Thursday, June 14, 2012

Minds and brains

Jeff Lower has attempted to respond to my criticisms of his original post.

Reply: Like the previous objection, this objection is not of obvious relevance to APM and for the same reason. The existence of mental states is included within the background information for APM.

Jeff is not entitled to treat the existence of mental states as a given within the framework of naturalism, if that background information is inconsistent with naturalism. He’s artificially isolating his argument from evidence to the contrary. Moreover, the counterevidence is internal to naturalism (i.e. eliminative materialism).

Reply: If human minds were independent of the physical brain, then brain injuries should not have much, if any, impact on mental activity since, ex hypothesi, mental activity does not occur in the brain to begin with. Thus, E1.2 is antecedently much more probable on the assumption that E is true than on the assumption that E is false, i.e., Pr(E1.2 | E) >! Pr(E1.2 | ~E).

The objector’s computer analogy does not explain E1.2. If my computer were stolen, no amount of fiddling with it would enable the burglar to affect my ability to speak, recognize faces, feel pain. In contrast, brain injuries can produce exactly these kinds of results.

Jeff is committing a level-confusion. According to my computer analogy, the soul is to the user as the computer is to the brain. On this analogy, the brain and soul are ontologically distinct or separable, but epistemologically indistinguishable or inseparable inasmuch as we lack direct or independent access to the soul. Our knowledge of the soul is mediated by the conduit of the brain.

To say damage to the computer doesn’t damage the user misses the point, for the analogy presupposes that very distinction. Because the soul or user is metaphysically distinct or separable from the brain or computer, damaging the medium doesn’t damage the soul or user. That’s not inconsistent with my argument; rather, that’s a presupposition of my argument.

However, you can’t get around the medium to find out what the soul is really like, apart from the medium. Damaging the computer does affect Jeff’s ability to communicate via the computer. It prevents him from sending or receiving information via the computer. So you have a correlation analogous to the correlation that Jeff stipulated between mental events and brain events. And if that were his only conduit to the outside world, a computer malfunction would be epistemically equivalent to neurological impairment.

Reply: Regarding E1.4, when we compare different classes within the animal kingdom (e.g., fish, amphibians, reptiles, birds, mammals), we find that more intelligent classes always have more complex brains.

Really? For instance, I’ve seen lots of nature shows about crocodiles. Crocs have a small “primitive” reptilian brain. Less “advanced” than the mammalian brain. Yet folks who study crocs say they are very cunning–especially older crocs with years of experience.

I am not a neuroscientist or biologist, but I suspect that there is some margin of error with the correlation between brain size and mental capacities, e.g., an animal with a 500cc brain is always going to be more intelligent than an animal with a 5cc brain, but there may be instances within the same species of an animal with a 450cc brain that is more intelligent than an animal with a 500cc brain.

But do we actually have any hard data comparing the brains of smart little dogs with big dumb dogs? Or neurological data comparing smarter dog breeds with dumber dog breeds? 

Reply: Again, I am not a biologist, but I doubt that any biologist thinks that the intelligence of social insects is even in the same league as that of chimpanzees.

Well, let’s get more specific. For instance:

Honeybees build nests with hexagonal combs for brood raising and food storage. All of them use dance language as their primary means of recruiting nest mates to valuable resources.

Apis floreata and A. adreniformis the smallest species build small cones hanging from tree branches, because the support for the comb supplies a flat surface on the top of the branch they can indicate direction directly by doing their dance on this part of the comb. The other species all do their dance on the vertical plane of the comb and thus indicate direction by transposing the direction of the sun to the direction relevant to gravity using a straight upwards direction to be equivalent to flying towards the sun. In all species the vigor with which the bee dances is directly correlated with the richness of the resource indicated, while the length of the straight run, or its omission are indications of the distance to be traveled to the resource, this need not be a straight line but may involve flying around some natural obstacle such as a small mountain.

The Honey Bee dance comes in three forms; A) the reversing circle or round dance used to indicate a resource close to the hive, B) the sickle dance, only used by the Italian race of Apis mellifera to indicate a slightly more distant resource, and C) the waggle dance, which is basically the circle dance with the addition of bisecting line down the middle during which the bee waggles her abdomen. The more intense the waggling the closer the food source. Different species and races use the round dance to indicate resources at different distances, thus A. floreata, A. dorsata and A. cerana only use the round dance for distances less than 50 meters while Apis mellifera (race) ligustica uses it for distances up to 200 meters and A. mellifera cornica for resources up to 1000 meters distant.

