Epiphenomenalism is the view that (1) physical events exert causal influence on mental events, but (2) mental events do not exert causal influence on physical events.
Pro
The primary argument in favor of epiphenomenalism comes from the mountains of evidence we have about the behavior of matter down to the level of subatomic particles, as summarized in the standard model and general relativity (forget about dark matter, dark energy, etc. for now). Given this evidence, some conclude it is unlikely that mental events could also be affecting the behavior of matter without our having noticed this in experiments.
There are two primary responses to this argument: First, most of our evidence here comes from observing ostensibly unconscious matter, and not e.g. living brain tissue. Second, if panpsychism is true, we would not expect to observe differences between the behavior of conscious matter and unconscious matter, since on panpsychism it’s just all conscious.
Con
The primary argument against epiphenomenalism comes from an apparent correlation between mental states and adaptive behaviors that seems tricky to explain without at least some causal arrows running from mental to physical. For example, our experience of unpleasant mental events like hunger, thirst, pain, and fear tends to be followed by adaptive physical behavior like eating food, drinking water, and avoiding damaging stimuli. Options for explaining this correlation include some combination of (a) mental events causally contributing to the physical behavior, (b) the physical behavior causally contributing to the mental events, or (c) some third factor causally affecting both. However, if (b) or (c) were the sole explanations, it’s difficult to see why the correlation exists at all - any old mental events will do so long as the right behavior occurs, so the connection seems arbitrary. Thus, it seems that (a) has to be at least part of the explanation for the correlation.
(For clarity, there are obviously plenty of cases where physical events causally influence physical behavior while the subject has the illusion that mental events are causing the behavior - e.g. split brain patients rationalizing why they moved their left hand. That’s not under dispute. What’s under dispute is whether such physical events account for 100% of the causal factors contributing to the physical behavior.)
There are two primary responses to this argument: First, some try to challenge whether the correlation actually exists. For example, hunger may also be followed by maladaptive behavior (overeating leading to obesity). I tend not to buy this response, though. For example, in this particular case at least it seems that there really was a correlation between hunger and adaptive behavior over most of human history, and that it is only in recent times when food was plenty that the correlation became muddled because selection forces have not had enough time to adequately address the overeating problem. Second, some would point out that the arbitrariness problem lodged against (b) and (c) above as sole explanations could also be levied against mental causation. That is, it seems like a surprising coincidence that the psychophysical laws are such that negative mental events would cause adaptive behavior. However, this doesn’t seem to solve the arbitrariness problem lodged against epiphenomenalism, but rather just shows that there is a second problem that needs to be solved by the anti-epiphenomenalist.
Taking all this into consideration, I tend to believe that epiphenomenalism is false with moderate confidence. It’s hard to find hard data on this, but anecdotally it seems that this is the majority opinion of philosophers of mind (although not anywhere close to a consensus).
Prompt
With all that in the background, here’s my challenge for discussion: If it’s really true that (A) all matter follows the behavior set forth in the standard model and general relativity (“causal closure”), and (B) epiphenomenalism is false for the reasons discussed above (“anti-epiphenomenalism”), that seems to constitute evidence for panpsychism.
In other words, if at least some physical events are causally affected by consciousness, we would expect conscious matter to behave differently from unconscious matter. If we then observe that all matter behaves the same (although granted it may be difficult to do that due to practical limits and eventually the perennially-annoying Heisenberg), that seems to constitute evidence that all matter is conscious.
To be extremely clear, I am using “evidence” here the way that epistemologists do, meaning anything that is probability-increasing with respect to a proposition. I’m not claiming anything like mathematical proof or hard scientific evidence that demands a consensus. Rather, it simply seems that if (A) and (B) are true, that should meaningfully increase our Bayesian posterior.