A view of my lemon and almond cake. And might that be me and the Philosophy teapot in the background? Rather a stodgy thing (sorry) – the cake, not me – though it seemed to go down well enough with (and into) the French Department later on.
Long time no Zaz.
Brilliant. (French students: translations on my desk, please, by 9am.)
I particularly like this comment:
Bu şarkı olduğu sürece mutsuz olmak imkansız gibi geliyor, diğer şarkılarında olduğu gibi büyüleyeci bir şekilde mutluluk veriyor. Zaza cok şey borçluyum mutluluğu onun şarkılarıyla hissettim. Bi’ gün ZAZ dinledim ve hayatım değişti…
For intrepid classical travellers:
We saw in the previous post on this topic that Determinism is sometimes thought to threaten rationality on the basis that the ‘true’ causal explanations of what happens, in terms of mechanistic causation involving items described in non-psychological terms (molecules, electrical discharges, keys turning etc etc), seem to exclude the possibility that reasons-explanations could offer true explanations of actions.
There are two main lines of argument that would offer solutions to the problem.
One line of argument would be to say that
- reference to reasons can be introduced to explain what people do; but
- reasons are not among the causes of actions.
This kind of reply tends to get dismissed quite quickly by a majority of philosophers, and you won’t find it in the textbook. This is because people assume that reasons consist of psychological states like beliefs and desires, and similar desire-like states; and they assume that such psychological states must have effects on behaviour.
The view that there are reasons, but that reasons don’t affect what happens, is usually rejected on the ground that it involves a commitment to epiphenomenalism about the psychological states. And epiphenomenalism is widely regarded as indefensible.
Epiphenomenalism about psychological states is the view that psychological states are a mere side-show – a pointless accompaniment – to the causal processes in the brain. It is the view that the psychological states are causally generated by the electrochemical processes in the brain; but that they do not themselves cause anything.
This is a very unpopular view, because it implies that your believing and wanting things makes no real difference to what you do. That seems deeply implausible to most people.
It also makes it a bit of mystery why psychological states like beliefs and desires exist, since accounting for them in evolutionary terms is not straightforward. (If they don’t play any causal role, why have they been preserved by processes of natural selection?)
By comparison, it is fairly easy to explain, in terms of natural selection, why there are creatures that have beliefs, desires, and other psychological states, if you assume that those states make a causal difference to the behaviour of their subjects: essentially, the genetic mutations that have led to the occurrence of the psychological states have been preserved, because it benefited their subjects to have those states, in the sense that it made them more likely to survive – and hence to pass on their genes. And being a subject of psychological states, and therefore having conscious awareness of your environment, and of your needs and desires, is intuitively the sort of thing that could confer a survival advantage.
For examination purposes, I think you should agree that there is no mileage in this first type of response. I myself think that there is a version of this response that can be made to stick, and that isn’t guilty of saying anything absurd. But it does involve some fairly controversial moves, including denying the reality of psychological states. (I see it as going hand in hand with my denial that there are actions…) I intend to post something at some point in which I outline how that response might be developed. But, for now, ignore that.
The other line of argument involves trying to explain how it can be that reasons are causes.
On this view, we note that the original problem would be solved if we could say that reasons were identical with – the very same things as – the physical causes. Then we could say that making reference to an agent’s beliefs and desires – to their reasons for action – was a matter of making reference to some set of brain states and/or brain events which were the very same states and/or events that could equally be referred to in non-psychological terminology.
If so, the two sorts of explanation would no longer be competing explanations of action. For they would be explanations referring to the very same causes – but referring to them under different descriptions.
Many things can be referred to in different ways, e.g.:
• ‘Phoenix’ or ‘David’s blue Volvo’
• ‘Richard Dawkins’ or ‘the author of The Selfish Gene‘
So if beliefs and desires just are brain states or brain events, then maybe they too can equally be referred to either in psychological terms (‘A’s desire to get to work on time’) or in purely physical, non-psychological terms (‘neural event XYZ’).
Some philosophers, however, have argued that we cannot identify reasons with physical causes, on the grounds that there seem to be important differences between them. So we need to consider and evaluate the sorts of objection that have been made.
Challenges to the theory that reasons are causes of actions
Argument 1: Causes necessarily produce their effects; reasons don’t
Example: I have a reason to replace the cylinder head (in the form of my desire to make the car work again, and the belief that it will work again with a replacement cylinder head).
But possession of such a reason is no guarantee that I will act on it. I may also have stronger reasons not to replace the cylinder head – for example:
- maybe it is difficult and time-consuming, and I have other more urgent projects that I want to complete;
- maybe the weather is horrible, and I prefer to be indoors by the fire, rather than outside in the cold;
- maybe it is easier and cheaper, all things considered, simply to buy another car.
So I can have the reason without acting on it.
By contrast, once the cause of what I do has been correctly specified in physical terms, it makes no sense (if determinism is true) to suppose that this cause could have been in place and yet that I should not have acted; for causes necessitate their effects.
The argument is flawed. What it shows is that one can have a reason for doing something, on some occasion, on which one does not act (on that occasion). It doesn’t show that, when a person does act for some reason, that reason is not the necessitating cause of her/his action.
The argument’s superficial appeal lies in the fact that we are slow to spot a distinction between tokens and types.