Chimpanzees don’t construct geodesic homes. And they don’t have this sophisticated signaling system. On the face of it, honeybee behavior is more advanced than chimp behavior in this respect. If chimps built geodesic homes or employed complex sign language to give directions, wouldn’t we be tempted to attribute that to their superior brainpower? Conversely, if honeybees were as big as chimps, wouldn’t we be inclined to attribute their behavior to their superior brainpower?

Or take slavemaker ants:

The young slavemaking queen will wait outside of the colony she is leaving and follow a group of raiding slave makers into her new colony. As the worker slavemakers raid this colony for eggs, the queen takes advantage of the battle by using it to sneak into the colony. Once it finds the queen, it kills her and takes her place. The new queen mimics the old queen by consuming pheromones from her body and releasing them to the attending ants. This new queen having mated with a slavemaking male earlier begins to produce new slave makers. Other variations on these hostile takeovers include one South American species whose workers secrete a chemical on a host colony that causes the ants of the host colony to evacuate the nest. In their haste to leave, pupae will be left behind. These developing ants are then taken back to the slave maker nest. Another variation is in a European species that attacks ants that are significantly larger in size. The queen invades a nest by clinging on the rightful queen and slowly chokes her to death.

That’s pretty clever. If higher animals did that, it would be tempting to ascribe that to their mammalian brainpower. Or if ants were as big as midsized mammals, wouldn’t we be inclined to attribute that behavior to their brain capacity?

Reply: Yes, but (I'm told) predatory insects are much more limited in their behavioral responses than social mammals like wolves, lions, leopards, and dogs. The latter are much more versatile in their behavioral responses because of their greater complexity. I could be wrong, but I believe this point is uncontroversial among entomologists and ethologists.

Take the trapdoor spider. If a monkey constructed a tunnel with a hinged, camouflaged lid, to ambush prey–wouldn’t we be inclined to chalk that up to the monkey’s superior brain development?

Reply: Given that my argument is an evidential argument, this is false. Here considerations like the explanatory virtues come into play. What is the most parsimonious, scientifically conservative, successfully predictive explanation that accounts for the widest range of facts? Judged like any other empirical hypothesis, E is the best explanation, hands down.

There are several difficulties with that glib appeal:

i) One problem with Occam’s razor is that it’s prejudicial and premature. Occam’s razor is really an admission of ignorance. If we already knew how simple or complex the world is, we wouldn’t resort to Occam’s razor in the first place. So we’re getting ahead of ourselves at that juncture.

ii) Apropos (i), a simpler theory is better than a complex theory–assuming that reality is simple. But why assume (at the outset) that reality is simple? That’s something you can't predict ahead of time.

A simpler explanation is better–provided that the world is as simple as your explanation. But that’s the very question at issue. You don’t want a theory that’s more complicated than reality, but by the same token you don’t want a theory that oversimplifies reality. Rather, you want a theory that matches reality.

And Occam’s razor is misleading, because you can’t say in advance what the world is like. That’s something we must discover.

iii) There’s a tradeoff between ontological simplicity and theoretical simplicity. More entities can simplify the explanation. Fewer entities can complicate the theory.

Most philosophers believe that, other things being equal, simpler theories are better. But what exactly does theoretical simplicity amount to? Syntactic simplicity, or elegance, measures the number and conciseness of the theory's basic principles. Ontological simplicity, or parsimony, measures the number of kinds of entities postulated by the theory. One issue concerns how these two forms of simplicity relate to one another.

 A distinction is often made between two fundamentally distinct senses of simplicity: syntactic simplicity (roughly, the number and complexity of hypotheses), and ontological simplicity (roughly, the number and complexity of things postulated).[3] These two facets of simplicity are often referred to as elegance and parsimony respectively.

For instance, Jeff is disregarding familiar, intractable problems confronting physicalism, viz. qualia, intentionality.


  1. In addition to syntactic simplicity and ontological simplicity, there's functional simplicity. Something could be ontological or part-wise simple but functionally complex. An electric razor is mechanically more complex than a straight razor, but the latter is functionally more complex in that I could use it to perform more tasks than shaving.

    Moreoever, there's serious debate about whether simplicity as a factor in choosing which theory to believe is truth-apt. Bas van Fraassen and others have raised serious challenges to it. Pragmatic vs. epistemic virtues. But does this just mean that the one who opts for simplicity is simply expressing some aesthetic preference? They just happen to like simpler theories? It makes things nice and easy to work with and manage? This is a debate about the pragmatics of theory choice.