Your complimentary articles

You’ve read one of your four complimentary articles for this month.

You can read four articles free per month. To have complete access to the thousands of philosophy articles on this site, please

SUBSCRIBE NOW

Metaphysics

Epiphenomenalism Explained

Norman Bacrac lets his brain do all the thinking.

“Conscious will is a symptom, not a cause; its roots… are invisible to it… material”
George Santayana, The Realm of Matter (1930)

“A disgrace… more awful than dualism” (Ted Honderich, Philosopher – A Kind of Life, 2001, pp.247, 278); “a dreaded relic” (Daniel C. Dennett, Brainchildren, 1998, p.65). These are just some of the epithets thrown at any position doubting the potency of conscious will. After all, don’t we clearly consciously decide to do something, and then do it? Our conscious will certainly seems to be the cause of our subsequent actions. Yet this piece of common sense is denied by epiphenomenalism, one of the classic theories in the philosophy of mind.

Spanish/American philosopher and essayist George Santayana (1863-1952), however, saw conscious will as only a symptom – the expression of the underlying activity of the brain. Consciousness is a phenomenon arising from and ‘sitting above’ the brain’s action – an epiphenomenon. Consciousness is caused by, but not itself the cause of, the changes effected by the brain.

There is a clear statement of epiphenomenalism towards the end of Thomas H. Huxley’s address to the British Association for the Advancement of Science in Belfast, 1874:

“all states of consciousness in us, as in [brutes], are immediately caused by molecular changes of the brain-substance. It seems to me that in men, as in brutes, there is no proof that any state of consciousness is [itself] the cause of change in the motion of the matter [brain] of the organism.”
(T.H. Huxley, On the Hypothesis that Animals are Automata)

I believe neuroscience is steadily accumulating detailed evidence to justify Huxley’s foresight – especially the close associations seen between scanned images of brain events and subjective experiences. Making straightforward extrapolations from ideas implicit in current brain science, we can extract two propositions which may be considered the two equally-important axioms of epiphenomenalism, and which together formulate it:

Axiom 1. Every conscious state is determined by a simultaneous brain state.

Axiom 2. Every brain state evolves solely in accordance with physical law.

Axiom 1 asserts that all the contents of a conscious state, however complex, are what they are because of the state of the brain. This is not to be thought of as an instance of ‘efficient causation’ in the Aristotelian sense – where the cause precedes the effect – but rather, of Aristotle’s ‘material causation’, where the effect, in this case, the experience, appears simultaneous with its material cause, in this case, the state of the brain.

Axiom 1’s hypothesis of psychoneural correlation was stated by the physicist John Tyndall in 1868 (for more information see my piece ‘John Tyndall’s Materialist Philosophy’ in Ethical Record, Vol 109, 4, 2004). The correlation is many-one, in that many slight variations in brain state could produce the conscious mind state – it is unlikely that the numerous differences at the smallest physical levels, within atoms, would make any difference to brain behaviour, and so what we consciously experience.

Axiom 2 asserts that nothing will take place in the brain which is not in conformity with the laws of physics, which Tyndall again supposed. This axiom is now sometimes referred to as ‘the causal closure of the physical’. It excludes the intervention into the physical realm of any non-physical entity (eg, a soul) and implies that the cause of each brain state is in practice the prior physical state of the brain. Axiom 2 is valid whether physics concludes that causality is deterministic (each state of the universe, that is, of everything in it, being the necessary consequence of its immediately previous state), or whether it confirms the indeterminism in current quantum theory. Quantum randomness is probably too small-scale to cause arbitrary change to the groups of neural switches on which brain action depends, so the brain is de facto deterministic.

In asserting that consciousness is dependent upon the brain, epiphenomenalism is a natural outcome of the materialist tradition in philosophy. Although by adding consciousness it specifies two attributes of matter, it is misleading to classify it as a form of dualism, because it doesn’t propose two entities with independent existence, nor two sources of efficacy.

Matter’s two properties or attributes are therefore:

a) Physical: the unified entity of space-time matter which constitutes the substance of the universe. Physical science, including chemistry and neuroscience, is charged with discovering the nature of the physical world and the laws it follows. Since the time of Galileo, no physics textbook has needed to include consciousness in the equations it expounds; current neuroscience would not be satisfied to take a ‘thought’ or a ‘sensation’ as the efficient cause of a brain event: it seeks instead for the neural (physical) process underlying them.

b) Conscious: an attribute which occurs automatically in human and possibly in animal brains when a certain (presently unspecifiable) neural organisation is established. When this occurs, the conscious content is immediately and certainly known to its subjects – and solely to them, because its content is nothing other than their subjectivity. Consciousness is real but not ‘substantial’ in the philosophically technical sense, as it cannot exist alone, that is, without that small portion of the space-time-matter world in the brain with which it is co-extensive.

