Think: A Compelling Introduction to Philosophy (15 page)

I think the best line for compatibilism, faced with this counterattack, is to query the word `set', when there is talk of the modules
being set to produce some outcome. This in effect repeats a similar
move to the one he made to distinguish decision-making from
drowning. There, he introduced a degree of flexibility into the
causal process, by highlighting modules that are capable of being
tuned or set differently. When the objector claims that in that case
the subject is a mere victim if the modules are `set' wrong, the reply
ought to be to introduce another level of flexibility. FRIC, We Call
say, in the case of the brainwashed teenager, or the mini-Martians,
the modules may really he set. We are imagining the modules badly
fixed by chemical or other processes. But these cases are special,
precisely because once they are in them subjects become inflexible:
immune to argument, or to additions or changes in the decisionmaking scenario. But normally agents are not so set in their ways.
Their freedom consists in the fact that they are responsive to new
information, and new differences in the situation. They are not
driven or hound to chuck dogs out of windows or to stand all day
at the cosmetics counter.

We might pursue the idea with something like this, that I shall
call the revised compatibilist defutition:

The subject acted freely if she could have done otherwise in
the right sense. This means that she would have done
otherwise if she had chosen differently and, under the impact of other thoughts or considerations, she would have
chosen differently.

Of course, on an occasion, it may have been had hick that the right
thoughts did not arise. Well, says the compatibilist once more, that
is indeed had luck. But perhaps my anger and the fact that I am
going to thump you will prevent it recurring.

Some philosophers (Baruch Spinoza [1632-77] is the most
famous example) like to associate freedom with increased knowledge and understanding. We are free, they say, in so far as we
understand things. This is in many ways an attractive idea: it ties
freedom of the will to things like political freedoms: freedom of information and freedom of speech. We are only free in so far as we
have opportunities open to us, and lack of information denies us
opportunities. We could add this thought to the revised compatibilist definition, by specifying that the `other thoughts or considerations', first, are accurate representations of the agent's situation
and options, and second, are available to the agent. That is, it is not
much use saying that under the impact of other thoughts or considerations she would have chosen differently, if those other
thoughts and considerations were simply not in the landscape.
Thus, suppose I set about to poison you and cunningly put arsenic
in your coffee. You drink it. It is not much use saying that you were
free not to do so. For although it is true that you would have
avoided the coffee if you had chosen differently, and true that the
thought or consideration that perhaps the coffee was laced with arsenic would have made you choose differently, nevertheless, since
there was no reason for that thought to enter your mind, you were
a victim rather than a free agent. We might incorporate that into a
revised revised compatibilist definition:

The subject acted freely if she could have done otherwise in
the right sense. This means that she would have done otherwise if she had chosen differently and, under the impact
of other true and available thoughts or considerations, she
would have chosen differently.True and available thoughts
and considerations are those that represent her situation
accurately, and are ones that she could reasonably be expected to have taken into account.

What of the person to whom the thoughts or considerations just
didn't occur? Is she a victim rather than a responsible agent? This
introduces a new twist to things.

So far we have talked as if'free choice, either of some mysterious
interventionist kind or of some substitute 'inside' or compatibilist
kind, is necessary for responsibility. But is this right? I said above
that it might be just bad luck that some crucial consideration does
not occur to someone at a moment of decision. But sometimes we
do not treat it as'nmere' bad luck. We say that the thought should
have arisen. The agent is liable to censure if it didn't. Someone setting fire to buildings for fun cannot seriously plead that 'it never
occurred to him' that someone might get hurt-not unless he is a
child or mentally deficient. Even if it is true that it never occurred
to him, so there was no free choice to put people at risk, he is still responsible. Recklessness and negligence are faults, and we can be
held responsible for them, just as much as we are for more controlled decisions. Some philosophers have found it hard to accept
that. Aristotle rather desperately held that negligent people have actually chosen to make themselves negligent, perhaps in early childhood, and that this is the only reason they can be held responsible.

There is actually a whole range of interesting thoughts that open
up here. Some kinds of bad luck are really incidental: things that do
not affect our relationship to the agent. But others in some way cast
a reflection on the agent. Imagine a golfer. Suppose on day one he
hits a fine ball, but, amazingly, a passing seagull gets in its way and
spoils the shot. Then on day two he hits an equally fine ball, but a
little breeze blows it off course and again spoils the shot. We might
say each of these is bad luck. The first is pure bad luck. But the second is not quite so simple. It is bad luck, yes, but the kind of bad
luck that a really good golfer is expected to foresee and play
around. It should be within the player's purview. Whereas the seagull represents a pure act of God. Enough bad luck of the second
kind, and we start to think less well of the golfer, and it is the same
with agency. Hence the reply made by a pianist whose admirer
gushed about how lucky he was to have so much talent: `Yes, and
the more I practice the luckier I get.'

The conceptual engineering we are doing at this point is supposed to tease out or make explicit real elements in our thinking.
We want to highlight and try to encapsulate things like this: we do
make a distinction between changing the past (cannot do) and acting differently than we do (sometimes can do); we do have discriminating practices of blame; we do make a distinction between
being ill and being bad; we do allow some excuses and disallow
others. The philosophical analysis is supposed to give us intellectual control of all this. It is supposed to exhibit it all, not just as
an irrational jumble of disconnected habits, but as the application
of a reasonable and defensible set of concepts and principles. It is
because it is hard to do this that the philosophy is hard. The compatibilist account is a piece of engineering, either plotting our
extant concepts, or designing improved ones. It has to answer to
the ways we often think, or think when we are best in control of the
problems that face us. Myself, I believe that the revised revised
compatibilist definition does that pretty well. But others take
Karat's objection more seriously. They think that our`interpersonal
reactions', which include the ways we hold each other and ourselves responsible for things, do depend upon some lingering affection for interventionist freedom. So if that is metaphysically
bankrupt, our attitudes ought to change. The philosophical problem would he that interventionist control is untenable, and inside
control inadequate.

Sometimes all analysis will settle hard cases. But sometimes it
leaves grey areas, and this may not be a had thing. Return to the
teenage girl spending an incredible amount of time and money on
cosmetics. Can she do otherwise? If we run the revised revised definition, we may find that the issue hinges now on what other
thoughts and considerations are'available'to her. In one sense, we
might want to say, it is possible that she should start realizing that
her popularity or attractiveness is not greatly improved by cosmetics (it would increase more if she got a decent mind, perhaps by
reading a book like this). This may be a true and potentially available thought. But in another sense, perhaps it is not. Perhaps people subjected to the influences she is subjected to just cannot get
themselves to believe this. The culture is awfully good at blinding
teenagers to this truth. So it would not be reasonable to expect her
to believe it. Myself, I would incline to this diagnosis, seeing her as
a victim rather than an agent. But the point is that even if the re vised revised analysis does not settle this issue, it certainly pinpoints
it. And this is itself a step towards getting the issue of responsibility
and freedom under control. But it must in fairness be added that
there is still a road to travel. An incompatibilist, for instance, might
insist that thoughts are only available if they are themselves the objects of free (interventionist) selection, and this would put us back
to square one.

Contemporary culture is not very good on responsibility. Consider the notorious `Twinkle defence'. One day in 1978, an exemployee of the city of San Francisco, Dan White, entered the City
Hall with a gun, evading metal detectors by going through a basement window. He went upstairs, and shot and killed Mayor George
Moscone and a supervisor, Harvey Milk. In court a defence psychiatrist, Martin Blinder, testified that White had been depressed,
which led to his eating too much, and in particular the high-sugar
junk food known as Twinkies. According to Blinder, this further
deepened his depression, since White was an ex-athlete and knew
that Twinkies were not good for him. Blinder claimed that the
emotional state White would have got into would have meant it
was impossible to have acted with premeditation or real intent,
both of which were necessary conditions for first-degree murder.
The jury were impressed by the argument, and acquitted White of
murder, finding him guilty instead of the lesser crime of'voluntary
manslaughter'.

California later revised its law to close the space for this kind of
defence, and on the face of it the state was right to do so. White obviously acted with intention and premeditation, since that is why
he procured a gun and went in through the basement. And we can see that the revised revised analysis is not at all hospitable to the
Twinkle defence. A defendant would have to work awfully hard to
show that enough sugar literally takes our behaviour out of the
range of our decision-making modules and our thoughts. It does
not seem to be true that with enough Twinkies inside us we hecome literally incapable of certain thoughts, so that we could not
reasonably be expected to realize that murdering people is a bad
idea, for example. Even a lot of sugar does not tend to do that. (But
then, contemporary juries are not very good on causation either. In
Michigan recently a man won a lawsuit for substantial damages because, he claimed, a rear-end collision in his car had made him a
homosexual.)

Before leaving compatibilism, it is worth noticing a difficulty in
front of all the definitions. Compatibilism tries to generate the
right notion of control out of the reflection that under different
circumstances the agent would have done otherwise. There are
nasty cases that suggest that these notions do not fit together quite
so tightly. These are called `causal overdetermination' cases. In such
a case something does control some outcome, although the outcome would have been the same anyway because of a `fail-safe'
mechanism. Thus, a thermostat might control the temperature
even if, because of a fail-safe mechanism, the temperature would
have been the same even ifthe thermostat had malfunctioned. If
the thermostat had malfunctioned, something else would have
clicked in to keep the temperature at its proper level. Similarly an
agent might do something had, be in control, be acting with intent
and responsibility, even if were he to choose to do otherwise unknown mechanisms would click in to ensure that he does the had thing anyhow. Imagine the mini-Martians sitting there not actually interfering with things, but ready to do so whenever the outcome looks set to be one that they don't want. These cases are
surprisingly tricky to handle. But the compatibilist can reflect that
they make it no harder to define the right sense of control for
human beings than they do for thermostats. Since the problem
must have a solution in the case of mechanical control, it must have
one for people as well.

OBJECTIFYING PEOPLE

Is there anything else to worry about? One might think like this:

The compatibilist vision describes the operation of organic
beings with brains in terms of decision-making modules.
But all this is just describing things in terms of what happens. It is not describing things in terms of agency, or of my
doing things. It is therefore leaving out something essential
to my humanity, and essential to my human regard for others, which is that we are not just passive patients and victims, but active agents.