Lecture Six: Mental Causation[1]

The general public tends to believe that the mind and the brain are the same thing. So given there is a mental state of believing it is raining, there will be a neural state, let’s call it ‘state n’, that IS the belief it is raining. Similarly given the mental state of experiencing red, there must be a neural state 'state s’, that IS the experience of red[2].

Those who believe this will not have any trouble with mental causation. If mental states are physical statesthen mental causation is just another form of physical causation. This means it will fall comfortably under RTC, CTC, STC or some hybrid account of causation. There will be noparticular problem of mental causation.

The public’s view of the mind might be right[3]. But it is harder to justify this view than you might think. Let’s start by having a look at why this should be so.

Problems For The View That Mental States Are Physical States

Broadly speaking there are two types of mental state[4]:

  1. Qualitative states
  2. Intentional states

Qualitative states include pains, experiences of red, love or jealousy, and emotions like hatred or anger. The property that characterises these states is their possession of a subjective quality to which the subject has ‘privileged access’. The qualia of an experience is essential to its being the experience it is[5].

Beliefs, on the other hand, are intentional states. The property that characterises beliefs is their content. It is in virtue of their contents that beliefs are true or false, and can enter into rational relations. The content of a belief is essential to its being the belief it is[6].

The possession of qualia or content is essential to a mental state’s being the mental state it is. If it didn’t feel bad it wouldn’t be a pain. If your belief wasn’t a belief about me, it wouldn’t be the belief that it is. All mental states have either qualia or content.

Physical states on the other hand, do not have qualia or content. At least they do not have them essentially. The only property that physical states have essentially is their extension – the fact they fill space, that they are three dimensional.

Mental states, however, do not have extension, at least not essentially. They do not have size or mass, and they cannot be next to each other, on top of each other or inside each other (except metaphorically). Mental states are not characterised by their spatial relations to anything (including each other).

So we have a two-sided dilemma:

  1. states are not mental unless they have qualia or content. Physical states have neither qualia nor content.
  1. States are not physical unless they have extension. Mental states do not have extension.

How can states with such different properties be identical? How can they be the same type of state? However obvious it might seem to Joe Public that mental states are physical states, it looks very much as if mental states are notphysical states.

The Problem of Mental Causation

But if mental states are not physical states, the problem of mental causation arises. If mental and physical states are quite different then how can theycausally interact[7]?

In the 40s and 50s some thought that mental states were contingently identical to physical states[8]. But in the 70s Saul Kripke came along and showed us that they couldn’t be; that if mental states were identical to physical states then they were necessarily identical to physical states[9]. This is so counterintuitive, we went back to the drawing board[10].

In the 1950s and 60s others thought that reasons couldn’t be causes because reasons (beliefs and desires) are linked logically to their behavioural effects[11]. We cross the road because we want an ice cream, and believe there is an ice cream van over the road. Hume, of course, showed us that there is no logical link between causes and their effects (no logically necessary connection between cause and effect).

But the use of the word ‘because’ in “we cross the road because we want an ice cream….” suggests that we do think of reasons as causes. We certainly talk as if we do.

It was Donald Davidson who showed us that reasons can be causes[12]. He did so by distinguishing between causation (a metaphysical relation between events), and explanation (an epistemological relation between descriptions of events). He pointed out that the statement ‘the cause of G caused G’ may be unexplanatory (because the descriptions ‘the cause of G’ and ‘G’ are logically linked), but it is still true.

Statements of the form ‘A caused B’ are made true by a causal relationship between A and B however A and B are described. But only certain descriptions of A and B will generate an explanation. So whilst it is true that:

“The event described on page 3 of today’s Times caused the house to collapse”

we only have an explanation of the house’s collapsing when we know that the event described on page 3 of today’s Times was an earthquake.

Reasons, said Davidson, are causes because reason explanations are a species of causal explanation. But whereas ordinary causal explanations make sense of an event as one that fits with our picture of nature as uniform, reason explanations make sense of a behaviour as one that fits with our picture of the agent as rational.

The Principle of the Uniformity of Nature and The Principle of Charity

According to Davidson reason explanations are special because they are constrained by the Principle of Charity(PC)[13],[14]. This tells us that our explanations of others’ behaviour can be deemed correct if and only if they portray others as rational. Evidence of irrationality and of false beliefs are, according to Davidson, evidence of error – our error not that of our subject.

This contrasts with normal causal explanation where our explanations can be correct if and only if they portray nature as uniform. Evidence of irregularity is, for a normal causal explanation, evidence of error. Ordinary causal explanations are, in other words, constrained by the Principle of the Uniformity of Nature (PUN)[15].

So reason explanations are causal explanations, albeit a special type of causal explanation in virtue of being constrained by the PC rather than the PUN.

