FREEDOM, DETERMINISM, AND MORAL RESPONSIBILITY
(with an Appendix on Freedom and Foreknowledge)
William A. Wisdom
March/April 2004
This paper will be an exercise in critical thinking about a cluster of
notions that are more often used than carefully considered. This
cluster of notions is interesting in part because it lies at the point
where the three main branches of Philosophy meet: Ethics, Metaphysics, and Epistemology
or the Theory of Knowledge. I will address the questions of what the
terms "freedom", "determinism", and "moral responsibility" mean--or
might mean, or ought to mean--and how they are related to each other.
While this is a philosophical article, I mean to avoid professional or
technical jargon. That is, I mean to write in plain English, though the
concepts considered and the distinctions made may require close
attention to follow. I can't help it that I'm a philosopher. But
philosophical thinking is no different from ordinary thinking, except
that it tries to be clearer, more rigorous, and more accurate than
everyday conversation. I agree with David Hume's dictum that
"philosophical decisions are nothing but the reflections of common
life, methodized and corrected". Of course that final qualification is
crucial. Without it, philosophical discussion would be so much careless
and often erroneous chitchat. We'll do our best to avoid that.
(By the way, I intend to write about freedom rather than about free will. For reasons that I hope I can make clear, the notion of free will makes little or no sense to me.)
There are actually two papers here. The first is the one on "Freedom,
Determinism, and Moral Responsibility". As an appendix to that paper, I
attach a closely related one on "Freedom and Foreknowledge".
___________________
The issues of freedom and determinism, and their relation to moral
responsibility, might preliminarily be illustrated by this bit of
fiction. I get angry at my psychotic partner
for destroying a project I was working on. My friend says: "You can't
blame her. You know she's not responsible for her behavior. She
couldn't help it. Remember how she was abused and abandoned as a child,
developed destructive patterns of conduct early on, and now is just
doing what she thinks the voice of God is telling her to do."
We hear this sort of thing all the time--perhaps most often from
defense attorneys in court. Some piece of obnoxious behavior is caused
by prior circumstances, which themselves are caused by prior
circumstances, and so on back. These causal chains eventually reach
back to situations beyond the evil-doer's control. If one's behavior is
controlled by conditions which are not under his control, then he is
not responsible for that behavior.
___________________
Let's take a look at the meanings of the key terms here, and then think
about how they are related. Perhaps the easiest term to understand is
"moral responsibility". A person is morally responsible if he "is
answerable for his actions and so is a proper subject for praise or
blame" [Antony Flew, A Dictionary of Philosophy
(2nd ed.), pg. 306]. If a person is not responsible for her behavior,
then it is inappropriate to praise or blame her for doing or not doing
something. In order for someone to be responsible for an act in this
sense, most people would say that the act has to be within the person's
power…that it has to be under the person's own control.
The term "freedom" is a bit more difficult to grasp, perhaps because it
means different things to different people. One major contemporary
philosopher [Flew, p. 125--Flew is a CSICOP
Fellow, and Professor of Philosophy at the University of Reading in
England] gives us a starting point: people are free if they are able to
choose and act according to the dictates of their own will. This
closely parallels David Hume's
formula: By freedom we can only mean "a power of acting or not acting,
according to the determinations of the will; that is, if we choose to
remain at rest, we may; if we choose to move, we also may. Now this
hypothetical liberty [freedom] is universally allowed to belong to
everyone, who is not a prisoner and in chains." Perhaps this expresses
the same point: A person's action is free just in case he would have
acted differently if he had chosen differently. (Let me briefly note
why I said that the notion of "free will" makes little or no sense to
me. Given what I've just said about freedom, to say that people have
freedom of the will is to say that they are able to will according to
the dictates of their own will; free will is a power of willing or not
willing, according to the determinations of the will; a person's will
is free just in case he would have willed differently if he had willed
differently. Not very helpful.)