Suppose that I have a broken car on two different occasions. On each occasion, I have reason to fix it, in the form of the belief and desire described above. On one occasion I do so; on the other I do not (because of stronger competing reasons not to fix it).
It is tempting to say that the reason I had for fixing the car was the ‘same‘ reason on each occasion. And this is true, but only in the sense that the reasons were ‘type-identical’; which is to say that they were describable in exactly similar psychological terms: on each occasion, I wanted to have a working car and I believed that replacing the cylinder head would give me a working car.
But because these were different occasions, they were nonetheless different ‘token’ reasons; they were different particular states of mind. They were not one and the same set of psychological states. Rather, one was the reason that I had on the one occasion; the other the (similar) reason that I had on the other occasion. They were therefore numerically different states of mind, occurring on different dates.
Because they were different states of mind, there is nothing to stop you insisting that, on the occasion when I did mend the car, the relevant state did necessitate the action.
Argument 2: Physical causation is law-like; reasons are not
This second argument is closely is related to the first.
To say that physical causation is law-like in its operation is to say that there are exceptionless regularities linking events described as physical causes and effects: that is, there are laws of the form:
Whenever a physical event of type A occurs, a physical event of type B will occur.
There are no such exceptionless regularities linking reasons and behaviour, as the examples above already show. Sometimes a reason will issue in action; sometimes the same reason will not (there may be a competing desire, or whatever).
As for the first argument. It shows only that the language that we use when we describe psychological states is insufficiently fine-grained to pick out differences in the causal properties of two numerically different psychological states that we label as states of the same type.
It doesn’t follow that, on the occasion when I act on my reason and mend the car, that particular reason on that occasion was not the cause of my action. All that the argument shows is that that reason would need to be described in some more precise language – probably the language of physics – if it were to be seen as fitting a pattern of law-like regularity.
Argument 3: Reasons rationalize action in the light of goals; mere physical causes don’t
An explanation in terms of reasons makes reference to the desires and beliefs of the agent. As such, there will be implicit reference to goals. Thus a reasons-explanation is teleological. The action will be represented, in the explanation, as an means to the achievement of some goal that the agent has in mind.
By contrast, an explanation in terms of physical causes will not be forwards-looking or in any way goal-based: it will always be a mere matter of citing earlier events as the causes of events.
So what? If you use two different forms of description of the same phenomenon, it is hardly surprising if these different forms of description achieve different things.
If I say ‘Phoenix is thirsty’, I am talking about the same phenomenon that I may describe by using the words ‘the blue Volvo is running low on petrol’. The first description is of course a playful metaphorical one, in which I anthropomorphize my car – I talk about it as if it were a person.
Suppose that I can find no petrol station, and Phoenix coasts to a halt.
Well, in non-metaphorical language, what happened was that all the fuel was used up and so the engine had nothing to burn; it therefore stopped. No rationalizing; no reference to goals. In metaphorical language, Phoenix’s need to drink was unsatisfied. I rationalize his stopping by reference to his goals. But it is the same phenomenon that I am describing. So the fact that one way of talking rationalizes, while the other does not, cannot possibly show that I am not talking about the same thing.
Argument 4: Reasons-explanations assume some human cultural or social background of rules or practices; physical causal explanations do not.
Example: to explain properly why the bowler dived to his left you need at some point to make reference, implicit or explicit, to the rules of the game of cricket. You need to explain that he was playing cricket and therefore that his aim was to take the batsman’s wicket and/or prevent the batsman from scoring runs. By contrast, an explanation in terms of physical causation will involve no such reference to the rules or aims of the game.
True: but it doesn’t follow that it isn’t one and the same thing that you are referring to when you refer to the physical cause and to the reason.
The solution, in each case, is to notice that what we are doing is describing one and the same set of phenomena in different ways.
It is true that, when we use the language of everyday psychology, we give a different kind of explanation from the sort of explanation that we give when we talk in terms of physical events in a non-psychological way. We explain it in terms of goals, and we rationalise it against some presumed background of social or cultural phenomena.
But it does not follow that the events to which we refer, when we give a psychological explanation of action, are not the very same events as those that we can equally refer to in non-psychological terms.
So we can answer the original objection as follows:
There are not, after all, two competing causal explanations of actions, one psychological, the other non-psychological. Both explanations refer to the very same causes and effects. It is merely that one explanation uses the non-psychological language of the physical sciences; the other uses the language of everyday psychological description.
Note: The view defended above is, broadly speaking, the view developed by the American philosopher Donald Davidson in a series of articles on actions and events published in the 1970s and 1980s.
Human actions are often explained by reference to reasons, where reasons are understood in psychological terms.
I am rolling pastry because I want to make a pear tart, and I know that you can’t make a pear tart without rolled pastry (and I haven’t got any pre-rolled pastry).
Notice, in passing, that reasons-explanations are ‘goal-based’, or ‘teleological’: in giving a reason for someone’s action, we specify some desire, or goal, or aim that the agent has – and that the action is aimed at fulfilling. Since this is so, all reasons-explanations look to the future, in that they are stated in terms of the agent’s desires or goals in acting as (s)he does.
This fact is sometimes disguised by the fact that we sometimes don’t state reasons in full.
For example, someone might reply to the question, “Why are you replacing the cylinder head?” simply by replying, “Because it is broken.”