Although correlatable with brain states, the contents of conscious awareness – all sensations, feelings and thoughts, including self-consciousness – are not, as many ‘physicalists’ claim, identical to brain substance: “consciousness is not cells” is how Ted Honderich puts it. Theorists who do assert a ‘mind-brain identity’ hope thereby both to dissipate the mystery of consciousness by asserting that it just is the physical brain, and thus guarantee consciousness a causal role. Epiphenomenalism denies this identity, asserting that the physical brain alone is responsible for the consciousness it generates, and for subsequent brain events. Mind-brain identity theorists maintain that consciousness is causal by analogy with the causal role of the temperature of a gas, which can be mathematically proved to be identical to its atoms’ average kinetic energy. Indeed, solid butter is caused to melt by raising its temperature, which turns out to be identical to the process of increasing the kinetic energy of its atoms, but no means are available to prove that consciousness is necessarily identical to neural events. Thus unlike identity theorists, epiphenomenalists can conceive of the logical possibility (in another world) of creatures physically identical to us but lacking sentience: zombies (see Objection 6).

Animal Choices

Unlike a plant, an animal can move about, and therefore has to incorporate a mechanism for making decisions about which way to go. A fly may spot a drop of water, alight near it to take a sip, and then fly off. Unless the Jains are right, insects are not conscious (that is, they can react to the world, but they have no contents of consciousness: there is nothing it is like to be a fly). Nevertheless, we find it quite easy to be anthropomorphic about them. We can say the fly was thirsty, it saw the water, chose to alight because it desired a sip, and took off satisfied. We have spattered the account with conscious intentions, while knowing that neurologists could give a purely physiological account of how the fly accomplished its objective without invoking ‘insect conscious purpose’ or ‘fly free will’.

If you see this account of the fly’s behaviour as reflecting the pattern of human projects, while realising that it was brain physiology that did the work, you risk becoming an epiphenomenalist. Of course human reasoning is considerably more complex than the fly’s, comparing many more factors whose strength we vaguely feel.

The epiphenomenalist sees the process of voluntary action as follows:

Thus a brain state A, having received a nerve signal that one’s throat is dry, gives rise to a simultaneous conscious experience a, the sensation of dryness, coupled with the desire to drink water. A unconsciously initiates (⇒) activity that evolves into brain state B, which gives rise to b, the satisfactory feeling produced when succeeding in one’s freely chosen aim, lifting a glass to drink. Hitting no impediment in raising the glass and sipping the water, B evolves into C, the brain state resulting from a fresh signal from the throat. C gives rise to c, the sensation of having slaked one’s thirst.

Note that the subjective experience of carrying out a decision to drink (a..b..c) is perfectly compatible with that decision’s physical correlates (A ⇒ B ⇒ C) being an integral part of the physical world’s causality. (The brain’s interaction with the rest of the world has been omitted for simplicity.) It’s plausible, but an error, to think that any intention or volition enters as a link in the chain of physical causation – that a physical brain event causes a conscious event, which then causes a further physical event, as if we had A ⇒ a ⇒ B. But as Huxley said, “the feeling we call volition is not the cause of a voluntary act, but the symbol of that state of the brain which is the immediate cause of that act.” (Collected Essays Vol.I, 1894). The author of the Bhagavad Gita seems to share that insight:

All things are everywhere by Nature wrought
In interaction of Nature’s qualities.
The fool, cheated by self, thinks, “This I did”
And “That I wrought”.
(from Ch.3, trans. Edwin Arnold)

If the two Axioms are true, our undoubted effect on the world is not accomplished by a conscious self or ‘free will’ acting by magically originating a sequence of brain events. The I that thinks (in ‘cogito ergo sum’) is therefore perhaps best regarded as the label our brain’s language programme uses when it refers to itself, and not, as Descartes supposed, an independent source of change in the brain. We are causal agents in the world by virtue of being physical beings. Our nervous system has evolved to manage our practical animal needs without our consciously knowing the details. We can work to achieve the future conscious states we desire and diminish the undesirable states. To this end, adapting Friedrich Engels, we shall find that the ‘knowledge of necessity’, ie, knowledge of the physical laws operating in the world, is a condition of our success. Epiphenomenalists proclaim the supreme importance of consciousness while accepting its total impotence.

Some Objections to Epiphenomenalism

1. We have free will, showing that consciousness exerts control over brain activity.

The term ‘free will’ is unfortunately used in two distinct senses, leading to interminable wrangles. I will refer to them as the Primary and Secondary senses.

