North American Philosophical Publications

Recent Work on the Free-Will Problem

Author(s): Harald Ofstad
Reviewed work(s):
Source: American Philosophical Quarterly, Vol. 4, No. 3 (Jul., 1967), pp. 179-207
Published by: University of Illinois Press on behalf of the North American Philosophical Publications
Stable URL: http://www.jstor.org/stable/20009243 .
Accessed: 17/12/2012 10:29

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

University of Illinois Press and North American Philosophical Publications are collaborating with JSTOR to
digitize, preserve and extend access to American Philosophical Quarterly.

http://www.jstor.org

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 American Philosophical Quarterly
Volume 4, Number 3, July 1967

I. RECENT WORK ON THE FREE-WILL

PROBLEM
HARALD OFSTAD

few of the so-called indeterminists and libertarians

I. Some Major Trends
accept that freedom in an indeterministic sense is
the problems formulated and the not a sufficient,
only a necessary condition of free?
ALTHOUGH answers given by contributors to the free-will dom in the sense required by moral
responsibility.
discussion during the last decade1 vary immensely, It is also necessary that the agent was able to do
there are nevertheless certain noticeable trends. something, e.g., to make what has been called a
For instance, very few of the authors take it as certain "effort of will."
a priori or certain that the so-called The period includes some able defenders of so
empirically
principle of causality can be applied to human called mild (soft) as well as hard determinism, but
decisions and actions.2 It ismostly considered as a they are, on the whole, less concerned about
working hypothesis.3 Moreover, due to the obvious proving that our decisions and actions are caused
difficulties of discussing the question of causation than trying to show that moral is?
responsibility
as an
ontological issue, the authors tend to con? or is not?compatible with this assumption. More?
centrate on the
question whether or not our over, also the determinists have become interested
decisions and actions can be predicted or explained. in the analysis of such expressions as "tried," "was
In this connection the Bergsonian point, that a able to," "had the ability to," "could have acted
chooser cannot predict his future decision without (decided) otherwise," "had it in one's power to act
making it,4 has been taken up again by several (decide) otherwise," rather than following their
authors. Some of them also stress the methodo? more traditional line only, defining "freedom" as
logical difference between things and human absence of compulsion or constraint.
Thereby they
beings, denying the program of the unity of science have come closer to the libertarian position in so
movement, and
advocating
some variation of the far as this stresses "the ability to make efforts of
old distinction between
verstehende and erkl?rende will." This new-won interest, both of some deter?
explanations,5 emphasizing the methodological minists and indeterminists in the analysis of the
dissimilarity between explanations of actions (in sense in which we have it in our power to decide
terms of reasons), and explanations of events (in and act otherwise than we do, and the extent to
terms of causes). Such concepts
as reason, choice, which we have such power, is one of the most pro?
decision, intention, and action have been sub? mising features of the recent discussion. My main
jected to numerous analyses, some of which in? reason for this evaluation is that this line of think?
directly may be of value to the free-will discussion. ing seems to me to focus on that sense of "free"
At the same time that we find this increased which is most relevant from the point of view of
scepticism toward certain interpretations of the so moral responsibility. So even disregarding the
called deterministic doctrine, we find that quite a semantic unclarity of such terms as "determinism"

1
This, of course, is quite an arbitrary delimitation of a period of debate.
2
Blanshard is one of the few exceptions. See his "The Case for Determinism" in Hook will
(1958). (My references include
only the name of the author, and, if convenient, the year of the publication of his work. The exact reference can be found
by consulting the bibliography. If one author has several relevant from the same year,
publications they will be marked,
e.g., 1960.1, 1960.2, etc.). I see no reason to define "the principle of causality" in an article of this kind. Different definitions
are discussed in the literature referred to.
3
See, e.g., Nagel (1961), pp. 316 ff.
4
Bergson, Time and Free Will (London, 1910), pp. 184 fr.
6
Gp. W. Dilthey, "Ideen ?ber eine beschreibende und zergliedernde Psychologie," published 1894, reprinted in Gesammelte
Schriften (Leipzig und Berlin, 1924), Band V, pp. 139-240, and H. Rickert,
Kulturwissenschaft und Naturwissenschaft (T?bingen,