There is no reference to any goal or desire here. The answer refers not to any future goals, but only to the present state of the cylinder head.
But this is only because the reasons-explanation is incomplete. After all, the mere fact that something is broken is not, in itself, a reason for fixing or replacing it: one might be perfectly content for it to remain broken, in which case one would have no reason to do anything about it.
The fact that something is broken is only a reason for fixing it if one additionally wants it not to be broken any more. So the answer given, “Because it is broken”, is really a shorthand for something like, “Because it is broken, and so the car won’t run properly in this condition; and I want the car to run properly.” We often omit the reference to our desires or goals, because it is often too obvious to need stating; or because we are so familiar with this kind of explanation that the goal can be left implicit in what one actually says.
Reasons-explanations of the kind described rationalize our actions. That is to say, they make sense of them in terms of reasons.
Replacing the cylinder head is a rational thing to do, if you know that you have a broken exhaust valve and if you want the car to run properly again.
Rolling pastry makes sense if you want a pear tart, know what is needed, and know that an unrolled block of pastry is in your possession.
The explanation, in each case, rationalizes what you are doing, in the light of what you know and want.
1. Does determinism threaten rationality?
A question that arises under the syllabus is whether determinism would undermine or threaten (human) rationality. Annoyingly, this is not an especially helpful way of putting the relevant question, and it is not immediately obvious what the problem is meant to be.
What they have in mind is the following:
Does determinism, if it is true, exclude the possibility of explaining human actions in terms of reasons?
2. How this question arises: competing causal explanations of action
As explained above, we often explain our actions, and those of others, in terms of reasons.
Furthermore, a widely-held theory in the Philosophy of Mind is that reasons for action are also the causes of actions. So, at least in our ordinary talk about actions, we explain them by reference to reasons, thought of as mental states like beliefs and desires; and we think of such explanations as giving accounts of the causal background to actions.
This picture may seem to be threatened, however, when we realise that if determinism is true, then the true causal explanation of all events can be stated without reference to mental states of the agent.
What is the explanation of the fact that the engine is running at 6.30am?
If determinism is true, then a complete causal account of this fact is available in terms of earlier states of the world and micro-physical events, especially brain events: towards the end of the explanation we would get something like the following:
…various electrochemical events occurred in A’s brain, including the firing of neurons XYZ, which triggered nervous impulses causing muscles in A’s hand to contract, which led to the key’s turning in the ignition, which caused the starter motor to run, turning the flywheel and causing fuel and air to be drawn into the cylinders, where it was compressed by the pistons and ignited by the spark plugs. The consequent explosions depressed the pistons, so that …. blah blah blah…
So, on the one hand, if determinism is true, all events have explanations that can be stated in such terms as these – terms that make no reference to anyone’s beliefs, desires or intentions.
On the other hand, since the engine’s running is the product of a human action, it seems also to have an explanation that is stated in psychological terms – in terms of reasons:
The engine is running because the car’s owner wanted to get to work on time and believed that turning the key and starting the engine at 6.30am was an appropriate way of getting to work on time.
So we have two proposed causal explanations of the same state of affairs. But this state of affairs, surely, cannot have two different causal explanations, both of which are true. That would be to suppose that every state of affairs generated by a human action was actually overdetermined. (An outcome is overdetermined if it has two (or more) different causes, each of which is sufficient on its own to produce the outcome.)
The supposed threat to rationality is then this:
If determinism is true, then the true causal account of the original of every action must be the one stated in non-psychological terms – the account that refers to physical events in the brain, and so on; and that, since this is so, the other (psychological) account cannot be correct: there cannot be room for more than one true causal explanation of any event.
So the concern is that determinism might exclude the possibility that human actions can properly be explained in terms of reasons; and that (in that sense) determinism would threaten human rationality: if the true explanation were not in terms of reasons, then reasons could not properly be cited as the explanations of why we do what we do.
Haven’t we been forgetting the view known as libertarianism?
We spent a fair amount of time looking at Incompatibilism, including Hard Determinism and the other view that I insist on making space for (the view that I call Agnostic Incompatibilism).
We have also looked at a number of approaches within Compatibilism (also labelled ‘Soft Determinism’).
But we merely mentioned, without seeming fully to examine, what we called ‘Libertarianism’.
Libertarianism, remember, is an incompatibilist view: libertarians hold that, if determinism is true, then we lack free will. They share this belief in incompatibilism with both Hard Determinists (people who think that determinism is true, and that we therefore lack free will) and with Agnostic Incompatibilists (people like me who say that they have no idea whether Determinism is true or not (for it is an empirical, scientific matter, and I am not a scientist but a philosopher), but that if it turns out that Determinism is true, then we lack free will).
Libertarians, then, are incompatibilists. But they assert that we do have free will. And so they are committed to the view that Determinism is false.
Indeed, the main task of libertarianism is to try to explain how it might be that we could have free will, given the truth of incompatibilism.
The principal libertarian manoeuvre is to say that, although material events may be subject to deterministic causation, not all things are material. We have material bodies; but we also have minds (or souls) that are immaterial and that are therefore not bound by the deterministic causation that operates in the material realm. Since our minds are not subject to determinism, and since our minds are the sources of our actions, we are, after all, free.
This view is known as ‘dualist’ libertarianism because it adopts what is known as ‘dualism’ in the Philosophy of Mind. Dualism is the view that there are two sorts of substances: (i) material/physical and (ii) immaterial.