Davidson also went onto argue that this would enable us to show that token mental states are token physical states, even if we cannot (for all the reasons above) identify types of mental states with types of physical state. Davidson believed that this would solve the problem of mental causation.

Token Identity One: Anomalous Monism[16]

Davidson points to an‘inconsistent triad’[17] of statements, and notes that as all three of the statements are true, the appearance of inconsistency must be explained[18]. He argues that the best way to explain it is to assume that every mental state[19]has a physical description – i.e. that every mental state is a physical state.

Here is the ‘inconsistent triad’:

1)All causation is law-governed

2)There is psychophysical causation

3)There are no psycho-physical laws

That 1) is true is suggested by almost every theory we have considered[20].

That 2) is true is suggested by our everyday and scientific practice. We do what we do because we believe what we believe, and because we experience what we experience.

3) is the only one of our triad that seems questionable. Davidson argues for it by appeal to his claim that reason explanation is governed by the PC. When reasons are implicated in the production of our behaviour, they are rationally so implicated. A belief isn’t just a cause of behaviourit is a reasonfor behaviour.

Because of the ‘disparate commitments’ of reason explanation and ordinary causal explanation, says Davidson, we will never be able to formulate laws combining mental and physical predicates – they are not, says Davidson, ‘fitted’ for one another.

But, he says, if all causation is law-governed, and there are no psychophysical laws, then the laws governing psychophysical causation must be physical laws. This tells us that every mental event must have a physical description. It is in virtue of this description that it falls under a physical law. By such means Davidson argued that mental states are physical states despite appearances to the contrary.

It is important to recognise that Davidson’s is a token identity theory, not a type identity theory. There may be causal laws relating states that are as a matter of fact mental with states that are as a matter of fact physical, but there are no bridge laws relating mental predicates to physical predicates.

Objections were almost immediately brought against Davidson’s claims, but before we look at them let’s look at another way of showing that token mental states aretoken physical states.

Token Identity Two: Functionalism[21]

The functionalist claims that mental states are functional states[22]. The functionalist believes that mental predicates are defined in terms of the causal role they play in ‘folk psychological’ theory, the theories we use to explain each others’ actions. So a pain is a state caused by tissue damage, and that causes moans, attempts to repair the damage, beliefs that one is in pain etc.

On this story mental states are ‘second order states’; to be in pain is to have inside you a state that plays the functional role characteristic of pain.

The important point about functionalism is that it permits ‘multiple realisability’. If c-fibre firing plays the role of pain in human beings, and we said that pain is c-fibre firing, we would have to deny that dogs can be in pain (at least assuming that dogs don’t have c-fibres). Functionalism says that pain is whatever state plays the functional role that defines pain. It might be realised by c-fibre firing in human beings, d-fibre firing in dogs, and something quite different in Martians[23].

Functionalists claim to have solved the problem of mental causation because pain is by definition that which is caused by tissue damage, and that causes moans, attempts to repair the damage, beliefs that one is in pain etc. How could there be any problem with mental causation if mental states are defined by their causal role?

A problem for functionalism is the fact that if mental states are defined by their causal roles, then there is a logical link between pains and their effects. This cannot be undermined in Davidson’s manner because there will be no independent properties to which we could be referring. According to functionalism pain is exhausted by its causal role[24]. Some functionalists respond to this by thinking of the pain as the physical realiser the causal role, rather than the causal role itself[25].

Non-Reductive Physicalism and Mental Causation

We can appeal to either AM or to Functionalism to argue that mental states are physical states. And this might seem to resolve the problem of mental causation too – after all if mental states are physical states why should there be a problem of mental causation?

But there are two problems[26] that haunt both types of non-reductive physicalism: exclusion and externalism.

The Exclusion Problem[27],[28]

Imagine that Fred’s belief at t that crossing the road will get him an ice cream iseither token-identical to, or realised by, neural state N. It is, says those who bring the exclusion problem, the fact that neural state N has the physical properties it does that causes Fred to cross the road, not that neural state N is a belief that crossing the road will secure an ice cream. It is the physical properties of the state that are causally efficacious, not its mental properties.

This argument is based on two premises:

P1: that physics is complete (i.e. every physical effect has a physical cause)

P2: there is no overdetermination (i.e. we only need one sufficient conditions for an effect to have found its cause)

But given these two premises, the fact that the physical properties of neural state N suffice for Fred’s road-crossing leaves no job for the mental properties to do. The causal efficacy of the mental properties of state N is excluded by the causal efficacy of its physical properties[29].

There are numerous responses to the exclusion problem – you can judge for yourself whether any of them are satisfactory. You will find them all in the SEP entry on Mental Causation (references below)

The Externalism Problem[30][31]

Intentional states have content – they represent the world. Arguably this ensures that intentional states are not the sort of states that are inside us. They are states of the sort we get into. They are, in other words, externally determined[32].

