so-called Discovery Institute and its intellectual toadies, because it is prediction-free. Given that the designer is truly unknown, with truly unknown powers, there is no telling what the designer will do. He did only make zebras in Africa, but, in the future, he could make whole groups of closely-related species such that each species within the group is separated from every other such species by untraversable mountains, oceans, deserts, or whatever. Further, he might do something similar to Haldane's rabbit fossil in the Precambrian, and disperse related species throughout time in such a way that none of them could be the ancestor of any of the others. Given an utterly unspecified designer, no predictions are possible. Thus, putting constraints on the designer as I do makes my theory potentially testable.

Problems with the Theory

The first and most obvious problem is that it violates Occam's Razor, as any design theory alternative to conventional evolutionary theory must, given the presently-available data. This is inherent in introducing a designer at all when the data doesn't require it. In principle, the real-world data could one day be too much for evolutionary theory to be able to handle without introducing a designer. Until that day comes, however, the Ockham's Razor problem remains.

The second problem is that we don't know the goals of the designer, so we can't make the kinds of long-term predictions about how life would change given different environmental conditions that we could with evolutionary theory. My designer is stuck with constraints, but even these could change (were this designer real at all), and the designer's goals could change (even assuming we knew what they were). Further, other factors could interfere and ruin the designer's plans, whatever they are. Note that this is no disadvantage with respect to the ID designer, who, being unspecified, could do anything at all in the future, including letting other factors interfere with what he initially had in mind.

Another major problem is that, as far as I can see at present, this theory makes no predictions that are distinctive to it as compared with ordinary evolutionary theory, except, possibly with respect to genetic statistics. Indeed, it has been made to "mimic" evolutionary theory, in a sense, because it had to encompass and integrate the same data and make the same predictions in cases that we already know about. Because it has been designed to mimic ordinary evolution, the only way I can see of giving it a distinctive test is via careful accumulation and analysis of genetic statistics, to see if something has been happening in nature that doesn't happen in the lab, such as an overabundance of certain kinds of genetic modifications.

The assumption that the designer is intelligent but not omnipotent or omniscient is added, independently of the one just described, in order to account for the frequent appearance of what would be design flaws if we assume that the designer is both omniscient and omnipotent (or at least extremely knowledgeable and easily able to create the genes necessary for any set of traits he wants). If we make the first constraint above sufficiently strong, we might not need this one. The assumption that the designer may have been distracted for much of the past few billion years by other things is made to explain the extremely slow changing of life on Earth for the first three plus billion years, though this could also be simply a result of having relatively little to work with.

Worse, I don't think we can ever make it distinctively testable in any clean-cut way, at least not without modifying it to include a specific further sub-theory about the goals and possible additional constraints of the designer. If we make the right set of assumptions about the designer, we may be able to make a theory that can be put to special tests that would distinguish its predictions from those of ordinary evolutionary theory. However, I wouldn't expect any such variation to do well in such tests.

Related to the Ockham's Razor issue is the issue of ad hoc components. The theory is a collection of assumptions based each on a different set of considerations. The assumption that the designer can't really move more than a few molecules around to make genetic modifications is added in order to make the theory fit biogeographical and temporal data about groupings of closely related species.

All of this ad hoc introduction of assumptions make the theory "work" for classes of data that it otherwise would not, but what we'd rather have is a set of basic principles that make the theory work "automatically" for these same classes of data, without having to stop and add a new assumption that explains what would otherwise not make a lot of sense from really intelligent designer.

I don't see any way out of this. Indeed, I expect we may have to add more ad hoc assumptions as time goes on and we find more things that the theory does not yet handle. That is, I don't see the theory being very much able to account for as-yet-unexamined classes of data without requiring such changes, each of them added only for the purpose of fitting that particular class of data into the system. This is far better than the ID/creationist approach of simply ignoring, denying, explaining away, or distorting the data, but it suggests that the theory is not really correct (which I wouldn't think it is anyway, but that's a different matter).

And, although I think we can reduce the designer's interference to a low enough level and have him do it in such a way as not to significantly "infect" genetic science, I'm not entirely happy with that aspect of the theory. In effect, this is almost a form of explaining away, in that the data of genetics suggests a view of what happens genetically that's quite different from what this theory suggests.

Prospects

This theory is not ideal because it posits so much for so little (as compared to conventional evolutionary theory), and because it has no scientific advantages over conventional evolutionary theory in terms of predictive power.

On the other hand, in comparison to the view that the designer is God, this theory is a huge improvement.

And yet, I don't think most ID advocates will find even this theory acceptable, even though it really is a scientific intelligent design theory.

Why will ID people reject it even though it is obviously better than their own theory? Why will they reject a theory that is actually scientific (if in a somewhat loose sense) whereas their own theory cannot make any testable predictions?

They will reject it because their position is not really a scientific one at all. It is, in fact, an almost purely religious view, and all the ID science talk is nothing more than propaganda aimed at trying to wedge their own religion into our public schools and into our government, at the expense of everyone else's religion (and the philosophical views of the non-religious).

They are religious dogmatists with all the scientific interest of the average abortion-clinic bomber. The ID proposal is their attempt to bypass our Constitution and our courts' prohibition on imposing their religion on our children in the public school system and thus further the process of indoctrinating everyone's innocent and intellectually relatively defenseless children into being right-wing fundamentalists like themselves.

How do we know this? I mean, aside from the fact that they reject any attempt to produce a genuinely scientific intelligent design theory? Because, that's exactly what they have said they want (though they didn't phrase it in exactly those terms, of course). Initially, when the Discovery Institute was starting out, they said this to each other and to religious people they were trying to recruit. Now that they have good funding, they've put on a (rather thin and more or less transparent) veneer of scientific respectability, while still espousing the same religious and political goals amongst themselves).

The point of my proposed scientific intelligent design theory is to expose the fact that their motives are not what they sometimes pretend them to be when they are trying to appear scientific (i.e., when they are trying to get ID into the public school science courses), and to expose the fact that their honesty is, to say the least, minimal.

If they were genuinely more interested in science than in pushing religion, they would welcome any theory such as the one I propose with open arms and jubilant enthusiasm, because it does exactly what they claim a designer theory can and should do (but which their own theory has not been able even to approach). They would adopt it, they would begin developing it further and formalizing it, they would begin working out further tests, they would begin applying it to empirical research, and writing computer emulations (based on different sub-hypotheses as to the goals and constraints on the designer) to implement and demonstrate its key points, and so on.

But, the basic idea of using a non-supernatural theory has been around for years. Behe has even used it as an evasion of the charge that he's just pushing religion (but always disdainfully, from what I hear). However, ID folks have never actually tried to develop this idea because that's not what they actually want. A non-theistic, naturalistic designer theory, if accepted, would destroy the value of ID for their social and political goals, so, though they may occasionally trot out this idea, they are grossly dishonest in doing so (if they were honest about it, they'd actually put some work into the idea and I and others wouldn't have to do it for them).

Feedback, discussion, comments, questions: Chris Cogan, ccogan@ou.edu