(Dualism is opposed to ‘Monism’, which is the view that there is just one kind of stuff. There are in fact two different monist views. One is ‘materialism’, according to which everything is material/physical – there is nothing to us other than our material nature: the existence of mentality consists in nothing other than the operations of our physical brains in the physical world. The other (less common) monist view is ‘immaterialism’: it is the view that there is no material nature at all: everything that there is is immaterial. This second view obviously involves saying that our ordinary belief in the existence of material objects is in error.)
According to dualist libertarianism, our immaterial souls are our minds: they are thus the sources, or the locations of our thoughts; and our actions flow from the mental events (thoughts, intentions, desires, decisions) that occur in our souls.
Actions, therefore (according to the dualist libertarian), are free because they are the products of entities (minds/souls) that are not subject to deterministic causal processes, and are therefore free.
As agents we can therefore make free decisions (ones which are not necessitated by earlier events) by which we can make future events (including physical events) go one way rather than another. In doing so we create new chains of causes and effects in in the physical world.
• I am confronted by coffee cake and lemon cake. These physical objects impinge on my senses in the supermarket and lead me to think about which to buy;
• I ponder the matter with my immaterial mind, and reach a decision;
• I act, moving my physical body so as to grasp the lemon cake and remove it from the shelf.
Central to this picture of things is the notion that my immaterial mind – being outside the deterministic system – creates breaks, or gaps, in universal causation. Nothing in the surrounding physical circumstances forces me to choose lemon over coffee cake – I alone do that.
And when I make my decision and act, I make later physical events occur one way rather than another way. I am the starting point in a chain of physical causation that begins with the removal of the lemon cake from the supermarket shelf. And so I have acted freely and in a way that makes me morally responsible for what I do.
Two types of objection to dualist libertarianism
There are essentially two different types of attack that critics of dualist libertarianism may make:
1. They may attack the libertarian’s embracing of dualism, on the ground that dualism faces insuperable philosophical objections.
2. They may accept dualism (or at any rate they may allow it for the sake of argument); but attack the claim that dualism, even if it is correct, yields either freedom or moral responsibility for action.
Consider these two types of attack in turn.
1. Objections to dualism
There are many objections to mind-body dualism. Here are a few.
(i) Interaction problem
On any dualist view, there has to be two-way causal interaction between immaterial mind and material world:
• The material world must have effects on the mind: for example, perceiving material objects is a causal process in which material events (light reflecting from the physical object, compression waves (sounds) in the air caused by the physical object, etc) have effects on the mind (we have perceptual experiences).
• The mind must have effects on the physical world: mental events occurring in my immaterial mind (such as my decisions to act) are able to move my material body (I decide to take the lemon cake, and reach out with my arm, which then moves the cake to my basket).
The problem is that such causal interaction seems incredible.
We have a pretty good understanding of how material events can cause other material events. To get a cake, or an arm, or a spanner, to move, you have to apply a physical force to it that is sufficient to shift this mass from a stationary position.
But we have no understanding at all of how you might get a material thing to do anything without applying some material, physical force to it. An immaterial thing has no mass, bulk, size, charge or anything of the kind; so it seems quite impossible for it to exert any force capable of getting a material thing to move.
Nor do we have any understanding of causation in the opposite direction – of how you might use a material thing to ‘move’, or otherwise affect, an immaterial one.
The problem is exacerbated by the fact that, if minds are immaterial, they would seem to have to lack spatial properties: material bodies can exist in space, but it is not clear how an immaterial body could. (If it isn’t made of matter it can’t take up any room; but if it doesn’t take up any room it isn’t in space at all.) But then it seems an insoluble mystery where causal interaction between mind and body is supposed to take place.
(ii) Location problem
The lack of spatial properties makes it equally a mystery where immaterial minds are supposed to be located at all.
And this seems linked to a further problem: what is it that makes my mind mine, and yours yours? The answer cannot (it seems) be the tempting suggestion that mine is in, or attached to, my body; for my mind isn’t anywhere. (Anyway, how do you go about attaching an immaterial thing to any material thing? What sort of fastener would you use?) But then what is the answer?
(iii) Violation of the law of the conservation of energy
Dualist libertarianism, if true, would force us to abandon well-supported scientific laws about the conservation of energy. Those laws imply that a complete causal account of what happens in the physical world can be given that does not require positing the addition of new energy from outside the physical system.
But if the dualist’s story is true, then my decision to choose the lemon cake must be an event outside the physical system; and if that decision genuinely makes a difference to what happens within the physical system, then its effects on the physical world must amount to the injection into that world of new energy from outside it.
(iv) Increased understanding of the relation between brain events and mentality
Although brain science is still in its infancy, the results of such science are already giving us evidence of what looks like a clear dependence of mental functioning on brain functioning.
We therefore have an increasingly well-supported hypothesis that the functioning of the mind, in all its intricacy, is dependent on the brain; and hence a plausible hypothesis that there is nothing more to the mind than the brain – that all of mentality can be accounted for without requiring anything other than material objects.
This materialist hypothesis seems to offer the most plausible explanation of data such as that damage to specific parts of the brain can damage specific aspects of mental functioning. More generally, if my mind is immaterial, then why does injury to the brain cause unconsciousness?
