@
Titus Rivas, I am revisiting this thread some six years (!) later to share some thoughts after reflection. In summary:
I continue to enthusiastically endorse the core argument of your paper, however, I think that you've missed or short-changed some other arguments against epiphenomenalism that are powerful enough that they also decisively falsify it, or at least are not easily dismissed.
Briefly, before getting to those arguments, you might find useful the summary form into which I put your argument when it became relevant
in a recent thread. I've given it a new name in this form:
The anti-epiphenomenalist argument from knowledge of consciousness
- If consciousness is epiphenomenal, then we do not (because we cannot) know we are conscious.
- We know we are conscious.
- Therefore, consciousness is not epiphenomenal.
Based also in the fatally problematic nature of the causal impotence of mind on epiphenomenalism, I think another compelling argument is:
The anti-epiphenomenalist argument from knowledge of the mind's causal efficacy
- If consciousness is epiphenomenal, then the mind is not casually efficacious.
- We know that the mind is causally efficacious.
- If we know that the mind is causally efficacious, then the mind is causally efficacious.
- The mind is causally efficacious.
- Therefore, consciousness is not epiphenomenal.
A defence of the argument
Premise one is true by definition. Premise three is unobjectionable, and premise four follows from premises two and three. The conclusion follows from premises one and four - that is, the argument is valid.
Thus, the only defence required is of the second premise. I defend it as follows:
It is apparent to us (me and the reader,) given our experience, that we know that the mind is causally efficacious. That we are mistaken in this apparent knowledge is the only way that the premise could be false. The task of the epiphenomenalist, then, is to provide an alternative account on which our apparent knowledge is, indeed, mistaken. I proceed by example to show that this task cannot succeed.
By example: the inexplicability of mental causation on epiphenomenalism
This example uses
the quality of feelings as the causally efficacious mental phenomena. For the purposes of this example, the quality of feelings (including sensations and emotions) can roughly be distinguished as being either positive (that of pleasure) or negative (that of pain).
Here is the example:
A person suffers a terribly agonising headache, then thinks afterwards to him/herself, "That felt awful", and then resolves to donate money to medical research into the causes and treatments for headaches.
The thought afterwards demonstrates the causal efficacy for which the epiphenomenalist needs to provide an alternative account: it makes sense given that the headache
had a negative (painful) quality to subsequently think, "That felt awful".
Given that negative quality of the headache, it would
not have made sense if the thought afterwards had instead been, "That felt wonderful". Thus, we
appear to know that the
quality of the feeling was causally efficacious. We experience this causal efficacy so consistently that this example is easily recognisable as representative
of that consistency of the mind's causal efficacy.
On epiphenomenalism, though, (the negative) quality (of the headache) is not inherent in the underlying (physical) brain state which gave rise to it; its (negative) quality is determined by a
metaphysical relation. Given solely the physical facts (about the brain state that gave rise to it), it
might as well have been positive as negative. Thus, there is nothing in that brain state itself that plausibly
could cause a subsequent brain state which in turn causes the
appropriate epiphenomenal thought, "That felt awful".
Potential response #1: placeholders
There are better and worse ways for an epiphenomenalist to respond to this simple point, but no successful ones of which I am aware. Perhaps the worst which is even considerable as a response is:
The word "awful" is simply a placeholder token that refers to whatever phenomenal experience happens to be associated with such brain states as in this case happen to have a negative quality.
We know that this isn't true though: words like "awful" are not mere placeholders; they really do refer to a negative phenomenal quality.
Potential response #2: encapsulation
The most promising response for the epiphenomenalist which makes minimally troublesome (for that epiphenomenalist) claims, then, it seems to me, is this:
OK, I grant that "awful" means what it means, and that its use following the headache is appropriate given the negative quality of that (epi)phenomenal experience, and that that appropriateness seems to be causally inexplicable on epiphenomenalism - but it's not inexplicable; not really. It's perfectly plausible that the brain state that gives rise to the headache is in some sense "encapsulated" in the brain state which gives rise to the reflection on its awfulness, and that it is this similarity - via encapsulation - that leads to the (epi)phenomenal correspondent of the thought "That felt awful", in which the same negative quality is embedded.
The problem is that there's a sleight of hand in that response: the requirement is to explain causally, given the causal impotence of the (negative) quality of the (epi)phenomenal experience of the headache, why the appropriate
physical substrate of the thought "That felt awful" follows (as opposed to, say, the -
inappropriate -
physical substrate of the thought "That felt wonderful"),
not to explain the (quality of the) (epi)phenomenal experience of that thought, which is irrelevant in this context. This response offers nothing additional by way of the (hitherto lack of) explanation as to why that appropriate physical substrate is caused and not some other inappropriate one given the causal impotence of the phenomenal quality of the headache.
