In Chapter 8 of “Where the Conflict Really Lies”, Alvin Plantinga advances the arguments of intelligent design proponent Michael Behe, argues for the design argument on the basis of the foundationalist epistemology, and concludes that evolutionary science does not defeat the design argument.
It’s been very beneficial to examine each chapter thoroughly as Plantinga starts building on his earlier argumentation in earnest in this chapter.
The arguments of Michael Behe make it very clear that proponents of intelligent design simply do not understand natural selection. It has been clear throughout that fundamental misunderstandings of natural selection are at the root of many arguments in favor of intelligent design and divine intervention. It seems necessary to restate again the principles of natural selection as it relates to evolution.
Evolution is the confluence of natural selection and deep time. Deep time provides an excellent probability that natural selection occurred.
By natural selection we mean unguided natural selection involving nothing more than the actions of the laws of nature and the universe. Otherwise it would perhaps be supernatural selection.
Natural selection does not mean that only adaptive traits get passed on. It means that organisms that survive get to pass on their traits, which include adaptive and nonadaptive traits.
Natural selection operates through descent with modification. The modification that occurs may occur in any direction. It may add characteristics, remove characteristics, or change characteristics.
This means that although natural selection tends to be adaptive in the long run, each modification is not necessarily adaptive. Thus natural selection is an unoptimized process. We would not expect to see natural selection resulting in optimal designs.
Sometimes non-adaptive traits get modified to become adaptive traits. There exist multiple paths to any given adaptation that do not always involve adaptation at each step.
Thus, natural selection predicts nothing. It is a bottom-up process that may or may not result in the survival of a species. Where it results in survival, the results will be seen to be very much in tune with the environment the surviving species find themselves in.
With that basis, the first order of business is to examine the “irreducible complexity” argument directly. Behe is not a pioneer here by any means, either logically or scientifically, merely repurposing ideas for his own conclusions. Logically, “irreducible complexity” is nothing more than the “watchmaker argument” at a smaller level. His argument then is really only (potentially) novel at the level of his scientific evidence.
Scientifically, “irreducible complexity” is design-inspired reimagining of what was understood as far back as 1918 as “interlocking complexity” by geneticist Hermann Muller. Muller argued that interlocking complexity is an expected result of evolution through natural selection. Evolution is not just gradual addition, but gradual modification, and this provides the mechanism through complexity can arise by degrees yet end up functionally “irreducible”. The image that comes to my mind is that of an ocean liner whose smokestacks have rusted away leaving only the many, many layers of paint applied through the years.
Sometimes an unnecessary part becomes necessary. In a primer on Muller’s idea, Dr. Douglas Theobald explains it this way: consider a bridge made of three blocks. You then overlay a flat surface over all three blocks. Finally, you remove the middle block. You still have a bridge, and the flat surface, initially unnecessary, has become necessary, while the middle block, initially necessary, has become unnecessary.
H. Allen Orr, in his review of “Darwin’s Black Box”, describes another potential mechanism in the form of increasingly symbiotic relationships. Consider that adaptation A is advantageous, and is subsequently joined by adaptation is B, which is advantageous in that it assists the function of A. Later, adaptation A further adapts to make better use of B, to the point where A now requires B. B, which at one point was unnecessary, served to turn both A and B into necessary parts of a system.
The fatal flaw in Behe’s concept of “irreducible complexity” is that natural selection, as I highlighted above, is not merely the addition of parts in a linear fashion. There are steps forward, steps back, additions, subtractions, modifications, etc. and these play together to give us the world we have today.
In his book “The Edge of Evolution”, Behe narrows the scope of intelligent design to the molecular level. In particular, he argues that evolution has produced no new protein-protein interactions (combinations of proteins that carry out cellular tasks) since the first cells showed up on the scene. These interactions are seemingly held to be “irreducibly complex.”
He further argues that natural selection can’t adapt to threats that require multiple mutations in the protein-protein interactions. He uses the example of malaria, which has multiplied massively since the introduction of anti-malarial drugs many years ago. Although it has acquired resistance to chloroquine in a number of cases, it has only done so after a massive number of generations, which demonstrates that in more complex organisms with longer life-times, it is highly unlikely that such a thing would happen.