2. Even if we allow dualism for the sake of argument, does the libertarian’s view give us freedom?
Even if dualism were plausible, it is not clear that it really helps the libertarian anyway.
If I do ever act freely, I surely act for a reason. (If I act for no reason at all, then what I do seems not even to qualify as a real action at all, let alone a free one.)
If I choose the lemon cake, then my reason is (something like) the fact that I prefer lemon to coffee cake (at least on this occasion; as an incidental autobiographical matter, I in fact prefer lemon cake to coffee cake on every occasion).
But if I act for a reason, then my reason (it seems) must be sufficient to make me act as I do.
This observation is the application to the mind of a more general principle: it is part of our ordinary understanding of the world that everything that happens happens for a reason. This is known as the ‘principle of sufficient reason’. Applying this principle to the thoughts and decisions that we make in our immaterial minds leads to the conclusion that whatever decision we have made, there must have been a reason why we made that decision; and that reason was enough to make us decide as we did.
Now: If we could exactly replicate the situation in the supermarket as it was before I chose one cake rather than another, then surely I would have exactly the same reason as before.
And if that reason was sufficient to make me act as I did first time around, then surely it must be sufficient in the identical circumstances of the replicated situation to make me do the same thing again.
In which case it seems that the world will still be deterministic, even if my mind is immaterial.
The lesson seems to be as follows:
First, the supposition that we have immaterial minds does not, in itself, mean that those minds are not subject to determinism. So even if we are dualists, determinism may still be true. The thoughts and decisions made in your immaterial mind may be just as much necessitated by earlier events and the laws of nature as any events involving material substances and objects.
Secondly, an important consideration about reasons for action (the principle of sufficient reason, as applied to human decisions and actions) suggests that, if we have immaterial minds, then our decisions and actions are in fact necessitated by earlier events and the laws of nature.
The only obvious way to avoid this conclusion is to suggest that my reasons aren’t sufficient to make me do what I do. But if that is true, then there are two huge problems for the claim that dualism gives us freedom.
First, I don’t seem to be in full rational control of what I do, which doesn’t sound much like freedom.
Secondly, it would appear to be a random matter whether my reasons lead me to do this or that. But (as noted elsewhere) randomness seems not to be a plausible basis for freedom.
I said yesterday (in class, not here) that my view is that there are no actions. Time for an explanation.
It is not merely that actions cannot be identified with the events that we call bodily movements. My view is that they cannot be identified with any events at all. For there is no such thing as an action.
A simple ontology
To explain why, it may help if I first say a little bit about what I think does exist.
Some people, possibly, would say that this ontology is hopelessly flawed. But even if it is, I would point out that a commitment to the ontology itself is not essential to the main arguments against the reality of actions that I go on to present; those arguments require only the widely-shared assumption that events, at least, are real. The inclusion of a summary of the assumed ontology is just intended as a helpful background to my way of thinking about reality.
Reality consists of nothing more than the existence of certain individual entities that persist through time (atoms and particles, and the composite objects that they constitute).
As a matter of completeness, I should add that if I am honest I don’t really believe in the supposed ‘composite’ entities either, whether natural or man-made (tables, lemon cakes, Volvos, people and other animals, Mount Vesuvius, the planet Saturn…). You don’t exist; and neither do I. What I really think is that only the most fundamental particles are real.
But this is an additional, and inessential, element of the view that there are no actions. Besides, since actions – if there are any – are essentially the actions of people, if I were to insist from the start that there are no people, it would be a very short argument to the conclusion that there are no actions. I want to present an argument against the reality of actions that doesn’t depend on denying the existence of people.
Properties and events
What about the properties that entities possess, and the events that occur? Aren’t these real?
Well, yes, in a sense, there are properties and events. But if I were to say that I was a realist about events and properties as well as entities, that would tend to make it sound as if I thought that reality is more full of real ‘things’ than I mean to imply. For my view is not that there are three sorts of thing – objects, properties, and events: my view is really that there is just one sort of thing: the only real ‘things’ are the entities.
Properties and events are not real ‘things’ over and above the particles and the composite entities. Both properties and events ‘come for free’ with the existence of the entities that there are; they are not ontologically additional to those entities.
Properties as ways of being
The world cannot contain an entity at all unless that entity ‘has properties’. The notion of an entity that had no mass, no size, no shape, no position, no movement, no colour, no temperature, no electrical charge, nor any other property, is, I think, incoherent. A supposed ‘entity’ that lacked all properties would in fact be nothing at all. So ‘properties’ are automatically there, given entities.
‘Having a property’, therefore, is not nearly as grand as the label makes it sound: it is not like possessing a valuable object, like a lemon cake, or a gold ring, or a slave-girl; indeed, it is not like possessing any object – not even a cheap and nasty one, like a rusty old Lada or a toenail clipping. For it is not a matter of possessing any object at all: rather, it is just a matter of being one way rather than another. ‘Having the property’ of being green is just a matter of being green rather than any other colour, or no colour. ‘Having the property’ of being a Volvo is a matter of having been made by one motor manufacturer rather than by another, or by no motor manufacturer at all. Properties are not ‘things’ (objects, entities); they are ways of being.