To put it another way: it might very well be that there is some "encapsulation" of the prior brain state in the subsequent brain state, and that this could result in some
similar epiphenomenal
reexperience of the negative quality of the headache while saying "That felt awful" - but none of this is present in or available to the prior physical substrate, so we still have no explanation as to why the appropriate
semantic content ("That felt awful") is present in the (epi)phenomenal experience caused by the subsequent brain state given that the prior brain state
had no access to the (awful) quality of feeling which might have explained that appropriateness.
Potential response #3: subsumption of the mental in the physical
The epiphenomenalist might then respond:
OK, I grant that that doesn't explain the appropriateness of the subsequent physical substrate of the "That felt awful" thought given the negative quality of the headache as experienced, but we can explain that by postulating that the negative quality of the headache is not determined by the metaphysical rules relating physical and mental events, but is part of the physical substrate of that which causes the (epi)phenomenal experience of the headache itself. That is how it can cause an appropriate physical substrate for the next (appropriate) mental event which is the thought "That felt awful".
At this point, though, the epiphenomenalist has given up on epiphenomenalism - and conceded the argument - because this amounts to causally efficacious experiences - the denial of which is a fundamental premise of epiphenomenalism - and premise one of the argument. It's also incoherent for, as this potential response implies, a mental fact simultaneously to
be a physical fact, but that's a point to take up elsewhere.
Final response: it just so happened
It seems, then, that the epiphenomenalist is reduced to this response:
OK, I grant that "awful" means what it means, and that its appropriate use following the headache has no adequate causal explanation on epiphenomenalism - but it just so happened that the physical causal chain led to what just so happened to be an appropriate reaction.
Here, too, the epiphenomenalist has lost the argument, because "it just so happened" is not an adequate causal explanation, especially for the third and final step in the causal chain of the above example - that it
just so happened that despite the negative quality of the (epi)phenomenal experience not being part of the causal chain (such that, from the perspective of the causal chain,
it might as well have been a positive quality as a negative one), the person anyway developed a totally appropriate intent predicated on that negative quality - to donate to medical research into treatment for experiences like it - which would
not have been appropriate if the (epi)phenomenal quality had instead been blissful.
More complex causal chains predicated on the quality of (epi)phenomenal experiences could be described that would render the "it just so happened" postulate even more utterly absurd than it already is in this example.
I maintain then that my defence of the argument's second premise succeeds, and thus that the argument is sound, and thus that epiphenomenalism is not a correct theory of consciousness.
Potential responses in the literature: #1
I have done some reading to try to see if there are any adequate epiphenomenalist responses to this argument. In particular, I have read (a rereading in part) the SEP's
article on epiphenomenalism, as I first mentioned doing in my earlier post (
#10) in this thread. I also reread the paper that I referenced in that post #10 as having found via the SEP article:
Causation, Sensations and Knowledge by William S. Robinson in
Mind (1982).
It turns out that William S. Robinson wrote the SEP article itself, and the relevant part of the article is addressed in more detail in that paper, so I'll briefly address the paper only. In short, in terms of my example above: in his figures 1 and 2 on page 533, his event e2 corresponds to the physical substrate of the headache, F corresponds to the headache itself, and event e3 corresponds to the physical substrate of the thought "That felt awful". He does not in any way seem to even be aware of the sort of argument I'm making, let alone address it in any way: he simply
assumes the causal appropriateness of e3.
Oddly, though, he
does indicate a keen awareness of the argument in discussing physicalism (as identity theory) on page 535:
"Recall that there is no explaining, from its physical properties, why e2 is an F sensation rather than one that is H or K, and that there is no explaining from any properties of e2 why e3 should be what follows it." It's peculiar that he doesn't recognise this as perfectly applicable to the epiphenomenalist account too!
The superseding of my earlier critique
The above supersedes my response in that earlier post of mine (again,
#10) to the line of thinking presented by William S. Robinson in defence against anti-epiphenomenalist arguments from the need to explain apparent mental causal efficacy, and in particular of the need to explain the apparent causal efficacy of the quality of feelings. I have refined my understanding of what requires explaining: it is not the subsequent mental event (as I'd contended in that post) per se, but the (causal appropriateness of the)
physical substrate of that subsequent mental event.