The Primary Sense: Free will in the libertarian sense alleges we possess a special faculty, the ‘will’, capable of initiating actions from various alternatives. No doubt there is in the brain (or mind) a continual jostling for dominance amongst the contenders. Presumably, the role of the ‘free’ will is to choose between the alternatives and select one action. On what basis is this done? Unless by pure chance (which the libertarians dislike because they discern, rightly, that this eliminates moral responsibility from the act), how can this act of will not automatically select whatever is the strongest motive at that moment, or the argument with the greatest weight? How can it fail to be determined by what one has become and wants at that moment? Perhaps the libertarian pictures free will as a homunculus with free will whose free will is itself a smaller homunculus with free will whose… In fact, the Russian doll ‘solution’.

Because the libertarian doctrine of free will rejects both the possible physical processes for events or thoughts – the causal/deterministic, and its negation, the random/indeterministic – it is beyond the realm of logic to which even the gods (say the theologians) are confined. Libertarian free will is bogus. The neurosychologist Benjamin Libet conducted experiments demonstrating that the brain prepares unconsciously what we are going to think and will before we consciously think and will it. He believed that at the penultimate moment, consciousness could veto any decision – thereby, he said, demonstrating free will in action. However, determinism has never meant our opinions cannot change; and according to Axiom 1, any such ‘second thoughts’ will also have been generated by their fluctuating neural basis. Such thoughts are thus an example of the secondary, but more common, sense of the term ‘free will’:

The Secondary Sense of free will is compatible with determinism. An epiphenomenalist contends that those who claim that free will is compatible with determinism are using the term ‘free will’ simply as a description of the conscious sensation of choosing accompanying a successful voluntary act. But to be effectively realised, the sensation of free choice a..b..c requires the determinism of A ⇒ B ⇒ C.

The sensation of acting freely, according to one’s own desire, is of course highly valued – unlike the sensation of being coerced. The cashier who hands cash to an armed robber can credibly say, “I acted under duress and not of my own free will.” He wants the court to understand the reason for his behaviour, and giving this explanation implies that his behaviour must be seen as determined by it – that is, viewed deterministically.

In dealing with the robber, and wanting to condemn anti-social actions, the law has found it expedient to employ the doctrine of free will in the primary sense: “The robber could have acted otherwise – he misused his free will, and deserves retribution.” But everyone’s current brain state is the inevitable result of its history of that person’s interactions with the world, including every benign and malign influence, so it’s clear the robber could not have done otherwise. Now, although every past event and past human action could not have been other than it was, the expression of our approval or disapproval of them affects future actions, and is necessary for progress. As society comes to appreciate this, it may realise (with reluctance, perhaps) that retributive punishment is unjustified because retrospective. Justifiable sanctions, such as the deprivation of liberty, should always involve attempts at reform, directed to the future – to their prospective effect on the individuals concerned.

2. Epiphenomenalism implies fatalism.

This is a mistake. Yes ‘what will be, will be’ – but this tautology contains no information about the future. No one is privy to a ‘Flash Forward’ (the name of a TV series where people get a glimpse of their future). We can afford to be optimistic, since the future is unknown to us.

3. Surely mental contents cause actions? It seems right to say “I shall visit the dentist because I have toothache,” assigning a conscious experience, the ache, as the cause of a physical action, going to the dentist.

For the epiphenomenalist, the word ‘toothache’ functions as a label in our language for that group of neural processes which are the cause not only of the ache, but which also initiate the process of visiting the dentist. These neural processes are the real cause of the behaviour. The label ‘toothache’, like a good computer file name, aids communication. We shall continue to say “because I have toothache” just as we still say “the sun rose” because these statements still convey our meaning – but not the objective truth.

4. Epiphenomenalism is self-refuting, because if the brain is driven by the inhuman forces of physics, any conclusion it reaches, eg ‘epiphenomenalism is true’, is unreliable, since good conclusions are the result of thought.

By applying Axiom 1, we can see that what appears in our consciousness as the compelling reasons for a particular conclusion corresponds with what at the neural level is the cause of those reasons and that conclusion. Just as a computer’s chess programme necessarily obeys the laws of physics as it derives lawful chess moves, so brains, driven by physics, can exhibit logical reasoning, if educated correctly.

5. Consciousness would not have evolved unless it were advantageous to living organisms. Consciousness must have survival value. It must have been specially selected for, and so must have specific biological effects.