Events as changes
Events are not ‘things’, either. An ‘event’ is just an occurrence. So typical events are changes in the properties of things (e.g. one particle’s moving closer to, or further away from, another; Nicolas’s changing from being 7 years old to being 8 years old; Genevieve’s changing from being a functioning car to being a non-functioning one, and back again).
Events also include the coming into existence, and the departure from existence, of things like cars and people and other objects. Genevieve’s manufacture, for example, was an event that took place in 1996: in that year, a thing (Genevieve) began to exist that did not previously exist. So understood, an event is not an entity like a mountain, or horse, or car, or person – not even a tiny one like a molecule, an atom, or a quark. An event is just a change in the world – a change in the complete inventory of the things that exist, or in the ways that those things are.
Insofar as events and properties are ‘real’, then, there are two main points that should be stressed: first, events and properties are not persisting entities, but rather aspects of entities, and/or modifications in them; secondly, their ‘existence’ depends on the existence of entities. Events and properties are ‘ontologically parasitic’ on the existence of particles and composite objects: they are dependent on them for their existence; and they contribute nothing to the existence of those particles and composite objects.
So the overall picture is as follows: there are various entities (physical particles, and (if you believe in them) the composite objects that they may constitute). These entities are all unavoidably some way or other – they ‘have properties’; and they change in various ways (and these changes we label ‘events’). There is thus a ‘hierarchy of being’ in my ontology: the only ‘really real’ entities – the ‘free-born citizens’ of my universe – are the particles and composite entities; properties and events are second-class citizens – slaves and attendants, ontologically speaking, of the free-born.
Where are the actions?
When I deny the reality of actions, what I am really saying is simply that none of the particular ‘things’ mentioned in the overall picture of reality can be identified as the ‘actions’ that people claim to believe in. An action is not any individual entity, or property, or event. Nor can any individual action be identified with any larger set of the ‘things’ mentioned in that picture. An action is not any plurality of entities, or properties, or events, nor any mixture or combination of these items.
When I say that there are no such things as actions, this is not to be understood as the (outrageous) claim that no one ever does anything. People do, of course, do things – that is, if we assume for the sake of argument that people do exist.
Nouns that don’t refer
My denial that there are actions is merely the claim that, although our language has a noun ‘action’ that enables us to talk about what people do in terms of the ‘actions’ that they perform, the term ‘action’ never refers to any thing that is listed in the complete inventory of what there really is, nor to any larger set of such things.
As an analogy: one can speak of doing something (paying a bill, making a speech etc) on someone else’s behalf, or of doing something for someone else’s sake, without thinking that the terms ‘behalf’ or ‘sake’ refer to any entities. My claim that there are no actions is analogous to the claim that there are no behalves, and no sakes: when a person does something (boils an egg, mends a car, falls in love, etc), I think that the aim of identifying any real ‘thing’ (entity, property, event) in the world that is the person’s action is an error of the same kind as trying to identify the ‘thing’ that is my behalf, when you speak of doing something on my behalf. We use nouns, or noun-phrases, in apparent reference to actions: ‘Smith’s action’; ‘A’s killing B’; but there is nothing in the world to which such nouns and noun-phrases refer.
Why, though, do I refuse to identify actions with any of the ‘things’ whose existence I do admit?
Actions would be events, if they were anything
Start by noticing that, if actions were to be identified with anything, it would have to be with certain events. An action, if it is anything at all, is an occurrence rather than an entity or a property. It is something that happens, rather than a persisting object, or a way of being. Actions, therefore, could not realistically be identified with entities (particles or composites), nor with properties. At most, they might be identified with certain of the events that occur.
Why actions aren’t events
So why do I refuse to identify actions with any of the events whose existence I allow?
Mainly because any attempt to make such an identification, in the case of any particular ‘action’, seems to run into insuperable difficulties. (We have already come across some of the obstacles to such identification, in discussing arguments for the non-identity of actions and bodily movements; but there is, I think, more to the problem than those points alone.)
The following is a small selection of the sorts of difficulties that seem to exist.
1. Actions and contexts
Consider, first, the fact that actions depend, for their being the actions that they are, on a background of shared social and cultural practices. An action is the action that it is only because of that background.
Merely moving your hands and arms in a certain way, for example, does not on its own suffice for executing the gesture known as the ‘quenelle’. Before the French comedian Dieudonné introduced and popularised this movement as a gesture, a person who moved his or her arms in that way would not have been making that gesture.
So it seems that the events are one thing, and the context another. The action – performing the quenelle – requires both the events (the bodily movements) and the context – for without the context of the social practice, it could not be the action that it is, but only a movement that was wholly devoid of meaning. Hence the action cannot be identified with the events of bodily movement.
Or consider my action of incurring a liability to income tax. This is something that can be done only by earning an (actual or deemed) income; but doing that, alone, does not suffice for incurring the tax liability. It is also essential that there be unrepealed legislation, passed by the relevant authority in the relevant jurisdiction, according to which my earning the income makes me liable to pay income tax. So you cannot (it seems) identify my action of incurring the liability with the events that (somehow) constitute my earning the income; for that would be to omit something that is essential to my action.
Now, there is a possible reply to these arguments about cultural and social context.
An objector could reply by asserting that events are individuated not by reference to their intrinsic features alone, but also by reference to the total context in which they occur. So the event or events which is/are my receipt of an income qualify as being those particular events only because they occur in the context of the facts about existing applicable tax legislation that in fact obtain; and those facts about applicable tax legislation are facts only because of other, earlier, events involving the legislature.