Potential responses in the literature: #2
Titus, under the section of your paper headed, "The argument from the knowledge of contents of consciousness", you reference (numbered 36) the paper
Meaning and the Impotence Hypothesis by Michael Hodges and John Lachs in The Review of Metaphysics (1979). I read the paper, and, to the extent that it is relevant to my argument, it falls into the same trap as the above paper: simply
assuming the appropriateness of the causal connection. Witness:
"Consider a particular sort of brain event B. Such events will in many cases have, according to the epiphenomenalist, two distinct sorts of effects. In the first place, through a complex set of intermediate causes, B's may generate some observable behavior. The behavior might well be the utterance of 'red', i.e., tokening behavior. In the second place, B's may give rise to mental events of the type Feigl and Meehl refer to as "raw feels." Of course, from the point of view of consciousness only the raw feel and the experienced tokening are evident. The B involved is hidden from view in the recesses of the brain."
Using the phenomenal experience of the colour red - rather than, say, that of the painful headache of my example - obscures that that assumption is being made, but the same principle as in my example applies: when we utter "red", we
mean by it
the quality of our phenomenal experience of the colour, just as the thought in my example, "That felt awful", refers to
the quality of feeling of our headache, and that cannot be derived from the prior physical substrate (whether brain event B or the physical substrate of the headache)
because it is not present in it.
The argument with respect to the SEP and Titus's paper
It seems that my argument above can best be categorised under that referred to in the SEP article as both
Obvious Absurdity and
Self-stultification. In your paper, Titus, it seems to fit both under "Intuitive objections" (which I think, given the above, you've short-changed as an argument), and under (again) "The argument from the knowledge of contents of consciousness", as you'd suggested in your response to me earlier in the thread (in
post #17).
I originally accepted (in
post #19) that that latter section of your paper adequately addressed my earlier post (again,
#10), but, given the above, I no longer do, and I think that here, again, you've short-changed an argument against epiphenomenalism.
Relatedness to psychophysical harmony
There seems to be some overlap between that to which I've referred above as "the mind's causal efficacy" and the "psychophysical harmony" of Sci's thread
The Mystery of Psycho-Physical Harmony, and, in particular, the causal efficacy of the quality of feeling as in my example of the headache seems to be closely comparable to that to which Brian Cutter and Dustin Crummet refer in their paper as "hedonic harmony".
A last way out: God?
The argument of their (Brian and Dustin's) paper suggests that the epiphenomenalist has one final way out of the argument from knowledge of the mind's causal efficacy: the apparent mental causation didn't "just so happen"; rather,
God made it happen.
I don't, though, think that this is a compelling way out, firstly because an epiphenomenal reality - in which mental causation is merely apparent rather than actual - is an unintuitive and unlikely one for a rational God to set up (surely, at the very least, God's mind
itself would/could not be epiphenomenal), and, secondly, because epiphenomenalism is typically motivated by the primacy (and causal closure) of the physical, which is violated by adding (a non-physical) God into the mix.
The evolutionary argument
Moving on: another argument which, Titus, I think your paper short-changes, is that to which it refers as "Argument based on evolution theory". You refer to William James but you seem to too readily discount the strongest form of his argument, when you write (footnote elided):
"With regards to James's argument of the "selective agent" that consciousness would be: this is explicitly attacked by Ray Jackendoff. In reality, Jackendoff holds, it is a subconscious, 'computational' process of concentration and selection of certain information, that would in many cases effectively lead to experiences of conscious attention. The real selection and choice would thus take place at a subconscious level, not based on subconscious objectives and motives, but on their hypothetical subconscious "substrates" (= the hypothetical physiological structures underlying them)."
Again, this does not explain how
the quality of phenomenal experiences have
apparent causal power, which (their phenomenal quality), again,
is not available to that to which you refer as
'their hypothetical subconscious "substrates"' (because it is
epiphenomenal).
A paper I discovered
William James’s Objection to Epiphenomenalism by Alexander Klein in Philosophy of Science (2019) offers this summary:
"The domain-specific objection [raised by James] targets the notion that phenomenal pleasures and pains associated with “fundamental vital processes” could have been shaped by selection", going on to assert that
"James wrote that there is a “set of facts which seem explicable on the supposition that consciousness has causal efficacy” (and not explicable by epiphenomenalism; 1890/1981, 146). The facts in question concern the link we typically find between what phenomenally feels good or bad and what physically benefits or harms us."
