Sunday, March 25, 2018

Mental States are Causally Irrelevant on Naturalism (p. 5)

Home > Philosophy > Metaphysics
  1. Problems for Naturalism
  2. Why Mental States are Causally Irrelevant on Naturalism
  3. The Symbolic Logic
  4. Proving It
  5. The Argument is Sound, but So What?

The Argument is Sound, but So What?



Causal Irrelevance



One objection is largely one of semantics: on naturalism, even if mental states are causally irrelevant in my sense of the term, mental states are not causally irrelevant in the ordinary sense of the term. A naturalist could believe that mental states are identical to brain states, and so when brain states cause stuff, so do mental states. Thus mental states are causally relevant in the ordinary sense of the term, and so there’s no problem for naturalism.

The problem for naturalism (as the symbolic logic argument suggests) is that the mental states associated with the brain states have nothing to do with the brain’s causal behavior. That said, whether mental states are causally irrelevant on naturalism requires some nuance. On an identity theory variant of naturalism (where brain states are identical to mental states), in one sense mental states are causally relevant to the brain’s behavior: mental states are identical brain states, and brain states cause behavior. Nonetheless, on naturalism the association between the brain state and the mental state would not be causally relevant to the brain’s causal behavior.

To illustrate the somewhat subtle idea behind this I’ll borrow an example from Alvin Plantinga.[6] Suppose a ball flies through the air and breaks a window by virtue of (among other things) having a mass of 0.2 kilograms. Suppose also that having a mass of 0.2 kilograms is Sam’s favorite property; Sam’s favorite property just is having a mass of 0.2 kilograms (this identity association does not hold by definition, but likewise it’s conceivable for a brain state to be identical to a different mental state or no mental state). In one sense, Sam’s favorite property is causally relevant to the ball’s behavior: Sam’s favorite property is identical to having a mass of 0.2 kilograms and having a mass of 0.2 kilograms is (part of) what cause the ball’s behavior (breaking the window).

On the other hand, the fact that the ball’s mass of 0.2 kilograms is Sam’s favorite property doesn’t seem to have anything to do with why the ball breaks the window. In other words, the identity association between the ball’s mass and Sam’s favorite property is not causally relevant to the ball’s causal behavior; the ball would still have broken the window even if Sam’s favorite property were identical to something else (e.g. the ball’s color). Similarly, the identity association between the brain state and the mental state is not causally relevant to the brain’s causal behavior; even if the brain state were identical to a different mental state or no mental state, the brain state’s causal behavior would be the same (as the symbolic logic proof helps demonstrate).

So if we define a brain’s mental content as the mental state associated with the brain state, such that a brain state having a different mental content means that a different mental state is associated with it, the brain’s mental content is causally irrelevant to the brain’s causal behavior; the brain would still behave the same way even if it had no mental content. But if on naturalism the mind’s mental content (as opposed to its physical properties) has nothing to do with why its causal behavior is the way it is, this will be a problem for naturalism if a mind’s mental content does have something to do with why it causes stuff, as common sense would seem to suggest. The above reasoning arguing for the mind’s mental content being causally irrelevant on naturalism in this way can be justified via symbolic logic.[7]

The naturalist could bite the bullet and deny that a mind’s mental content has something to do with why it causes stuff, but not everyone will find the bullet plausible. Even so, that is not (in my opinion) the biggest problem for the naturalist that results from the causal irrelevance problem the symbolic logic argument describes. There’s also the Probability Thesis of the EAAN.

The Probability Thesis



The naturalist could believe that mental states supervene on brain states in the sense that there cannot possibly be a difference in mental states without a difference in brain states. What type of “possibility” are we talking about here? It varies, but we can assume arguendo that it’s metaphysical possibility; i.e. that mental states strongly supervene on brain states (perhaps because brain states are identical to mental states by metaphysical necessity). Will this solve the problem the alien scenario brings up when it argues Pr(RA|N&NE) is low?

