Sunday 5 January 2014

Why dualism and ephiphenomenalism are incoherent positions

There is a widely accepted principle called 'causal closure' which says that what happens in the world has a complete physical explanation. There are no gaps in the world for magic. The behaviour of the pieces of our world are well-understood, even though the behaviour of those pieces in large numbers can be intractable: we know the physical principles that are involved in water and air molecules and yet long-term weather prediction may be forever beyond us. That intractability doesn't mean that any new physical forces or substances are involved: weather is only molecules doing what molecules do.

Causal closure has implications for what we can sensibly say about our minds, because no matter what we think or believe or feel, causal closure means that there is a complete physical explanation for those mental states. Causal closure also means that whatever we say about our mental states has to have a complete physical explanation too – after all, the transferring of mental states into movements of the mouth, tongue, throat and lungs is a physical process! It might seem pendantic to go into such detail about what is involved in speaking, but the sheer physicality of our conversations about the nature of mind is an important part of understanding what is going on in such conversations: it's a mistake to think of human conversation as hovering some way between the physical and mental realms – it's a physical event, known particles and forces doing what they do, albeit in vast numbers.

The physicality of conversation means that it is subject to the laws that apply to all physical systems, and some of those laws deal with the nature of information. The activity of the brain is information processing, this information being present as the states and connections of brain cells. Each cell is a switch with tens of thousands of inputs. The signals from the inputs are summed and the cell either sends out its own signal or not depending on that sum. Cells can also change the number and character if their inputs based on aspects of the signals they receive. The brain is a vast and complex web of information processing units, yet there is nothing more going on than millions and billions of cells communicating and changing connections (although 'nothing more' seems dismissive of the richness and complexity of what brains actually do).

How does information end up in physical brain states? There are limited possibilities. One way is through the activity of brain cells: existing brain states can be transformed to new brain states. Another way is via sensory input: a variety of physical inputs from the world are encoded by sense organs into neuron activity and this is fed into the brain via nerves. Those are the only ways. Thinking and imagination have to be brain state transformations because of causal closure, and sense inputs are brain state transformations because of external signals. Of course, sensory input can itself contain coded information, as in speech or writing, but the purely physical nature of that information remains, no matter what the degree of encoding.

These limits of the possible sources of information about what is real are a real problem for any kind of dualism or epiphenomenalism, because these limits prevent thoughts from representing knowledge of dualism or epiphenomenalism. Thoughts can, of course, represent beliefs about things that are or might be real. Through some amazing co-incidence such beliefs might be true, but they can never be true because of knowledge. When someone comes to believe that some kind of extra non-physical aspect of mind exists, that change in belief cannot be caused by any property of that aspect, because a change of belief is a change in information, and that requires physical interactions: given causal closure, none of the reasons for believing in any non-physical aspects of mind are good reasons.

Some positions about consciousness consider that consciousness could exist as a parallel form of reality that exactly follows the physical. An example of this is pan-psychism, the idea that every particle in the universe has some minute quantity of consciousness, but less broad suggestions of the relationship between consciousness and the physical are rather more respectable. However, even these positions which respect the physical and causal closure fail in terms of knowledge. If some aspect of reality is completely mapped on to the physical, then it cannot differ in information content from the physical: every state change of the physical is accompanied by a parallel state change in the non-physical. This means that the non-physical aspect of reality cannot contain knowledge of itself, only beliefs about itself. It's not possible, because of causal closure, for either the existence of the non-physical or the relationship between the physical and non-physical to have causal influence on the physical, and so knowledge of the existence of even this relatively respectable form of semi-dualism is not possible – beliefs in this dualism cannot be justified.


Dualism and ephiphenomenalism are not just dead but well-buried. It's hard to justify the belief in such aspects of reality as even coherent. It may be reasonable to talk of 'what it is like' to be a conscious being, but, given causal closure, it isn't coherent to talk of 'what it is like' as being any justification for there being a non-physical extra part to minds. It just doesn't work.

6 comments:

Diacanu said...


It's incoherent, because the people positing this bullshit really don't give two shits about science and philosophy.

The end goal isn't to revolutionize science, and make a magic dualism box that makes e-mails send faster.

The goal is to keep religion shoved into every sphere of life for as long as possible.

Steve Zara said...

That's not true - I know of at least two decent atheist philosophers who support some sort of epiphenomenalism!

Quine said...

The "little bit of consciousness in everything" reminds me of saying that every molecule of air has a "little bit of tornado" in it. This represents a serous lack of understanding of the nature of emergent phenomena. There may be some levels of complexity re consciousness, but a system has to have a minimum capability to get going.

Steve Zara said...

Yes. I like Ramachandran's view of consciousness - self-empathy. You need at least more than the mental apparatus to feel empathy. Atoms don't have that!

Anonymous said...

I know nothing, but can't epiphenomenalism still be valid even if causal closure were true?
The brain states and changes in brain states are caused by sensory input. These brain state transformations cause physical movements and changes in our body.
At the same time, these weird complex brain cell relations and interactions are so complex that they evoke a feeling of consciousness and such in us. Our consciousness is caused completely by physical changes in our minds. However consciousness is still not a physical thing, it is an emergent phenomenon.
I don't know if I'm making any sense at all, and I only understood half of this article, but I don't get exactly why causal closure ensures that epiphenomenalism is not possible.

Steve Zara said...

Thanks for the question. Causal closure means that all influence is physical. That means that the feeling of consciousness cannot be due to something which is beyond physics. Emergent phenomena are physical things, just as, say, waves on the sea are physical things. There's a difference between not being physical and not needing anything extra to exist - waves are physical even though they don't involve anything more than water molecules.

I hope this helps.