Hence, if what we loosely call the ‘same thing’ (receipt of income) had happened in a different context (one in which there was no relevant tax legislation in force to which my income was subject), my receipt of the income would have been a different event. It might have been intrinsically indistinguishable (the same amount of money comes into my possession, and for the same reasons), but (since events are individuated by reference to matters that are extrinsic to them) it would nonetheless have been a different event: a case of ‘receipt-of-income-in-the-absence-of-certain-income-tax-legislation’ as opposed to a ‘receipt-of-income-in-the-context-of-the-existence-of-such-legislation’.
Similarly for the case of the gesture. We could say that the events in which your hands and arms move thus and so, in the context of Dieudonné’s contribution to social practices, are different events from those that would be occurring if your hands and arms were moving in exactly the same (intrinsic) way, but in the absence of the social context.
This reply would solve the problem presented by the fact that context is essential to an action’s being the action that it is. It solves it at the cost of commitment to the view that every event depends, for its identity, on all other events; but that is a price that some will be willing to pay.
Other problems, however, are less easily solved.
2. Temporal dimensions
Take, for example, the well-known problem of temporal dimensions.
- A shoots B;
- B dies 24 hours later; and
- in the interim (say, 12 hours after shooting B), A himself dies.
If there is such a thing as A’s action of killing B, when did it happen?
No answer, I think, is remotely acceptable:
If you say that the action has occurred by the time A’s finger has moved and the bullet has entered B’s body, then you have to say that the killing is over before B is dead. And that is absurd.
But if you say, instead, that the action of killing extends further in time, so that it is not complete until B is dead, then you have to say that A’s action continues after A’s death; and that is no less absurd, for it implies that A is still acting when A is already dead.
Denying the reality of actions solves the problem at a stroke. If there is no such event as A’s action of killing B, then there is no pressure to say precisely when it happens.
All that there is to be said, I think, is that A did something that caused B to die. This makes it true that A killed B. The fact that he did this consists in the fact that certain events occurred; and the fact that he did this is the same fact as the fact that he ‘performed the action of killing B’, as we sometimes put it. But that ‘action’ is not to be identified with any such set of events; and hence it has no duration, and there is no precise answer to the question when it happened. For there is no such thing as ‘that action’.
3. The relevance of conditionals
Consider my action of waiting for a train on some occasion; and imagine that you have made a complete list of all the events that occur in the relevant locale (presumably, the railway station) on the relevant day.
Which of these events can be identified with my action of waiting for a train?
The whole set? Certainly not: for (since there were countless other people also in the station, all of them going about their own business of waiting for trains of their own, selling tickets, buying and selling coffee and newspapers, driving trains, making announcements, waiting for friends to arrive on other trains, and so on) the complete inventory of events in the station includes countless events that have nothing whatever to do with my action.
Can we, then, identify my action with some subset of the total inventory of events in the station?
I think not. The fact that I am waiting for a train consists not just in the occurrence of the events in the station, but also, in part, in the truth of certain counterfactual and other conditionals: for example, that if my train arrives, I will board it (other things being equal); that if an announcement were made that all trains were cancelled, I would leave the station (other things being equal); and so on.
Any proposed identification of my action with any subset of the events inside the station must inevitably omit the relevance of these conditionals to the fact that I am waiting for a train. But if so, the proposed identification will have missed its target: it will not have picked out my action, but only some of the events that are in some sense relevant to it.
There is also the problem of what I call idle action, or active idleness, or inactive action, or active inaction. (In fact I can’t yet decide quite what to call it.)
Stay with the station example. Suppose that I have arrived rather early at the station, so that there are several hours for me to wait before the expected arrival of my train. So I am waiting for my train for several hours. I don’t at any point in that period cease to be waiting for my train, as I would if I were to abandon my intention of catching it. So at all times during that period of hours I am waiting for my train.
Since I have several hours to kill, it is unsurprising that I get involved in various other activities. Perhaps I buy a newspaper with the idea of tackling the crossword. Maybe I go to the station café to have something to eat and drink. Possibly there is a rather delightful waitress in the café with whom I fall into conversation… maybe she is coming to the end of her shift so that we are able to continue our conversation in the warm Spring sunshine… and then I wake up. I must have fallen asleep in the café, and the business with the waitress was just a dream… I look at my watch, and see that there are still another two hours to wait before my train is expected.
The point is that, since I have been waiting for my train throughout the whole period, the action of waiting for a train doesn’t have to involve any events that are in any particular way directed towards catching the train. I don’t have to be constantly watching the departures board, listening to announcements, or scanning the railway tracks. I can get on with other things quite unrelated to train-catching. I can even go to sleep. So I don’t need to be doing things that are train-related; I don’t need to be doing, really, anything at all. I can be more or less completely idle; but it is nonetheless true of me that I am waiting for a train. All that has to be true is that the conditionals referred to above remain true. So I can do something (wait for a train) without doing anything.
If actions are events, then the fact that I am waiting for a train means that there are some events going on that are identifiable as that action. But if I am not doing anything relevant to catching a train, but merely sleeping, then there seem to be no candidates for identification as the events that are my action of waiting.