Plantinga notes two good objections to this; first, that what happens with host-parasite interactions such as malaria and HIV might not apply more generally. Second, that if two protein-protein mutations are required, a single mutation may be preserved by natural selection (either in conjunction with another adaptive trait or by virtue of conferring some adaptive benefit on its own) until the second mutation comes into being. He does not reply to these arguments but instead turns to the probabilities and realizes we have no good way of evaluating that.
So Behe’s argument here seems to just lose its legs, especially given that there is evidence that new protein-protein interactions have evolved. From a review of “The Edge of Evolution” by David Levin:
“Behe is likely aware of at least some of the existing evidence that new protein-to-protein interactions have evolved. One must look no further than one of his acknowledged examples of evolutionary prowess. Under the heading of "What Darwinism Can Do," he describes the stepwise evolution of an antifreeze protein from a digestive enzyme in Antarctic fish. This was an important evolutionary adaptation that allowed fish that possess this protein to survive in frigid Antarctic waters. However, he omits an interesting detail from his description - the antifreeze protein has sugars added to it (by an enzyme), whereas the protein from which it evolved does not. Therefore, a new protein-to-protein interaction must also have evolved to allow modification of the antifreeze protein. In fact, this beautiful example of evolution involves the construction of significant complexity.”
Plantinga skips by defending the watchmaker argument as an argument and defends it on the basis of the foundationalist “given”. That is to say, that if we perceive something by direct observation, the truth of that perception is a given. Foundationalism (or, at least, a specific version of it) then will regard that that perception as a proper ground of knowledge. He goes on to argue that when we perceive nature’s complexity as the work of a designer, that perception then, is valid as a basis of knowledge.
This particular brand of foundationalism involves what is called phenomenal conservatism. From the article on Foundationalist Theories of Epistemic Justification in the Stanford Encyclopedia of Philosophy:
“the justifying state is a “seeming” or “appearance”: if it seems to S that P, then, in the absence of defeaters, S thereby has some degree of justification for believing that P”
Plantinga defends such phenomenal conservatism and the impression of nature as intelligently designed as an independently sound justification under this form of epistemology.
It is unlikely that the impression of nature as intelligently designed qualifies as a justification under the measure of phenomenal conservatism, since it is inferential itself. For example, one would not approach such an impression unless one decided that it first seemed like nature was complex. In the sense that we are discussing, the concept that nature is complex might be a justified belief (or might not), but derivations of that belief are by definition not foundationally justifiable on their own.
The impression of intelligent design also would not count as a foundationally justifiable seeming/perception since it is affected by propositional attitudes. Especially for the theist, they would have prior justification to believe there is a designer. For the non-theist, the impression of intelligent design would be based in a prior justification that certain levels of complexity require agency.
As an inferential belief, the watchmaker/design argument is fair game for logical argumentation. The perception of design cannot stand on its own as a foundational belief.
Having bypassed dealing with the widely accepted defeat of the watchmaker argument by claiming it finds its basis in perception, he then claims that Darwinism can only provide a rebuttal if it demonstrates that evolutionary science has proven that natural selection is unguided. He rests on his arguments against Dawkins and Dennett in chapters 1 and 2 to claim that Darwinism has not, in fact, shown this.
I am also unsure how a theistic evidence base fits into this. He never really told us what defines a theistic evidence base; if I assume it is revelation, how does that fit into his foundationalism? Or do we really have the truth “written on our hearts” and a proper application of foundationalism will reveal that all men understand the truth of this theistic evidence base?
He moves on to rebutting and undercutting defeaters. He claims that Darwinism does not rebut guided natural selection since it only describes the mechanism. Here we have an admission from Dennett himself (which I am surprised wasn’t quoted) in chapter 11 of “Darwin’s Dangerous Idea” (pp.317-318):
“It would be foolhardy, however, for any defender of neo-Darwinism to claim that contemporary evolution theory gives one the power to read history so finely from present data as to rule out the earlier historical presence of rational designers—a wildly implausible fantasy, but a possibility after all.”
