I have argued before that thoroughgoing determinism has no way of accounting for penal responsibility (and thus much less for penal culpability). Determinists, e.g., like Derk Pereboom and Ted Honderich, admit as much in the plainest terms. I have also argued a few times against perdurantism and now I want to consider both topics in light of some ethical considerations. Since I believe the issues are decisive by virtue of their basicality, I hope this post will not be too long.
1) Consider Nicholas Humphrey's famous-ish attempt to "solve the mind-body problem." He argues that conscious sensation occurs when "I" afferently "reach out" to an efferent stimulus on my somatic boundary. Humphrey argues further that consciousness is a temporally 'thickened' "loop" of representation within an organism, a recursion which allows the organism to respond to its own response to the boundary stimulus. This response to one's own response is Humphrey's core explanation of consciousness. As many of the replies in that book indicate, Humphrey insinuates consciousness into his very account of the evolutionary emergence of consciousness, a defect I have recently criticized (e.g. here and here), but that is not the flaw on which I want to focus. Instead, I want to consider what it means for people like Humphrey and Dennett, inter alia, to reduce the "I" of any action A to a variable brain module VBM at the time t:A of said action A.
If it is not "I" that, for example, perceives (A:p:m) a maple leaf fall from the tree in October in Monmouth, Oregon, but rather that perception is the combined A of, say, an anonymous occipital lobe and thalamus at time t:A:p, then we must ask if A:p actually happens. After all, by all accounts, no occipital lobe ever perceives anything. Rather, it responds 'mechanically' to the photonic stimulus of the light reflected off the maple leaf. If occipital lobes perceived the world as such, we would immediately find our theory of cognition in the quagmire of "the (inner) Cartesian theater" from which Dennett so desperately wants to escape. For then the occipital lobe would be the "I" of subjective experience, the tiny homunculus which materialist cognitivism seeks to vanquish. The same goes for any other lobe or module of my brain: there is no there there. My entire "I" is rigidly determined by the internecine natural selection of neural modules in my brain, so I am not in charge of my brain's actions, but vice versa.
But then, if there is no subject of A:p:m then A:p:m is not a legitimate experience and did not therefore ever happen. Extend this to the supposition that the thalamus ostensibly within 'my' skull compels 'my' arms and hands to approach a bystander and strangle him in broad daylight. Two consequences follow: a) "I" am not culpable for this deed, since there is no "I" which committed the murder, b) there was no victim, since no one actually experienced being strangled in broad daylight, nor 3) were there any witnesses to the murder. Unless, that is, we are willing to start interrogating occipital lobes.
Obviously, however, I would be culpable for strangling a man in broad daylight, even if I ripped my own arms off and lobotomized myself (maybe not in that order!) just after doing do, so non-subjective determinism is obviously false.
2) Let us define myself in perdurantist terms as 4D:I. This is the aggregate of time-slices which are conventionally referred to as "Elliot". Now let us suppose the time-slices of 4D:I, ranging from times t'a to t'x, overlap with the aggregate of a "man," which we shall call 4D:H, in such a way that the actions of 4D:(t'a…t'x)I bring about the termination of 4D:H by strangulation. Sometime later (or some place farther) t'x+n, 4D:I is brought to trial for the murder of 4D:H. 4D:I's defense is that, since at time t'x+n he is not wholly identical with 4D:(t'x)I, therefore he is not wholly identical with the agent culpable for the murder of 4D:(t'x)H. Convicting 4D:I at time t'n.v, 4D:(t'x+n)I argues, would amount to the court convicting an innocent spatiotemporal entity, since it is 4D:(t'x)I, neither 4D:(t'x+n)I nor 4D:(t'x.v)I, who strangled 4D:(t'x)H. As such, only if 4D:I is sentenced and convicted in the same instant that 4D:(t'x)I strangles 4D:(t'x)H can there be any justice in a perdurantist world. But if the court convicts 4D:(t'x)I while 4D:H still exists, it could sentence 4D:(t'x)I to a sentence only coterminous with the act of murder itself (A:t'x). After t'x, therefore, 4D:I would be an innocent and free entity and would not have to serve any time in prison.
Yet more bullshit brought to you by perdurantism.