Greg Detre
Thursday, 16 November, 2000
Mr Snowdon, Mind V
McGinn discusses the mind as having two components, one passive, one active, will and perception. Functionally, our usual perception-action loop of �Perceive the current state of the world. Act accordingly. Then perceive the current state of the world now�� requires both elements of mind. But if we were to sever the connection, could we imagine a mind equipped with only one of perception or action?
Certainly in evolutionary terms an animal that can act without a perceptive faculty would be hopelessly maladaptive � it would run into trees, get swallowed up with ease by predators, be unable to find food etc. But is it even conceivable? Such an animal would have no contact with its external world, certainly. But even if it had no predators and could itself automatically on thin air, it would be unaware of itself. It would get no somatic feedback, so it would not know what it wanted or needed. But if it wasn�t even aware of its own acts of will, then it seems that we cannot really ascribe such a being with mentality at all. A mental life consisting wholly of acts of will would have no sense of self, or time passing, or desires. I would be reluctant to describe it as mental at all.
On the other hand, McGinn considers that perhaps an animal without will is conceivable, i.e. we can imagine that God could create one if he chose. A hapless animal without will would be able to see the hungry predator strolling towards it, might even have time for its life to flash before its eyes, but would be helpless to fight off the predator, to run away or even to hide. We might imagine a mushroom with eyes, living in a field of hungry, mushroom-loving mushroom-eaters. Unlikely to evolve, and very unlikely to survive unless something changes, but still conceivable. Galen Strawson addresses the question with his lengthy discussion of the �Weather Watchers�, beings with mental life but no means whatsoever of exhibiting outward behaviour. They might desire that the weather that they watched continuously be a certain way, but he argues that this desire does not entail a desire to act, to influence the outcome. But to talk of mentality without will is an even stronger constraint than to talk of mentality without behaviour. A mental life without will would be robbed of the ability to direct its attention or to initiate a train of thought. Though Strawson is persuasive in the case of his behaviourless beings, beings with mentality but without will might be more akin to rocks. After all, if we were to take some sort of extreme functionalist viewpoint that sees consciousness as arising out of the process of neural activity in some way, then why might not all of nature�s processes be conscious to some degree. This leads to an inverse statement of the problem of other minds � how do we know that everything around us isn�t conscious? We should perhaps remind ourselves that thoughts spring into consciousness unbidden, and that even if we can somehow subconsciously vet them or gauge their direction, we are in effect observers to our own thought processes. Crucially though, we can certainly direct our attention with relative freedom and we can end or attempt to consider separate trains of thought, and so we won�t digress any further about the wholly different case of a mentality with no will whatsoever.
In both cases, conation without cognition or vice versa, we find that we can conceive of the one aspect dominating the other, but the idea of being equipped for one entirely without the other is inconceivable. So we can state a conceptual, as well as an evolutionary imperative for their intimate relatedness. As McGinn concludes, to emphasise one at the expense of the other misses their inseparability.
He moves on to discuss actions, starting with the basic action of raising one�s hand, and contrasts it with an identical-seeming movement resulting from, say, a muscle spasm. We would want to regard the two as different, differentiable by how active or passive the �agent� was in initiating the action. When I raise my hand, I consciously choose to, there is volition. When my hand is flung upwards by a gust of wind, this aim for the action to occur just before it does, is lacking. But to simply label the distinction, �volition�, does not really help us. Indeed, it is important that we don�t just regard human movements as actions � animals and babies are capable of actions. An animal that leaps over a river is performing an action, whereas an animal whose body is catapulted into the air by an impact is not. There need not be a rational, expressible reason for an action, though often we can make sense of it in rational terms. Indeed, here we are coming closer to pinning down actions: they are bodily events which originate in the agent which can be accorded a purpose. I raise my arm to get the blood flowing or to wave to a friend.
The problem with defining action in purely teleological terms is that we can see a purpose in my stomach�s digestion and my heart�s beating. Yet we would like to exclude them from willed movements because they are involuntary, i.e. they still lack this mysterious �volition�. It is obviously enough to simply talk of observable events, because my reflex of withdrawing my hand from a hot object is a bodily event that has a purpose (not burning my hand) and is observable � but though the gross movement was generated from a multitude of motor signals originating in my brain, it did not somehow satisfy the requirement of �[originating] in the agent�.
We are seeking a means of restricting willed actions more narrowly than simply purposive behaviour and less narrowly than rational, human intent. We can use the criteria of �trying� � whether we can non-metaphorically say of the agent in question that it is trying to do something without being guilty of over-anthromorphising it in some way. Neither worms nor thermostats are trying to achieve the purposes we ascribe them, though it is convenient and predictive to take what Dennett terms the �intentional stance� in pretending that they do. We can identify �actions� by being able to describe an agent as �trying� to do something. This is narrower than simply being able to see their behaviour in teleological terms, and wider than claiming they are being guided by full-blown rationality.
