Chapter
Nine
Agency and free will
So, is free will always an illusion? Whether free will is what it seems or not, we may
draw one firm conclusion. The fact that we may feel as though we have free will is
not convincing evidence either way.
THE CONSEQUENCES OF BELIEF
A common argument against taking a deterministic view of the universe is that
we need belief in the possibility of exercising conscious will to stop us behaving
immorally. But does believing in free will really make a difference to how we act,
or is the belief that it must make a difference just one more aspect of our illusions
about consciousness?
Levels of belief in free will are high in the few surveys that have been done, with
scales being developed to measure constructs such as scientific determinism,
fatalistic determinism, and perceptions of the world’s unpredictability (Rakos
et al., 2008; Paulhus and Carey, 2011), but reactions can vary with context. For
example, if you ask people whether someone can be free and morally responsible
in a deterministic world, they usually say no. But if you ask people whether John,
who murdered his wife and children so he could be with his lover, can be free
and morally responsible in a deterministic world, they usually say yes. This effect
has been repeated with similar results across a number of different deterministic
scenarios, and in different languages and cultures (Sarkissian et al., 2010).
This difference has been attributed to emotional reactions to John and his
behaviours, which are absent in the abstract case. But a recent meta-analysis of
thirty studies found that the size of such emotional reactions is not large enough
to explain the effect (Feltz and Cova, 2014). Another possibility is that the mental
states of the protagonists are bypassed in the abstract case but may be explicitly
given as a cause of action (John wanted to be with his lover) in the concrete case,
in a way that mirrors how we think about our own motivations and behaviours.
This means that small details of phrasing can make crucial differences in how peo-
ple interpret the statements. If they read a sentence as implying people cannot
act on the basis of their mental states, they give what appear to be incompatibilist
answers. But this arguably has nothing to do with incompatibilism, and every-
thing to do with simply not believing that free will is possible if mental states have
no impact on action.
Despite these difficulties, we can at least conclude that belief in free will is wide-
spread, but does this belief have consequences for behaviour? It might seem that
we could find out by comparing the actions of those who do and do not believe
in free will. But this will provide only correlations and not evidence for causality.
For example, people who tend to cruel or criminal behaviour might be inclined to
reject free will in order to claim that ‘my genes made me do it’ or ‘I couldn’t help
lying’ to avoid the consequences of their actions. Religious believers may behave
better because they believe in hell. What we need is experiments in which belief
is manipulated.
Many such experiments have been done, mostly priming participants by ask-
ing them to read statements provoking either determinist or free-will beliefs.
Some have used sections from Crick’s The Astonishing Hypothesis (1994) (see
our Chapter 7). Others give one group of participants such statements as ‘Sci-
ence has demonstrated that free will is an illusion’, or ‘Like everything else in