179

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 l80 AMERICAN PHILOSOPHICAL QUARTERLY

and "indeterminism," they have become less signi? Mr. Smith shot his aunt Olga to death. He admits
ficant also because they do not refer us to the whatever can be shown by a film of the happenings,
crucial issues.6 but he maintains that he is not responsible because
One of the great weaknesses of the recent dis? his will was not free. How would the recent con?
cussion is the insufficient clarification of the moral tributors to the free-will debate deal with this
context in relation to which the authors maintain contention ?
that we are (or are not) free. Though quite a few Most of them would consider it impossible to
are aware of the necessity of clarifying such expres? take any standpoint to it until it has been clarified
sions as "moral responsibility" in connection with what Smith meant by saying "my will was not
the free-will dispute, the analysis is still at a pre? free." By the term "will" he may refer, among
liminary level, and for some strange
reason the other things, to his so-called volitional capacity,
overwhelming majority of writers7 consider moral his choice, his decision or action. Let us suppose
responsibility equivalent with moral blameworthi that he meant that his action was not free. But
ness?as if every philosophy of responsibility what did he mean by "not free"? Perhaps he
commits us to some version of a meant that his action was in some
"you-naughty compelled way,
boy"-philosophy. Obviously, we may hold a person or that it was caused, or caused by certain special
morally responsible without holding him blame? factors, e.g., neuro-physiological
events or the so

worthy,
we may, e.g., consider him morally guilty called irrational part of himself, or perhaps he
without considering him blameworthy. Moreover, meant that he did not have it in his power to act
in
spite of the awareness of the role of excuses, very otherwise than he did.
few have seen the moral relativity of the free-will Let us suppose that Smith meant that his action
issue. Some philosophers even seem to believe that was caused (T). We can then raise the questions :
they can solve the so-called free-will problem by (i) Is T sufficiently precise, and (ii) Is T true ? But,
finding out how we use such sentences as "He as indicated above, a more fundamental question
could have decided (acted) otherwise." But even is: (Hi) Is the lack of freedom in this sense of "free?
those who understand that such investigations
are dom" sufficient to exclude moral responsibility?
not decisive and that we must find out whether or Or, formulated in more general terms: In what
not we are free in the sense required by moral sense of "freedom" is absence of freedom a sufficient

responsibility are insufficiently aware of the fact reason for holding that Smith is not morally re?
that this sense may vary with the nature of the
sponsible? The answer will depend upon our
moral system within which the concept of respon? concept of responsibility and the way in which we
sibility is defined. According to the moral system justify statements ascribing responsibility to agents.
Ex,
a
person is perhaps not
responsible
unless he It is impossible in this article to deal with these
decided freely in sense x, whereas, according
to ?"25
problems in any thorough way, but certain basic
he is not responsible unless he decided freely in distinctions must be made.
sense y. Let us suppose that we are free in sense x When someone says that Smith is responsible
but not in sensey. Do we then have a free will? for the death of Olga, he may mean only (i) that
That depends upon whether the system E1 is some of Smith's muscular movements were a causal

"more correct than" the system E2. Hence, there factor in that series of events which led to her death.
is an important sense in which a solution of the The statement that Smith is responsible in this
free-will issue presupposes a clarification and sense is empirically true or false, and there is no
solution of certain basic problems within ethics. plausible sense of "freedom" in which freedom is a
condition of responsibility in this sense. Next, one

II. Freedom and Moral Responsibility may mean (ii) that her death was due to something
Smith did, where "doing something" at least
To simplify the comparisons of different implies that Smith was sufficiently conscious to
positions, let us?whenever it is convenient? understand that his behavior was dangerous.8 In
concentrate on an
example of an action which we relation to this sense it is appropriate to ask
shall assume to be morally wrong: on May 1, 1965, whether freedom in any sense is a condition of
6
Cp., e.g., Pears (1963), ch. 4.
7 is one of the exceptions. See his "Determinism and the Theory of Agency" in Hook
Taylor (1958), p. 226.
8 Somewhat are made by Feinberg
similar distinctions (1965). See also, e.g., Brandt (1958), Friedrich (i960), and Hospers (1961).
A useful discussion of the concept of criminal responsibility can be found in Wootton (1959).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 RECENT WORK ON THE FREE-WILL PROBLEM l8l