This highlights two halves to an �action� � the willing and the movement, i.e. the mental and the physical acts. Every action has this inherent psychophysical duality. Yet when I psychologically try to move my arm (i.e. aim to produce that movement in the way that I do every day as a direct precursor to that movement actually happening), but find that nothing happens, I have still tried. How then is the trying related to the action? There are four theories, all taking the definition of action as a �willed movement�:
1. The action is the movement. But in order for the movement to have been counted as an action, �trying� had to have been one of its antecedents. The action and movement are identical, and the trying was not part of the movement, so it cannot be said to be part of the action. But, it is a requirement that for a movement to be considered an action, it must be preceded by trying. Thus, movements and actions are intrinsically the same, but distinguished by their relations to an extrinsic object.
2. This time, an action is identical with the trying. In this case then, the resulting bodily event is a contingent extrinsic matter. The trying and the action are one and the same, but a trying becomes an action by virtue of its consequences (i.e. a corresponding bodily movement).
Both these theories can be dismissed as regarding either the trying or the movement as being contingent and extrinsic, but actions need to be more closely related to both the psychological and bodily aspects. �An action is no more the cause of a movement than it is the effect of a trying�.
3. Action is successful trying. Thus, both trying and movement are constitutive of the action. This way of thinking about actions encompasses the whole psychophysical process, from the psychological, inner mental event of trying to raise my arm, all the way through to the interface with the physical which results in my arm actually rising.
Although this seems to be along the right lines, thinking about actions as successful tryings results in tryings as contingently successful and actions as necessarily successful. So actions necessarily have an outer aspect whereas tryings do not.
4. The action is composed of both trying and movement, related causally. The action is a complex event with these two constituents. Thus the action is related to the trying and the movement, but constitutively and not causally. This does require a dualist ontology of mind and body as a means of unifying the first and third person aspects of action.
We can try and examine more closely what is meant by �trying�. Is it a sensation or a propositional attitude? Unlike sensations, it is active, and it doesn�t have a phenomenological feel (it doesn�t really feel like anything to try). Yet, like sensations, trying is necessarily conscious and also pre-rational. Moreover, trying is not a propositional attitude (I try to, not that, I raise my arm). Hence, trying cannot be categorised within either of the cognitive classes, and has to be ascribed a nature of its own.
Like perception, trying is directed on something beyond itself. In the case of trying, its object is the body, or at least the parts where we have motor control. However, it cannot be said that I try to move my arm, so much that my trying is directed towards a pointer which refers to my arm. But could we perform willed movements without this acquaintance with our body which seems to derive from our proprioceptive and kinaesthetic perceptions? If we were deprived of our (internal perceptual) awareness of our body, then would we able to direct our trying towards it still? If the answer is �no�, then a creature with will but lacking the perception to form this proprioceptive awareness would be unable to act, to move.
It remains to investigate the antecedents of action. Minimally, these include some feeling of need and some sort of belief state which connects an action with fulfilling that need. For rational creatures, the antecedents of action are desire, belief and intention. The desire provides the motivation for the action, the belief state is that a causal relation holds between the motivation and the action, and the intention provides the resolve to proceed. None of these three antecedents can be reduced or expressed in terms of the other two, though desire and belief combined make up a �reason�.
In Davidsonian terms, for an event to be an action, the event must be describable in a specific way, i.e. an event that an agent performs with intentions and for reasons. The same action may be intentional under one description and not under another. For an action to be an event, we must be able to conceive of at least one description under which it is intentional. Davidson claims that reason explanation is a form of causal explanation. In this way, reasons are the causes of the actions they are reason for. The solution to the conflict depends on the distinction between events and their descriptions. The explanation of an agent�s action can only be considered adequate if it shows the action in question to be reasonable (i.e. consistent with) the background of the agent's beliefs and motivations. This conditioned, combined with the idea that propositional and rational attitudes attributed to an agent must be true (the truth condition) form the necessary conditions for the justification model of explanation. The explanation of action by reference to reasons is also a form of causal explanation, i.e. when we refer to an agent�s intentions or motives in acting, we are attempting to offer a causal explanation of that action. Explanation by reference to reasons was seen to be distinct from explanation by reference to causes as rational explanation was held not to involve reference to laws, but instead required showing how the action fitted into some larger pattern of rational behaviour.
Significant in Davidson�s account is the notion of a �primary reason�, comprised of the belief relating to the action and the desire. A second important idea is that of action �under a description�. This simply means that one single action can always be explained in terms of more than one correct description. This is important as it provides an explanation of how the same action (item of behaviour) can be understood as being intentional under some descriptions and not under others.
If we understand the connection between reason and action as rational, then it cannot be described in terms of a strict law. However, in understanding such a connection to be causal in nature, we are assuming the existence of some law-like regularity. Such a law-like connection would not by classical definitions of causality also be rational. Davidson claims that such a nomological connection would not be describable in the language of rationality and can therefore be causal even though it does not specify any strict law. He maintains therefore that rational explanation need not involve explicit reference to any law-like regularity, while nevertheless also maintaining some such regularity must exist due to the fact that there is an underlying causal property to the rational connection. Moreover, since Davidson resists the idea that rational explanations can be formulated in terms of a predictive science, so he seems committed to denying that there can be any reduction of rational to non-rational explanation.