Review of TLS -- Do you mind my science?

This is the twelfth part of my review of Dr. Edward Feser's The Last Superstition. You can find parts one, two, three, four, five, six, seven, eight, nine, ten, and eleven at those links. In the final chapter, Dr. Feser attempts to make the case that science is inherently Aristotelian in practice. Along the way he does a little ridiculing of eliminative materialism and two of its principle proponents, the Churchlands, derides the notions of the mind operating as a computer does, and equivocates the notion that there is regularity in the properties of matter as being indicative of final cause. He relies upon excluding a few middles, the literal interpretations of metaphors, and lots of uncompromising, evidence-free rhetoric. In other words, he uses pretty much the same methodology as the rest of his book. Below the fold I'll go over the various sections of this chapter. The next piece in this series will be a comparison of the promises I listed in part 2 to what Dr. Feser actually demonstrated, and unless Dr. Feser decides to contribute a response to this series, it will be the last installment.

The chapter actually starts in a manner I could agree with, by pointing out how nonsensical the Churchlands sound in a conversation where, instead of Pat saying she is in a bad mood, she'll say she has depressed serotonin levels (or something similar). However, I think the reason their conversation, which replaces things like a bad mood with a low serotonin level in normal conversation, is ridiculous for an entirely different reason. He thinks that their attempt is nonsense because the mind is not eliminable; I think it's nonsense because things like human emotions rarely boil down to a single source, and so identifying your good mood to natural opiates, or whatever, says something you can't possibly know is true at the moment. Of course, for Dr. Feser the Churchlands and eliminative materialism (the claim that their is no such thing as a mind) play the same role he claims Paley and the Design Argument play for the New Atheists, something that it's easy to kick around. Then, by claiming that eliminative materialism is the inevitable result of rejecting final causes (using the negative definition of the mind he set forth in Chapter 5 and I discussed in part ten, linked to above), he attempts to equate any notion the mind exists to the full-fledged acceptance of Aristotelian causes. He graciously concedes that for an acorn, having a final cause of an being an oak doesn't mean that the acorn has a function or purpose to be an oak. Here he undercuts his Chapter 3 Supreme Intelligence proof (which I discussed in part 5) that the Supreme Intelligence is directing the acorn to become the oak; giving the acorn a final cause of being an oak is giving it a purpose just by maintaining the final cause. However, to make his case in this chapter, final causes now become mere directions or tendencies for an event to occur, with no need to invoke something directing it. Maybe he thought no one would notice, or maybe he can't tell the difference himself, I really don't know.

The metaphor of the mind as a computer

Dr. Feser discusses four ways that he thinks the metaphor of the mind as a computer fails.

1) The brain can't really be storing symbols of outside events, because symbols require an interpreter. Since prior to modern times no one thought of the brain as storing symbols, it obviously is not storing symbols [no, I'm not making that up, he really says that]. The whole process idea is circular: the mind is made of symbols; the symbols need the mind to interpret them. Nor is this point of view rescued by referring to memes, because it takes a mind to identify what a meme is, so minds must precede memes [yes, he really says that, too].

2) It can't be said that human minds are running algorithms, because algorithms require intent to follow and algorithm. A repetitive or cyclical behavior by itself is not an algorithm. This, we again have a circular process; the mind can't be built on algorithms if the mind is needed to run algorithms.

3) Causal connections between the brain and the outside world become incoherent under this view. Looking at the process of vision, there is no objective reason to give favored status points the causal chain that start with light bouncing off an object and terminate with brain nerves being stimulated. Thus, the items that demarcate events are not truly objective.

4) Computers can't formulate rational arguments; they can only simulate the following of valid logical forms. If you arbitrarily change the meaning of the forms and symbols, the computer will not change how the forms are processed. For example, if you assign A=>B to mean "Grass is green", B=>C to mean "Cleveland has a Hall of Fame", and A=>C to mean "James doesn't eat fish sticks", then since formally ((A=>B) & (B=>C))=>(A=>C), the computer will interpret this sentenced as being valid: "Grass is green and Cleveland has a Hall of Fame, this means James doesn't eat fish sticks". This shows you need to understand the meaning behind the logical symbols and that logical reasoning is not possible without a mind. This is a version of the Argument from Reason.

Together, these arguments show you can't eliminate goals from human actions. This factors into the main reason why you can't explain the mind in mechanistic terms; it was created to hold the unexplainable. So, if you are going to hold to mechanism at all and not eliminate the mind entirely, your only choice is dualism.