In considering counter-arguments, it asks
"when James claims that epiphenomenalism makes the evolution of our native LEPPP [life-essential, phenomenal pleasures and pains] patterns inexplicable, did he simply overlook the possibility that those patterns could have been evolutionary by-products rather than adaptations?
"The answer is an emphatic no. Elsewhere in the Principles
James repeatedly explained other mental traits as evolutionary by-products. He also identified clear and broadly acceptable criteria for distinguishing between by-products and adaptations. When one reads James’s domain-specific objection to epiphenomenalism in light of his criteria for distinguishing by-products from adaptations, a much more powerful objection emerges than has been appreciated in the literature, namely, that our BHBSs [beneficial and harmful brain states] have phenomenal effects that have clear adaptive hallmarks, according to widely accepted standards in biology, and so contra epiphenomenalism, those effects are highly likely to have evolved through direct selection."
Earlier, it had summarised those criteria (footnote elided) in the context of an overall argument:
"if epiphenomenalism were true, LEPPPs could have no “efficacy” and so (recall James’s uncontroversial argument) could have made no difference to reproductive success. But LEPPPs have three features that do suggest that they were shaped by selection: our LEPPPs are (a) natively patterned (they have a characteristic “distribution”), (b) those patterns are systematically linked with antecedent BHBSs (this is the “well-known fact”), and (c) the patterns are “universal” among humans. James concludes that epiphenomenalism cannot make sense of how our natively patterned LEPPPs (with their systematic connections to antecedent BHBSs) could have evolved."
Potential responses in the literature
In response to this argument, I have read a couple of other papers by - you guessed it - William S. Robinson, and in my assessment they fail to counter it.
One is
EVOLUTION AND EPIPHENOMENALISM (2006), in which the crucial passage is this:
"A likely response from those who are sympathetic to James’s argument is that the epiphenomenalist reply falls short, because it does not give an evolutionary account (or any other account) of why N(P) causes P. In this context, however, this criticism amounts to an equivocation on “explanation”. To see this, notice that James’s argument likewise makes an assumption about a causal connection – namely, the one that holds between P and its alleged effects in us. This connection is not explained by evolution (or in any other way)."
Here is essentially an acceptance that the evolutionary argument against epiphenomenalism succeeds, and that no direct counter-argument is available. Instead, an indirect counter-argument is presented: that interactionist dualism has an equivalent problem. I don't think that it does though. I can see at least three ways for an interactionist dualist to provide a sufficient solution.
Firstly, the best response in context is probably: the causal connections from the mind to the brain/body were naturally selected for by evolutionary processes. After all, on interactionist dualism, natural selection
can operate here.
Secondly, the dualist can posit that the causal connections from the mind to the brain/body were designed by a (presumably intelligent) higher power.
Thirdly, the dualist can posit that assuming that the causal connections from the mind to the brain/body are arbitrary/random, embodied beings
learn by trial and error the right way to operate these otherwise arbitrary/random levers so as to achieve the effects that they desire in response to the quality of phenomenal experiences.
The other paper is
What Is it Like to Like? in Philosophical Psychology (2006), but I read it a month or two back and by now I don't recall it in enough detail to summarise what I perceived when I read it to be its failings - and this post is long and overdue enough that I am not going to take the time to revisit it. Others can read it and make up their own minds if they wish to.
Other potentially short-changed arguments?
Your paper perhaps short-changes other arguments, such as those to which it refers as "The argument from the origin of the concept of consciousness" and "The argument from wondering about consciousness", but I haven't followed up on any of the references provided there, so I'll somewhat withhold judgement. I'm inclined, though, to think that the broader claim on the Zombie Meditations page that I quoted in
post #2 is fair:
"if epiphenomenalism were true, no one would ever be able to write about it. In fact: no one would ever be able to write—nor think—about consciousness in general. No one would ever once in the history of universe have had a single thought about a single one of the questions posed by philosophy of mind. Not a single philosophical position on the nature of consciousness, epiphenomenalist or otherwise, would ever have been defined, believed, or defended by anyone."
In conclusion
Your paper, Titus, has been a great springboard for me to explore in more depth the fatal problems with epiphenomenalism - fatalities which equally apply to those variants of physicalism which deny, or imply, the causal impotence of mind - and to work through my own thinking on and formulations of arguments against these positions. The paper's core argument still seems cogent to me, but it also now seems to me - in its minimisation of other sound arguments - to cede waaay too much ground to epiphenomenalists.
![[-]](https://psiencequest.net/forums/images/collapse.png)