It seems fair to say that, according to Flew and Hume at any rate, the
absence of freedom is constraint or compulsion. In this sense, people
are often and obviously free, or have freedom. Putting these last two
notions together--freedom and moral responsibility--I'd be willing to
say that a person who is not free--a person who is constrained from
acting or compelled to act--cannot properly be held morally
responsible, whereas a person who is free in the above sense is
properly subject to praise or blame.
Consider two parallel stories in which I might be said to kill George Bush. In the first story, I'm looking down on the street from my sixth story window. I see Dubya
walking down the sidewalk in front of my house. I climb onto the ledge
outside my window with murderous glee. I perform a few rough
calculations and I jump, landing square on top of him, killing him
instantly. I think that most people would say that I killed him, that I
was responsible for his death.
Now change the story a little. While looking down, I see him coming. I
think to myself how easy it would be to squash him. I like the idea.
But before I can do anything about it, Joe Lieberman and Al Sharpton
grab me from behind and throw me out the window right onto the
President, with the same result as before. In either case, the
headlines read: "Logician crushes President." But in the first case,
I'm a free agent--or so it would seem--and in the second I'm a missile,
with no more responsibility for the President's death than had I been a
boulder or a piano. In the second case, the piano or I am not
responsible for his death because we would be moving wholly under the
influence of laws of nature--most obviously, gravitation--over which we
have no control. As I hurtle downward, it makes no difference to the
outcome whether or not I "choose" to land on the President. I probably
would not even be said to act at all when I plummet to the ground. It's
the difference between "I jumped" and "I fell".
Reference to laws of nature brings us to the third and perhaps most
difficult and controversial of our topic notions: determinism. Briefly,
determinism is the thesis that all events (including human actions and
decisions) are caused by prior circumstances. Before we can get a good
grasp on the notion of determinism, we clearly have to understand the
notion of causation--and more specifically, the notion that one event C
caused another event E, its effect.
I'd be reluctant to say that a cause makes its effect happen, because
the idea of one event making another happen is exactly as obscure as
the notion of one event causing another to happen. Similarly for a
cause bringing about its effect. These seem to me to be synonyms of
causing, and hence of no help in advancing our understanding.
Do we want to say that a cause necessitates its effect, obliges its
occurrence so that that outcome is inevitable and any other outcome
impossible? I think that's too strong. Necessity (and impossibility)
attach to sentences. I understand clearly what it means for it to be
necessary or impossible that some sentence is true. Some sentences are
logically true--true solely by virtue of the meanings of its
constituent terms--and others logically false--false solely by virtue
of the meanings of its constituent terms--though most sentences are
neither. But a sentence of the sort C caused E is not necessarily true
in the sense of being logically true. Nor is a sentence of the sort 'It
is necessarily true that effect E occurred, given that C occurred'
likely to be true. Analysis of causation in terms of necessitation
seems to me doomed to failure. Necessitation in any properly
intelligible sense is stronger than causation, and hence too strong to
provide us with an adequate analysis of causation.
We don't want causation to be some invisible and otherwise mysterious
relationship binding certain events together. We want it to be the sort
of "thing" for whose presence or occurrence we can get evidence in
experience. We want some specifiable observations, were they to occur,
to count in favor of a claim of the sort C caused E, and other
specifiable observations, were they to occur, to count against it. Let
me propose a candidate for an analysis (in the spirit of Hume) of the
claim that an event C caused an event E. This analysis is oversimple,
but ought to convey the point intended.
No doubt for evolutionary reasons, we tend to recognize similarities
between things and patterns in the occurrence of events. We note the
occurrence of an event--call it a(1). Then some time later we note the
occurrence of another event--call it a(2). And later, yet another event
a(3). In time we come to notice some similarities between these events,
by virtue of which we call them all A-ish events. We've begun to
generalize, giving a name to events of a certain kind. We also notice
the occurrence of another kind of event, call it b(1). Then some time
later we note the occurrence of another event--call it b(2). And later,
yet another event b(3). In time we come to notice some similarities
between these events, by virtue of which we call them all B-ish events.
More time passes, and we notice that A-ish events are always followed,
nearby and soon after, by B-ish events. It's easy to suppose that the
next thing that happens is that whenever we experience an A-ish event,
we come to anticipate the occurrence, nearby and soon after, of a B-ish
event. (That's called conditioning.)