The content of your beliefs are a function not just of their intrinsic properties, but also of their relations to the environment, history, society, culture and your community. You could not have the beliefs you actually have, without your being located as you are actually located in the world.

If contents are essentially properties extrinsic to the subject, however, then again we have a causal problem. Surely it is only the intrinsic properties of our brain states that are causally efficacious in the production of our behavior?

Consider a vending machine. When you put a pound coin in it, the machine responds only to the intrinsic properties of the pound coin. It is oblivious to the fact it is a coin of the realm, that it has the value of £1, that it is part of the British system of sterling etc. None of its extrinsic properties are relevant to its causal efficacy.

Surely what is true of the vending machine is also true of us? We may have beliefs that essentially have content. And this content might be essentially a function of our environments. But if so, surely this means that the content of our beliefs is not causally efficacious in the production of our behavior?

How can content, if it is essentially external, make a difference of any kind to our behaviour?

Again there are numerous possible responses to the externalism problem, and again I will leave you to judge for yourselves whether any of them are satisfactory (SEP Entry on Mental Causation again)

Something Wrong Surely?

It looks as if it is extremely difficult to justify the claim that mental states are causally implicated in the production of our behavior. Everywhere we looks our mental properties seems to be screened off from, or excluded from, the possibility of playing any causal role at all in our behavior.

But if we do not do what we do because we believe what we believe, and because we experience what we experience, then how can we justify the claim that we are choosing what we do? How do we justify the claim that we are responsible for what we do? How, indeed, do we justify the claim that we are rational?

Of course there is a long tradition of denying that we choose our own behavior and that we are responsible for what we do. The tradition is called determinism[33]. According to it we do not have free will, and everything we do is causally determined by the laws of nature and the environmental conditions in which we act[34]. On this story if we have moral responsibility at all, then we do not have it in virtue of choosing our own behaviours.

There is a more recent tradition of denying that we are rational: Eliminativism[35]. But some philosophers, and cognitive psychologists are claiming that folk psychology is a false theory and as such that it should be eliminated along with all its theoretical states and relations. According to the Eliminativists, we have reason to believe we do not have beliefs, and that we are not rational.

Perhaps you are attracted by such views? I am not. But if we are to reject them we must find some account of causation according to which mental states are causally efficacious in the production of our behaviours. As we have seen we do not yet have any candidate that has made its case to the satisfaction of any significant numbers of philosophers.

Marianne Talbot

University Of Oxford

Michaelmas 2015

1

[1]Robb, David and Heil, John, "Mental Causation", The Stanford Encyclopedia of Philosophy (Spring 2014 Edition), Edward N. Zalta(ed.), .

[2] The Internet Encyclopaedia of Philosophy on Identity Theories of Mind:

[3]The Stanford Encyclopaedia of Philosophy on Mind-Brain Identity Theory:

[4] Jim Pryor’s excellent summary of the nature of mental states:

[5] The Internet Encyclopaedia of Philosophy on qualia:

[6] The Internet Encyclopaedia of Philosophy on intentionality:

[7] This problem arose with a vengeance for Descartes who believed that minds and brains are different substances see Descartes, R: Meditations On First Philosophy, 3 & 6, or Williams, B Descartes: The Project of Pure Enquiry, chapters 4 & 10. Or the IEP: These days few people think minds and brains are different substances. But we still think they are different properties, or different aspects of some physical substance.

[8]U.T. Place, Is Consciousness a Brain Process

[9]Kripke, S: Naming and Necessity:

[10]It is counterintuitive because if pains in humans are necessarily c-fibre-firings and dogs don’t have CFF then dogs can’t be in pain.

[11]Melden, A. I., 1961, Free Action, London: Routledge & Kegan Paul

[12]Davidson; ‘Actions, Reasons and Causes’ (1963) in J. Phil 1963, and also in Davidson’s Essays on Actions and Events (OUP, 1980)

[13]If you want to learn more about the Principle of Charity try this:

[14]

[15]This phrase is taken from Hume.

[16] The Internet Encyclopaedia of Philosophy on Anomalous Monism:

[17]Statements are inconsistent if they can’t all be true together.

[18]Davidson, D., “Mental Events”, inEssays on Actions and Events, Oxford: Clarendon Press (1980).

[19]Davidson argued that only intentional states are truly mental. He believed that qualitative states are in fact physical states. In this he echoed the French philosopher Brentano, who first said that intentionality is the ‘mark of the mental’.

[20] STC could reject the idea that laws are at all involved in causation – but it would have to insist that causation is always observable in the individual case.

[21]The Internet Encyclopaedia of Philosophy on Functionalism:

[22]The Stanford Encyclopaedia of Philosophy on Functionalism:

[23]The Internet Encyclopaedia of Philosophy on Multiple Realisability: