It is widely thought that externalist theories entail that the mind is epiphenomenal (unable to be a real cause of events), which is a good reason to reject them since we know that the mind isn’t epiphenomenal (both from experience and because it would lead to the absurd conclusion that there is no such thing as human agency). Briefly, externalist theories claim that two people (and their brains) can be physically identical but have different mental content (or represent different things). But, since the physical world is completely causally closed, these people will perform the exact same actions, and thus it would seem that the mind, which content / representations are part of, played no role in the outcome. Dretske considers his theory to be a kind of externalism*, and thus in his book Naturalizing the Mind he attempts to defend his kind of externalism from the claim that it would make the mind epiphenomenal.
Dretske’s solution is to argue that in the case of mental causation there is a kind of over-determination, where the effects of an action are caused not only by the proximate physical features of the world, but also by remote historical events. Specifically, Dretske has been arguing that the reasons / selective pressures behind certain physical properties are what give them their power to represent the world, and he thinks that these historical reasons can also be said to be causes of events. He also describes them as having explanatory power, such that when we want to explain why a person performed an action we can appeal to these historical reasons (along with the proximate causes).
Let’s first tackle the argument that historical events can be considered to be additional causes in addition to the proximate physical causes. In one sense this is certainly true, since if the historical events hadn’t occurred it is conceivable that the current events wouldn’t have happened. The real problem is that when we ask what the cause for a certain event is we aren’t looking for a list of everything at all preceding moments that was a cause. As we go farther and farther back in time there are more and more factors that contributed to bring an event about (especially when you consider the butterfly effect), ending ultimately with the big bang, which is the cause of everything. But when people ask “what was the cause of her action?” they certainly wouldn’t consider the big bang as an acceptable answer, and the historical forces Dretske appeals to are no better. The relevant causes are recent ones, and thus Dretske’s theory would still be epiphenomenal, at least inasmuch as it doesn’t provide the kind of causes we are looking for when we inquire about people’s actions. But why should we be looking for causes that are close to their effects? Generally causes that are farther away from their results (in time) are more likely to be perturbed by other factors, just as dropping a rock from a great height makes it more likely to miss the target due to air currents. We know from experience that it is rare for our intentions not to be reflected appropriately in actions (the exceptions being diseases that affect parts of the brain, epilepsy, and paralysis), and thus we know that the mind, whatever it is, is temporally close to our actions (although not the absolutely immediate cause, since then it wouldn’t be possible for it to be disrupted). In other words Dretske’s appeal to historical forces as playing the causal role we expect of the mind seems a lot like a criminal blaming his unhappy childhood for his behavior. True it certainly was a cause, but not the kind of cause we were looking for.
Another way to see why Dretske’s externalism is epiphenomenal is to consider why we judge certain internalist theories about the mind to be epiphenomenal. Such theories, for example the theory that consciousness is simply a by-product of the brain, admit that it is conceivable for the mental properties to be removed from the situation and that events would proceed in the same manner. They also admit that it is not possible for the physical causes to be removed and have events proceed in the same manner. Thus we judge that only the physical causes are real, and that the others are causes only in name; they are epiphenomenal. We should judge Dretske’s historical forces to be epiphenomenal for the same reasons. Since Dretske admits that there might be perfectly identical copies of subjects but without their mental properties (if they were not brought about by the correct kind of historical forces) we could imagine a world that came into being at the previous moment (or a device that “destroyed” the history of this world). In such a world events would proceed in exactly the same manner as ours, but since there would be no historical facts about why things are the way they are there would be no mental properties, according to Dretske. This is half of the criterion for being epiphenomenal, and it should be obvious that under Dretske’s theory the physical properties of the world are still required for events to take place. Thus Dretske’s externalism meets the requirements for being epiphenomenal, even if we do occasionally speak of historical forces as causes.**
The same things can be said for the proposal that these historical forces explain our behavior (instead of causing it). When thinking about the mind and wondering how we explain a person’s behavior the selective forces that existed long before that person came into existence simply aren’t the right kind of answers (kind of a category mistake). The explanation we seek for behavior needs to be some factor that exists only when the person exists. (Actually we do sometimes seek these kinds of historical explanations, but usually when considering why people have the kinds of minds that they do, not in the context of why they perform specific actions.)
Finally I would like to mention that, although he attempts to address the epiphenomenalism objection to externalism, Dretske does not attempt to tackle the problem of other minds. Because externalism allows that there could be physically identical people, one with mental properties and one without, it is impossible even in principle to determine who has a mind. And an inability, even in principle, to resolve the problem is an equally serious failing for a theory about the mind to have, since it has undesirable consequences for other branches of philosophy (ex: ethics). The problem of other minds can only be resolved by theories that allow the mental to be a cause of physical events, since we could deduce the existence of the mental from its physical consequences. The inability of externalism to del with the problem of other minds is another good indication that it must also hold that the mind is epiphenomenal.
* Even though it would be easy to turn his theory into an internalist one.
** Could the same objections be raised to a materialist theory, since behavior is directly caused by neurons outside the brain? Yes and no. Yes in the sense that if materialists claimed that the mind directly caused behavior (with no intervening factors) the mind would be epiphenomenal. No because what materialists really claim is that the mind is the cause of (really identical with) the neural precursors of behavior. Making such a move would not help externalism, however, because it is epiphenomenal even with respect to the neural precursors.