If that expectation is sometimes unfulfilled, we refine our notions of
A-ishness and B-ishness to exclude the exceptions to the developing
rule: A-ish events are always followed, nearby and soon after, by B-ish
events. Regarding that rule as an hypothesis, we can often test it in
experience, and further refine it until it proves exceptionless. We now
have a law of nature--a statement of a regularity in the observable
world. Indeed, being a regularity in the ways events of one kind are
followed by events of another kind, we call it a causal law. And now
when some A-ish event a(n) is followed by a B-ish event b(n)
, we say that the former event caused the latter. We deploy the law A's
cause B's both to explain the occurrence of past B's and to predict the
occurrence of future B's.
If something like that is what causation means, then someone a bit of
whose behavior is caused (or all of whose behavior is caused) is
certainly not deprived of freedom of choice. To say that the behavior
is caused is just to say that someone who knew enough general facts
about the world (laws of nature) and enough specific facts about the
world (prior conditions) could infallibly have predicted that behavior.
Causation in this sense is not compulsion: it's in-principle
predictability ("in-principle" because no one actually knows, nor ever
will know, enough general and specific facts about the world to make
such predictions in all circumstances).
Determinism as the claim that all events are caused
amounts, then, to nothing more than the claim that each event is
related to some part of its preceding circumstances as the consequent
to the antecedent of a true law about the way events of certain kinds
invariably succeed each other. I don't know whether determinism is true
or not. I suppose that the expanding success of science gives us some
reason to think that it may well be true, at least at the macroscopic
level at which we live and move and have our being. But for our
purposes it doesn't matter. Certainly determinism is compatible with
freedom in the senses understood here.
The key question is whether or not determinism is compatible with moral
responsibility. Surely there is nothing about in-principle
predictability that is incompatible with the appropriateness of
judgments of moral responsibility, if (as we've seen) in-principle
predictability is compatible with freedom. Certainly a person's choices
and actions can be within that person's power, under that person's own
control, and still be entirely predictable by someone who knew all the
relevant facts both general and particular.
__________________
But Hume makes a further, and perhaps startling or controversial,
point: determinism is absolutely essential to the justification of
moral judgments; determinism is presupposed by those judgments. He
gives two arguments for this startling claim.
(1) Moral judgments--praise and other rewards for desirable behavior;
blame and punishment for undesirable behavior--assume that praise and
blame, rewards and punishment, "have a regular and uniform influence on
the mind, and both produce good and prevent the evil actions." That is
to say, praise and blame, rewards and punishment, are causes which make
some behavior more likely to occur and other behavior less likely to
occur. If rewards and punishment weren't causes of people's choices and
conduct, it would be at best pointless and at worst cruel to employ
them.
(2) We said early on that a person is a proper subject for praise or
blame and hence is morally responsible if he is answerable for his
acts. In order for someone to be responsible for an act, the act has to
be within the person's power…it has to be under the person's own
control. What can this mean but that the act is caused by the person's
character, nature, prejudices, opinions, preferences, decisions, and so
on--her make-up. Think of the alternative. If one's actions were not
caused by her make-up, then behavior would pop out of her at random,
and there could be no sense in which it was her behavior. All behavior would be like the outbursts of Tourette's Syndrome,
for which we don't blame people, since those outbursts are not under
the person's own control. (Of course I'm not saying that such outbursts
are uncaused. I'm only saying that they're not caused by the person
with the disorder.)
___________________
So where does that leave us?
(1) The absence of freedom--compulsion or constraint--is incompatible
with moral responsibility: a person whose behavior is compelled or
constrained is neither to be praised nor blamed for that behavior.
(2) Determinism (the doctrine that all events, including human
behavior, are caused by prior circumstances) may be true or false. But
even if it's true, it's wholly compatible with human freedom and
judgments of moral responsibility (praise and blame).
(3) Beyond that, it has been argued that if determinism is false, and
particularly if human behavior is not caused, then judgments of moral
responsibility (praise and blame) would be thoroughly inappropriate.