Recall that the scenario is one in which naturalism and evolution is true for our aliens, and we’re given no further relevant information, but let’s add to this scenario that mental states strongly supervene on brain states. What can we tell from this information? On naturalism the mental states of the alien humanoids are causally irrelevant in the sense that which mental states are associated with the physical states wouldn’t affect the outcome; e.g. even if mental states supervene on brain states, no matter which mental states supervene on the brain states, the behavior would still be the same. Since mental states are causally irrelevant in this way, even if the supervenience relation yielded garbage mental states, this would not affect behavior, and many different supervenience relations are conceivable. Even if mental states supervene on brain states, we don’t know which supervenience relation holds for these aliens, and it would be a serendipitous coincidence that the supervenience relation would yield an accurate mental state instead of one of the far more numerous garbage mental states. All things considered, it seems more likely that our aliens are effectively living in a dream world. Thus, the likelihood that the alien cognitive faculties are reliable given just the information we have (naturalism and evolution) is still low.

So even if mental states supervene on brain states by metaphysical necessity, mental states being causally irrelevant in the way I described still provides ammunition for the following argument:

R = Our cognitive faculties are reliable
N = naturalism is true
E = evolution is true
Pr(R|N&E) = the probability of R given N&E


The argument that Pr(R|N&E) is low with RA representing “the cognitive faculties of the aliens are reliable” is this:
  1. If Pr(RA|N&E) is low, then Pr(R|N&E) is low.
  2. Pr(RA|N&E) is low.
  3. Therefore, Pr(R|N&E) is low.
Justification for (1): What’s true for the humanoid aliens here is also true for us, since we are basically considering the probability of R on just N&E (we considered Pr(RA|N&E) merely so we could try thinking about the issue in a way that avoids bias towards our own species).

Justification for (2): Here’s what we know about our alien species on N&E:
  1. Mental states are a causally irrelevant and useless byproduct of the biological process, such that if different mental states supervened on brain states, the physical outcome (e.g. behavior) would be the same.
  2. Garbage mental states vastly outnumber ones that accurately resemble one’s external reality.
  3. We have no a priori reason to believe an alien would have an accurate mental state (one that accurately reflects their external reality) instead of one the far more numerous garbage mental states; e.g. we have no a priori reason to believe the supervenience relation yields an accurate mental state instead of one of the far more numerous garbage mental states.
We’re basically trying to evaluate the likelihood of RA given just N&E, so points (a), (b), and (c) together suggest that given just the information we have here, it is considerably more likely that the aliens have unreliable cognitive faculties (garbage mental states being far more numerous) than reliable ones. Mental states strongly supervening on brains states doesn’t save the naturalist from this sort of argument.

An argument for the Defeater Thesis (the claim that the naturalist who sees that Pr(R|N&E) is low undercuts the naturalist’s justification for believing R) is beyond the scope of this article, but you can see a quick one on page 3 of my EAAN post-debate reflection.

Conclusion



On naturalism, mental states are causally irrelevant in the sense that which mental states are associated with the physical states wouldn’t affect the outcome. Symbolic logic helps prove this with just two premises; one of which is true by definition (it follows by definition of what laws are; descriptions of how physical reality behaves in the absence of supernatural intervention), and one of which makes the very modest claim that it is logically possible (basically, not self-contradictory) for a physical state to have a different set of mental associations with it.

This sort of causal irrelevancy poses (at least) two problems for naturalism. On naturalism, the mental content of a brain state (i.e. the mental state that’s associated with the brain state) doesn’t have anything to do with why its causal behavior is the way it is (even if brain states are identical to mental states, brain states do not behave the way they do because they are identical to a particular mental state; a brain state would behave the same way if it was identical to a different mental state or no mental state). Thus on naturalism, a mind’s mental content doesn’t have anything to do with why its causal behavior is the way it is, whereas common sense suggests a mind’s mental content does have something to do with why it causes stuff. Another problem is that mental states being causally irrelevant in the way I suggested leads to Pr(R|N&E) being low (the Probability Thesis of the EAAN), and that’s true even if mental states strongly supervene on brain states. It seems to me then that mental states being causally irrelevant on naturalism poses a serious problem to naturalism.

 |   1   |   2   |   3   |   4   |   5   |   Next >




[6] Plantinga, Alvin. “A New Argument against Materialism” Philosophia Christi 14.1 (Summer 2012) pp. 23-24.