We won’t argue something these two men agree on.
However, the inability to rule out something does not make an idea worth holding.
So then moving past rebutting we come to what Plantinga calls undercutting defeaters. This is where is has become vitally important for him to have established the truth of the design argument, since it is now presented as a justified belief that is not sufficiently undercut by Darwinism. Let us remember that his attempt to put the design argument on foundationalist footing failed, and he did not attempt a refutation of the well established arguments against it. So the design argument fails, leaving us with no argument for Darwinism to undercut. We are merely left with a battle of probabilities between guided and unguided evolution.
He goes back to his foundationalist argumentation to argue that fine-tuning and Behe’s irreducible complexity can be accepted on the same grounds; that is, we perceive fine-tuning and irreducible complexity, therefore, we can accept them as justified beliefs in much the same way that we accept what we perceive what is around us. The same objections apply--these concepts are derivative of other basic perceptions and involve propositional attitudes and so ought to be dealt with on the level of argument and not, as he calls it, “discourse”.
Having no design argument for Darwinism to undercut, he then pulls out Behe’s biological evidence for irreducible complexity. Now keep in mind that Behe’s argument here is that you can’t explain how this came to be by unguided natural selection so it must be designed. Besides being a non-sequitur on the face of it, since then science has explained more, and gathered more evidence, to substantiate unguided evolution in those areas Behe mentioned. It just seems like hanging on to Behe is another god-of-the-gaps bit of thinking.
Plantinga seems to back off to the level of saying that Darwinism doesn’t provide significant undercutters and so does not offer any defeaters. But again, defeaters for what? What argument has Plantinga successfully defended that Darwinism is undercutting? Argument from design? Fine-tuning? Irreducible complexity? None of these has he successfully defended outside of trying to claim some are epistemologically valid on their own, which doesn’t work either.
He wraps up the chapter with a discussion of what he calls “deflector beliefs”, those beliefs that result in not being able to arrive at a separate specific belief. At a very simple level the concept is one of presuppositions; the beliefs you bring into a situation will influence the beliefs you form about a situation. This is just as much an argument against the argument from design as anything else. If my children are taught evolution through natural selection at a young age and do not receive religious teachings, the argument from design will have no foothold. The argument from design will be, quite simply, a non-starter. They won’t be employing the same propositional attitudes as a theist. Plantinga offers that a hypothetical neutral observer would likely be pre-disposed to conclude design, but doesn’t thoroughly argue that point. I don’t think it can be well defended in any case.
At this point I think he is far from disproving unguided natural selection (UNS). I think it is very important to argue here for it. In Dennett’s book, “Darwin’s Dangerous Idea”, he notes Aristotle’s four basic questions that man asks about anything, also known as the four causes. From page 23:
“(1) We may be curious about what something is made of, its matter or material cause.
(2) We may be curious about the form (or structure or shape) that that matter takes, its formal cause.
(3) We may be curious about its beginning, how it got started, or its efficient cause.
(4) We may be curious about its purpose or goal or end (as in "Do the ends justify the means?" ), which Aristotle called its telos, sometimes translated in English, awkwardly, as "final cause."”
#1 asks what is it? #2 asks what is it made of? #3 asks how did come about? #4 asks why is it, or what is its purpose?
Let’s take for our subject the universe as it exists today, and reflect on these four questions, as they would be answered by UNS, versus how they would be answered by natural selection guided by a designer (TE, theistic evolution).
#1: The material and final result of TE is indistinguishable from UNS.
#2: The mechanism of TE is indistinguishable from UNS.
#3: TE involves an unnecessary hypothesis that is not required by UNS.
#4: UNS is by definition purposeless. TE, on the other hand, requires purpose, but does not tell us what that purpose is. Where it attempts to do so, it establishes religion and fails by way of the argument from pluralism. For example, why did god use evolution, not 6 days of creation? Why did god cause the world to come into being the way it did?
Unguided natural selection is the better explanation. Arguing for TE is a grand exercise in scientific proof texting.