Linguistic Philosophy. The linguisticism now pre-
dominant in American and British academic philosophy
offers no contribution, perhaps, to the development of
the free will idea; but it offers a fresh approach to
the free will-determinism issue, seeing it as a question
of adjusting to one another two modes of speech,
through a careful study of the natural uses proper to
each. We use the language of sheer personal action—
with or without implications of alternative choice—
and we use the language of event, process, etc. Actions
we talk of as what we simply or freely do; events we
talk of as happening. To actions we assign intentions,
to events we assign causes. Of an event we ask, “What
led to it?” Of a man's action, “What is he up to?”
The problem of free will (excluding its theological
aspects) will be the problem of relating these two
ranges of speech to one another.
One point must first be made clear. Language which
describes or mentions choiceful action is secondary to
language in which we do our choiceful thinking or
make our choice itself, as when you say to someone
placing alternatives before you “I opt for” A or B
(J. L. Austin, How to Do Things with Words, Oxford,
1962). Having made this point, we may proceed to
show that the correlation of event-style and personal
action-style speech, so far from being an oddity, is
common form. In the case of your talking choice, or
spoken option, you choose in view of an opinion of
facts or events, which, if you do not explicitly state
it, you still take for granted. Indeed, your making of
your choice is a virtual affirmation of the facts in view
of which you make it. Voluntary decision to act upon
(supposed) facts will often be taken as a more serious
assertion of those facts than a direct statement of them.
If you did make the mere statement, it would be in
event-style form.
Or take the case in which you are simply describing
events or their interrelations. It remains true that you
do the describing, and you go about it as you choose,
talking backwards, forwards, or across the event-
process you describe, picking one word and rejecting
another. And if I wish to understand your speech, I
do not look for the causes of the vocal inflections, I
look to your expressive intention. Whereas, to under-
stand what you are talking about, I attend to the causal
sequences you are describing. In fact, I am bound to
do both. To understand you as speaking I must under-
stand the facts you state and to understand the facts
you state I must understand you as speaking; so in-
separably are the two modes or logics connected.
But if the matter is as straightforward as this, how
can we ever come to think of denying either mode
of speech its rights? Why should anyone dream of
reducing voluntary-action statements to the happen-
ing-by-cause category? The answer is this. So long as
we are talking our way into or through choiceful acts
we can use no category but the category proper to
such talk—I cannot talk to myself the choice I am
making, as being an event which occurs. But often we
speak of choiceful actions from outside, as having been
done or as likely to be done, and then it is possible
for us to switch categories, and to talk of them as
events—or, more accurately put, to talk, instead, about
the events in which they take effect. And events as
such are subject, we assume, to the category of causal-
ity—to exposition in terms of uniform sequence—and
are in principle predictable from a knowledge of their
antecedents. If, then, the event in which a choiceful
act takes form is predictable, and causally determinate,
how can that act itself be free? Such appears to be
the puzzle.
The determinist case is that the causal regularity of
choice-produced events should be accepted. There are
two stories—one descriptive from “within” of the way
in which the choice is made, another descriptive from
“outside” of the event's position in the event-sequence.
Each story is veridical in its own sphere, and there
is no difficulty in letting them run parallel.
The freewill rejoinder is that a solution in that sense
rests on a falsification. Two stories covering the same
ground and expressed in different logical idioms may
rightly be tolerated when they are both objectively
descriptive stories. I may tell a story of past voluntary
activity in cold detachment, and feel no great mental
discomfort in doubling it with a causal account of the
same behavior. But that is because in imagination I
degrade a personal story to the level of a story about
a process which unrolled as it did unroll. And that is
to depersonalize the story. It is only personal insofar
as I identify myself in some measure with the charac-
ters or agents in it and express them as personally
active from “within.” And then the acquiescence in
a parallel cause-and-effect story becomes impossible.
All the rejoinder achieves is to set aside the deter-
minist's soothing compromise. Three possible positions
remain. We may say (1) So much the worse for the
ultimate validity of the free-action language—a deter-
minist conclusion; (2) So much the worse for that of
the caused-event language—a libertarian conclusion;
(3) So much the worse for both, our language in either
case having a purely pragmatic value, in serving our
purposes—an agnostic conclusion.
The determinist will speak slightingly of the “sub-
jective character” of personal experience and its ex-
pressions, the libertarian of the “abstract and diagram-
matic character” of causality-constructions; while the
agnostic will cite the agelong inconclusiveness of the
debate between the two other parties, and the inade-
quacy of language as such to the nature of things.
The defender of free will ruins his case if he over-
plays his hand. He must not deny the validity of
causal-regularity interpretations so far as they go; but
he will maintain that we have no reason to suppose,
and much reason to disbelieve, that the grid of natural
uniformity fits so tightly upon living processes as to
deny scope to free personal action. On the other side
of his case, he must avoid exaggerating human liberty.
The individual is constantly subject to pressures, visible
and invisible, which he often has no motive and some-
times no ability to resist; and the free options he does
exercise are mostly within a range of choice narrowly
circumscribed by conditions outside his control. So
human conduct may often be broadly predictable. On
the other hand, the libertarian is not going to admit
that all the predictability in a man's conduct is de-
pendent upon the operation of determining causes
which restrict his freedom of choice. For in taking a
decision, a man will follow his usual policy in such
matters unless he now sees reason to revise it. If we,
his friends, have formulated his policy to ourselves, we
may think of the policy-rule, taken in conjunction with
the circumstances invoking its application, as causing
his action. But voluntary consistency is not subjection
to any determining causality. His policy only guides
the man because he goes on choosing to maintain it.
It is a hard case, if voluntary freedom is only to be
evinced by wild and continuous caprice.
Most difficult of solution along linguistic lines is the
theological problem of free will in face of a sovereign
divine will, insofar as religious conviction puts forward
statements about divine initiative in the origination of
human free acts which are in formal conflict with
statements about the human agent's own initiative.
Appeal may be made to the believer's practical under-
standing of what it is to exercise his will in prolongation
(as it were) of God's. But no formal solution can be
attempted without a prior examination of the special
sense and status of statements about the Divine Subject
(traditionally known as the topic of Analogy).