[7] Let’s define a brain’s mental content as the mental state associated with the brain state, such that a brain state having a different mental content means that a different mental state is associated with it.

W = (On naturalism) If brain states were identical to a different mental state or no mental state, the outcome (e.g. the brain’s causal behavior) would be the same; to the point where the same brain state (in the same physical state of the universe) entails the same causal behavior regardless of which conceivable mental state is associated with the brain state. On naturalism, if the physical state of the world had any other conceivable set of mental states associated with that physical state (e.g. different mental states associated with the world's brains), the physical outcome would have been the same.
A = (On naturalism) even if the association between brain states and mental states were an identity association, the association between brain states and mental states would not affect the the brain’s causal behavior (i.e. it would not affect what the brain does or what it causes).
B = (On naturalism) the mental content of a brain state does not affect the brain’s causal behavior (if the brain state had a different mental content—a different mental state associated with it—this would not affect the brain’s causal behavior).
M = (On naturalism) the mind’s mental content is causally irrelevant in the sense that the mind’s mental content does not affect the brain’s causal behavior.


  1. W
  2. W → A
  3. (W ∧ A) → B
  4. B → M

  1. A 1, 2, modus ponens
  2. W ∧ A 1, 5, conjunction
  3. B 3, 6, modus ponens
  4. M 4, 7, modus ponens
Note that premises 2, 3, and 4 are material conditionals (i.e. the only they can be false is if there is both a true antecedent and a false consequent).

Justification for (1): the main symbolic logic proof showed that these two statements are true:
  1. □((M2 ∧ N ∧ L ∧ I) → O)
  2. (M2 ∧ N ∧ L ∧ I) □→ O
Since conceivable mental state associations to brain states (M2) include the identity association, the proof (if sound) via statement (11) implies that on naturalism, if brain states were identical to a different mental state or no mental state, the outcome (e.g. the brain’s causal behavior) would be the same. Statement (10) implies the same brain state (in the same physical state of the universe) entails the same causal behavior regardless of which conceivable mental state is associated with the brain state.

Justification for (2): Assuming naturalism: W claims the brain state would still cause all the stuff it does regardless of whether it was associated to a mental state (even if it’s an identity association) to the point where the same brain state (in the same physical state of the universe) entails the same causal behavior regardless of which conceivable mental state is associated with the brain state. From this and the fact that W implies that even if the brain state were identical to a different mental state or no mental state, the brain's causal behavior would remain the same; it seems to follow that even if the association between brain states and mental states were an identity association, the association between brain states and mental states does not affect the brain’s causal behavior (as A claims) i.e. it doesn’t affect what the brain does or what it causes.

Justification for (3): One approach is to assume arguendo that W and A to see if we can reasonably infer B. W and A imply that the association between brain states and mental states does not affect the brain’s causal behavior (it doesn’t matter which mental state is associated with the brain state; the brain’s causal behavior would be the same). By definition, the mental content of a brain state is the mental state associated with that brain state. So if the association between brain states and mental states does not affect the brain’s causal behavior, then the mental content of a brain state does not affect the brain’s causal behavior; if the brain state had a different mental content—a different mental state associated with it—this would not affect the brain’s causal behavior (as B claims).

Justification for (4): One approach is to assume arguendo that B is true and see if we can reasonably infer M from it. On naturalism, it is the mind’s brain that must do the causal work of the mind, since any nonphysical component the mind might have would be outside the natural, physical world; and anything outside the natural, physical world causally influencing the physical world would by definition be supernatural. Thus, if the brain’s mental content is does not affect what the brain does, the mind’s mental content likewise does not affect the brain’s causal behavior (as M claims).

No comments:

Post a Comment