We can have no direct indication of animal consciousness. Other humans are presumed conscious because they are members of one’s own species – bats are not. We therefore have no sure ground for assuming, as we are wont to do, that an animal does what it does because of its possession of consciousness. If a species becomes successful in the Darwinian sense, we have no right to assume that it was guided by pleasure and pain experiences – only by appetitive and aversive reactions. We continue to refer to the alleged causal powers of pain even when we know it is not a cause, for example in the case of a reflex action. That pain is not the cause of the action which ends it was pointed out by Shadworth Hodgson in 1870.

A change in the architecture of an animal’s nervous system resulting from a spontaneous variation in its DNA may increase its relative fitness, giving it a selective advantage. If the new architecture happens to produce consciousness as an epiphenomenon, then consciousness will automatically be passed down the generations without having to be coded for explicitly. If consciousness increases in quality and variety as brains increase in complexity, and animals with complex brains are more likely to survive, then a sufficient condition for the evolution of consciousness will have been satisfied. Not that nature cares for what we may consider its finest product – dolphins may experience sensations beyond our imagination, but living in their evolutionary cul-de-sac the sea, gives them no opportunity to explore sophisticated intellectual realms; and large-brained Neanderthals became extinct in the last ice age.

6. If humans were not conscious this article would not have been written, not least because brains would not have needed to create words for the components of consciousness, nor the word ‘consciousness’. This article’s very existence demonstrates the causal efficacy of consciousness.

The formation of the abstract noun ‘consciousness’ requires the prior existence in language of such nouns as ‘colour’, ‘sound’, ‘taste’, ‘feeling’ etc. These in turn, arise from their constituents, eg particular colours such as green, particular tastes such as bitter, etc. For the epiphenomenalist, a word such as ‘red’ labels not only the sensation but also that sensation’s neural basis, and the word earns its keep by performing a useful function in the brain’s economy.

The brain’s language system could evolve words labelling neural entities whether or not they are experienced as sensation. Thus zombies – physically human replicas without sentience (which, by Axiom 1, cannot exist in this universe) – would invent sensation words. They could say they’d ‘seen a red light’ whenever their non-conscious brains detected photons of a certain energy, even though they never experienced red. Similarly, our sensation of redness is not required for us to speak of ‘seeing red’. This applies to all sensations and the abstract nouns based on them: no sensation is necessary for the brain’s reference to its neural correlates.

The word ‘consciousness’, the collective noun for the totality of sensory and emotional experience, may be culturally important. Ideologies and cults can be built around concepts (eg ‘god’), and wars can be fought for them. Once the word ‘consciousness’ exists, consciousness gets treated in our language as if it has causal powers, but like phlogiston, the fictitious fuel, it has none. Consciousness of one’s ‘self’ is a further instance of the brain’s ability to create concepts (see the Gita above). Yet the power of a concept derives from its human patrons. Its impact on our thinking is not proof that it can move matter. Thus all the detritus left by our talk of ‘consciousness’, and even of ‘self-consciousness’, is no more evidence of their causal effect than a library of books on fairies is evidence of fairy activity.

7. Epiphenomenalism cannot explain how the brain generates consciousness.

True, but at least it acknowledges a problem other philosophies of mind don’t. Eliminativists deny the existence of the contents of consciousness; physicalists award consciousness a dubious identity with neurons to grant it a spurious efficacy; and Cartesian dualists, heedless that it cannot survive a whiff of ether, readily (but unjustifiably) accord consciousness immortality …

In 1868 the epiphenomenalist and Alpine climber John Tyndall anticipated what in 1995 David Chalmers was to call the ‘hard problem’ of consciousness. Tyndall saw an “intellectually impassable chasm” existing between the physics of the brain and the corresponding states of consciousness, even if all the details about both were known: “If we knew that a right-handed spiral motion of molecules of the brain was associated with the feeling of love, and a left-handed spiral motion with the feeling of hate, the question Why? would remain as unanswerable as before.” (‘Scientific Materialism’ in Fragments of Science Vol II, 1889.) Neuroscience has not yet even solved Chalmers’ ‘easy’ problem, which is specific knowledge of the above-mentioned associations. If we eventually reach that stage and arrive at the edge of the chasm of the hard problem, will we have the intellect to find it passable?

© Norman Bacrac 2010

Norman Bacrac is Editor of The Ethical Record and a Trustee of the South Place Ethical Society, Conway Hall, London.

Breakthrough

‘Breakthrough’ © Orna Ben Shoshan, 2010. Orna’s metaphysical work infuses deep spiritual experience with subtle humor, and has been exhibited in numerous locations in the USA, Europe and Israel. To see more of her artwork and buy art or prints, please visit: www.ben-shoshan.com.