APPENDIX ON
FREEDOM AND FOREKNOWLEDGE
Here I intend not to address further the compatibility of freedom and
determinism, but to address a closely related issue: the compatibility
of freedom and foreknowledge. People are free
if they are able to choose and act according to the dictates of their
own will. This closely parallels David Hume's formula: By freedom we
can only mean "a power of acting or not acting, according to the
determinations of the will; that is, if we choose to remain at rest, we
may; if we choose to move, we also may." Perhaps this expresses the
same point: A person's action is free just in case he would have acted
differently if he had chosen differently.
When I say that some person's action is foreknown, I mean
that prior to that action, somebody knows (not just has good reason
strongly to believe, but knows) that it will occur. By virtue of what
knowledge is, any statement that is known is true; you can't know
something that is false. So, to use Aristotle's
example, if I now know that there will be a sea-battle tomorrow, then
it is now true that there will be a sea-battle tomorrow.
The third component of this problem is God's omniscience--he knows everything that is the case. Whoever believes that God is omniscient believes that He knows everything that has happened, is happening, and ever will happen.
It certainly seems, at first blush, that these theses can't all be
true. If the omniscient God has always known that I would write this
paper, then it has been true from all eternity that I would do so. It
certainly looks like I have no choice--at least no effective choice--in
the matter. It makes no difference at all what I choose to do. I will
write the paper no matter what. It looks like my choices are
irrelevant, and that the writing of the paper is inevitable. The same
applies to everything that everybody does. Given that, it certainly
looks like we are never free.
This conclusion has led some people to deny that God is omniscient. On the other hand, it has led some Christians, focusing on Scriptural talk about election and predestination,
to deny that we are free. I mean to show that such dodges are quite
unnecessary, since God's foreknowledge of our actions is compatible
with our freedom.
Possible worlds
Here we take some cues from G. W. Leibniz
(1646-1716), German philosopher, mathematician, and logician. A
possible world is a way things might unfold or develop in reality. We
live in one possible world. But there are infinitely (denumerably)
many other possible worlds--other collections of things that would
develop in other ways. If you need to picture them, suppose that they
are all held in God's imagination, none of them yet "actualized". Some
are similar to ours and others are quite different. Some have more
things in them than ours, and some less. Some are governed by simpler
or fewer principles or "laws of nature", and some by more complex or
more numerous laws. If you want to picture a possible world other than
our own, just imagine this world different in some major or minor way.
And some possible worlds are so radically different from our own that
we can't even imagine them. All that's required is that they be
possible worlds--that the concepts thereof contain no
self-contradictions. None is yet a real world.
In some but not all of the possible worlds, laws of nature
obtain--principles of regularity in the ways that events succeed one
another. In some but not all of these, every event follows prior
circumstances in accordance with laws of nature. In some but not all of
these possible worlds, each event is necessitated by what went before,
so that under those particular antecedent conditions, nothing else
could have happened. Now disregard these last "necessitated" worlds.
There are infinitely many different possible worlds remaining--in God's
imagination, as it were. In some of them, some human conduct is free,
though perhaps not in all of them.
God scans the infinitely many details of each of these infinitely many
possible worlds. For reasons that are irrelevant here, He decides to
"actualize" or create or bring into reality one of these possible
worlds. As it turns out, He chooses one of the possibilities in which
some human conduct is free. And of course He knows everything that will
ever happen in that world. After all, He understood every detail before
He decided to bring it into existence.
Now think about it. His having chosen to actualize a possible world in
which action is free doesn't change that action from free to
necessitated or compelled. Why? Because the actualized world is exactly
the same world the concept of which was in God's mind prior to
creation--a world of free creatures. If people's behavior in this real
world were necessitated, it wouldn't be the possible world that God
selected; it would be a different possible world.
I'm not arguing that this is the way the world came to be. Nor am I
arguing that some human action is free. I have only established that
human freedom is fully compatible with God's foreknowledge of all our
(mis)behavior.
Copyright © 2004, William A. Wisdom