The simple resolution of this apparent paradox, it seems to me, is to deny the reality of actions. If actions aren’t any events at all, then one can be doing something without there being any events that are one’s doing that thing. And so it doesn’t matter that one isn’t doing anything, and that there are no plausible candidate events for the identification.
5. Negative actions
Consider, finally, what might be called ‘negative actions’, like stopping smoking.
The fact that a person is stopping smoking consists purely in the fact that (s)he smoked in the past but has now ceased to do so.
With which event might the person’s action of stopping smoking be identified? There aren’t any particular events that are specially relevant to the stopping, so that they deserve to be identified with it, whereas others are not relevant and are not part of the action of stopping. All that happens is that the person does countless other things, none of them being smoking tobacco. (S)he doesn’t do anything special; (s)he merely omits to light another cigarette.
So it seems to me that you cannot identify the person’s action of stopping smoking with any event or events, unless you are willing so say that the person’s stopping smoking is identical with the set of all the events in this person’s life after the moment when (s)he extinguished the final cigarette.
But that would, surely, be absurd.
If that absurdity is not immediately apparent, suppose that the person simultaneously gives up both tobacco and alcohol. You would have to say that each of two actions – the person’s action of giving up smoking and his/her different action of giving up drinking – was identical with the set of all the events in that person’s life after the relevant moment. But two things cannot be one and the same thing. So you would have to say that the person’s giving up smoking was the same action as his/her giving up alcohol. Surely, however, it is not: there are two things that this person does, not one: she gives up both tobacco and alcohol, and these are different achievements.
What is really going on
Our talk of actions is, I think, just one particular kind of descriptive ‘overlay’ that we place on the real world in which persisting individuals exist and change. The reality is just that particles exist, persist and change.
We, however, cannot help seeing this rather scant-sounding reality as containing larger composite individuals, including people, cars, newspapers, railway stations, and so on.
So one mistake that we make is that of being misled by our ‘take’ on the world into believing in these larger composites.
Furthermore, we see some of these larger individuals as bringing certain things about (killing one another, mending cars), and as involved in various other performances (falling in love with one another, waiting for trains, stopping smoking).
Now, insofar as the larger individuals are real at all, it is true that people act in these ways. I think it is strictly speaking false to say that they do, for strictly speaking there are no such larger individuals. And so the talk of people’s doing things is an ontologically erroneous, but understandable, way of describing the world with which we are confronted.
Because we see the world in this way, we think and talk of people’s doing various things. We would not, of course, see things in this way if there were not, at the bottom level, real particles and real events constituted by changes in the natures of these particles. Hence the association that we make between the occurrence of events and the supposed occurrence of ‘actions’ is inevitable, given our way of looking at the world.
But the language of ‘actions’ is an instance of nominalisation – forming a noun from a verb – in circumstances in which there is literally nothing to which that noun can refer. So even if you believe in the composite entities that we call people, and even if it is true to say that those people ‘do things’, there are still no actions. For all ‘action’ terms are nouns or noun-phrases that lack a reference.
Note to my students: As I said, this view is certainly not compulsory. Indeed, you do not need to have any view – realist or anti-realist – about whether actions are real for the purpose of your syllabus. (But slightly bizarrely, in these circumstances, you are required to have an opinion on whether actions can be identified with bodily movements.)
I just think that it is interesting, and since I mentioned it yesterday, it seemed fair to outline what it is that I think. I’m working on some of my thoughts in this area for the purposes of the Austrian conference paper – or that is the idea at present; so any comments, criticisms etc would be very welcome.
Explain and illustrate the claim that our experience is intelligible only because we possess a predetermined conceptual scheme
To say that our experience is ‘intelligible’ is to say that it ‘makes sense’ to us. It is thus to say that, when we are experiencing, we understand what it is that our experiences represent to us.
It is clearly true that we do make sense of (at least most of) our experiences in this way. For example, when I have a typical perceptual experience, it seems at once to me that I am confronted with a range of objects which I understand:
Visually: I seem to be confronted with a number of recognisable objects (tables, chairs, people, etc) that stand in certain causal, spatial, and temporal relations to me and to one another. For example: it looks now as if there is a table which is supporting a laptop, a bottle of red wine, and a glass. The level of the wine in the bottle has been gradually decreasing over time as I repeatedly fill the glass and consume the wine.
As to the auditory aspect: there are sounds coming from the radio that I understand as the words of politicians and other public figures stating their views on various issues of public interest. For example, I hear a man defending the Scots’ alleged right to self-determination.
If I were a mere device for recording light and sound, which was conscious but lacked a conceptual scheme – a conscious combined digital camera/voice recorder – then I would be sensitive to light and sound in my environment, but I would not experience reality as containing objects, or people, or words/arguments. I would not understand why the area filled with red colour in my visual field kept diminishing. My ‘experience’ would consist of no more than an undifferentiated ‘blur’ or ‘buzz’ of sounds and colours.
The fact that my experience is not like that suggests that I am not a mere undiscriminating recipient of perceptual data. Rather, there must be some feature of my perceptual system that ‘filters’ and interprets the raw incoming data so that my experience is intelligible in the way described. The claim that we have a predetermined conceptual scheme is the claim that there is, built into us, a system that ‘filters’ the raw data, somehow subsuming it under concepts including those of object, space, time, and causation, so that we are able to have experiences as we do, rather than as the imagined conscious sound and light recorder would.