My response

The circularity argument in part 1) ignores a more obvious interpretation of a spiral. Even in animals that we presume have no intentionality at all, such as insects, at a very basic level translate exterior information (such as the position or the sun) and process symbols (such as the dance of a bumblebee telling other bees where to find flowers). So we clearly don't need a full-fledged rational mind to process symbols. This allows for the process to happen in a spiral, where storing the symbols leads to a certain level of mind, which interprets more symbols to create a more complex mind, which can interpret still more symbols to create a yet more complex mind, etc. By trying to present the argument as an all-or-nothing, Dr. Feser attempts to hide the in-between stance of an emergent phenomenon.

Argument 2) is just playing a game with definitions. Whether you can call a basic prcedss in the mind does an "algorithm" or not, the behavior is the same.

Argument 3) ignores that sight developed in a feedback loop. The earliest sighted animals did not distinguish points in the causal chain as being favored, however, those that followed later who did distinguish those points were better able to evaluate the environment, and passed that tendency along.

Argument 4) would be much more compelling if it did not take so much effort to teach logical thinking to humans. Humans are also prone to see form over meaning, confuse correlation with causation, etc. So it's not surprising that computers should have the same issues.

Overall, I agree that in many ways the metaphor of the mind as a computer does not do it justice. However, it's still a good metaphor for specific purposes, and Dr. Feser's objections to it are ill-founded.

Science and final causes

Dr. Feser discusses three areas of science that show final causes are essential.

A) "Darwinian biologists" use teleological language to describe biological systems, and could not do research without it. Phrases like the "function of the heart" and "in order to signal [there are] predators [near]" need to be explained without teleological language. It's not possible to do this without referring to final causes, and scientists don't even try. This is just as well, because the nonsense that philosophers write about science is only exceeded by the nonsense scientists write about philosophy. Any attempts to explain functions in terms of natural selection fails because it means (1) you can't know the function of an organism without knowing it's evolutionary history, (2) nothing that is unevolved can have a function (so the first kidneys had no function, the organs of Swampman have no functions, etc.), and (3) using natural selection as an explanation removes functionality rather than explaining it. You can't remove teleology from your explanations because the mind doing the explaining is teleological [once more: yes, he really says that]. You can't explain teleology in nature without final causes, you can only eliminate it. Descriptions of DNA as holding information, as a blueprint, etc. are fundamentally teleological and show that some interpreter is required; this leads to the notion of form. Note that Paleyan red herrings like irreducible complexity are not a part of this argument. When Dawkins discusses concepts like the selfish gene he pushes science beyond where it can go. When you remove teleology genes/DNA they are no longer useful as biological explanations.

B) Complex inorganic systems, such as the water cycle and the rock cycle, also show final causes, in that each stage in the cycle has a role to play in leading to the next stage. Efficient causes don't explain this, because each stage also has effects that are not part of the cycle. Nor can you say the cycles are just human constructions, they are objective effects. Thus, they are results of goals within each stage.

C) The basic laws of nature make sense only if you interpret final causes as being within objects. The modern view is that the brick shattering a window is not based on anything inherent to the brick or the window, but that event like a brick being thrown are simply followed by event like a window shattering. Hume's empiricism provided cover for these notions, but it is refuted by noting that mental concepts do exist. Humean notions undermine science by saying there are no necessary causal connections between events and denying the existence of law-like correlations. However, scientific regularities are hard to obscure. In practice, scientists draw conclusions after a few experiments, confirming that they accept law-like correlations. Even the notions of "ideal conditions" and "interference" that are needed for various thought-experiments are not Humean.

In summary, because of these reasons we are witnessing the return of a "physical intentionality" to science. This of course means the restoration of Aristotle has been accomplished in fact, if not name.

My response

I know I am being redundant, but I wanted to emphasize this. Back in chapters 2 and 3, formal and final causes were portrayed as having a shaping effect upon objects, where formal causes shaped what they are and final causes what they could become. However, in this chapter, any sort of regularity in behavior is treated as a final cause. Brittleness leads to a tendency to be shattered because the final cause of brittleness is shattering. Matches strike flames because their final cause is to strike a flame. This is done to emphasize repeatedly: science needs final causes. However, the final causes used in this section do not need the Supreme Intelligence to guide them; they merely need a principle of uniformity: under identical conditions, matter behaves identically. This falls short of the Aristotelian causes as described earlier. Dr. Feser is understandably anxious to exclude this middle view.

It's difficult to imagine any thing said about philosophy that is sillier than the notion there was a first kidney, and on top of that to pile on the notions of it being unevolved and without function. So apparently philosphers can win that contest after all. More generally, the feedback loop on survival creates exactly the teleological impetus that Dr. Feser sees as an emergent phenomenon. Beings that act in ways that favor the survival of their offspring see increased survival in their offspring. The goal of survival emerges naturally from this loop. It does not need to be imposed from the outside.

Complex inorganic systems certainly exhibit regularity. However, I have already pointed out that regularity is not the same as a final cause. This is also true basic natures and properties of objects, and the physical laws these natures and properties have them exhibit.