responsibility. Furthermore, one may mean (Hi) do they defend that this condition is?or is not?
that Smith ismorally guilty. The concept of moral fulfilled?
guilt is rather neglected in recent literature, and in Connected with the debate concerning the cor?
so far as it is taken up at all, the authors tend to rect
analysis of "free," there is another
concerning
conceive of it as identical with blame worthiness. the object of freedom.9 Some maintain that the
But this is not for a person who is is whether or not our actions
acceptable, important question
against moral blame as a matter of principle?and can be free, others that the relevant issue iswhether
therefore considers nobody blameworthy?may or not the self, the acting agent, or our decisions
nevertheless consider someone guilty. If he finds are free.
that the utterance of the sentence "You are guilty" In view of the concepts used by the
majority of
will be experienced as moral blame, he may stop the participants in the recent debate, it seems
saying such things, but still think them. preferable to choose "decision" and "action" as
To say that Smith ismorally guilty of the death our key terms. But if we take "freedom" in the
of Olga is to give a moral evaluation of Smith as sense of power to decide or act otherwise,
talking
the producer of this result in view of all the factors about the freedom of decisions or actions is really
in favor and disfavor of him. And, analogously to to talk about the agent9s power to decide or act.
criminal guilt, moral guilt can vary in degrees.
Hence, saying that Smith ismorally guilty of killing III. Types of Positions
Olga can be expressed thus: "In killing Olga you
were morally bad to such and such a degree." To Before dealing with some representative contri?
analyze the sense, if any, in which freedom is a butors, it may be useful to present a short survey
condition of moral guilt, is a crucial task within of the major types of positions.
the free-will debate. i. Indeterminism. Let us
distinguish two variations :
To say that Smith is morally responsible may i.i. Ontological indeterminism: decisions are not
also mean (iv) that he ismorally blameworthy. By caused, or, at least, some of their characteristics are
one often seems to mean that the not caused.
"blameworthy"
agent?in my terminology?is morally guilty, and 1.2. Epistemic indeterminism: it is to
impossible
that we consequently have a reason for blaming predict (and explain) decisions, or, at least, some
him. I have already mentioned the relation be? of their characteristics cannot be predicted (and
tween freedom and guilt, and the only new explained), at least not by the chooser when he is
question raised by (iv) is therefore concerned with deliberating.
the relation between freedom and blame. This 2. Two-domainism : the causal
principle does not
problem, however, can be more adequately dealt apply to decisions or actions, they are neither caused
with by turning to our fifth interpretation of saying nor uncaused, but it applies to
bodily movements
that Smith is responsible : someone not neces? are either caused or uncaused.
(but ?they
sarily you or /) ought to express moral disapproval 3. Libertarianism. Let us distinguish two types:
of him (v). And the final question is this : In what 3.1. Moral libertarianism: our decisions between
sense, if any, of the term "freedom" is freedom a following our interests or doing what we consider
condition of moral blame? our duty are made by a self which is not com?
My exposition must be limited to some of these pletely causally determined, and we are always
senses of "responsible," and by concentrating
on able (unless we are mentally sick) to put forth the
sense (Hi)?moral guilt, and (v)?moral blame, effort required in order to decide in favor of the
we seem to have the best chance of getting at the latter.
most important issues. Our problems can now be 3.2. Neutral libertarianism: our decisions between
thus: How do the contributors to the different courses of actions are made
formulated by a self
recent free-will debate deal with the question which is not completely causally determined, and
whether freedom in any sense is a condition of we are always able (unless we are
mentally sick) to
moral guilt and/or moral blame? What do they put forth the effort required in order to decide in
mean by "moral guilt," "moral blame," and by favor of anyone of them.
"freedom" ?How do they defend that freedom is? 4. Fatalism : I distinguish between :
or is not?a condition of guilt or blame ?And how 4.1. Logical fatalism : it is logically necessary that

9
See, e.g., Thalberg (1964).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 l82 AMERICAN PHILOSOPHICAL QUARTERLY

the future must be what it is going to be, and sibility requires that the wrongdoer would have
logically impossible that it can be anything other decided or acted otherwise if this or that condition
than what it is going to be. had been fulfilled, e.g., if he had chosen otherwise.
4.2. Empirical fatalism : our deliberations do not According to this view, we should, ideally speak?
influence our decisions and actions.10 ing, know
something about the agent's past.
5. Hard determinism: moral responsibility pre? 7.3. Freedom as categorical power : according to this
supposes that we, in a categorical sense of "could," view, moral guilt, in order to be maximal, requires
could have decided or acted otherwise, but since that the agent had it in his power to decide or act
our decisions and actions are caused, it follows otherwise?in the fullest sense in which we can
that we are not free in this sense. ever have anything in our power. And although
Two variations may be distinguished, depending our power to decide and act may be limited by
upon the moral consequences drawn. According to many different factors, it is never the principle of
the first: causality which annihilates or delimits it.
5.1. Hard determinism of the anti-moral-blame type:
the above view implies that we are never justified
IV. Ontological or Epistemic Indeterminism
in morally blaming anyone, but we may be justified
in making use of non-moral blame. The main arguments recent authors
presented by
According to the second: accepting ontological or epistemic indeterminism
5.2. Hard determinism of the anti-merit-blame type: are not based on empirical research or
introspec?
the above view implies that we are never justified tive reports about what we find ourselves believing
in maintaining that anyone deserves or merits blame, when we choose, but on some kind of philosophical
but blame, moral as well as non-moral, may be
analysis of such concepts as intention, decision, and
justified, e.g., in terms of utility. voluntary action. By the use of such analyses, they
Some authors have defended the position that have tried to discover a logical proof for indeter?
the moral consequences mentioned in 5.1 and 5.2 minacy or unpredictability.
follow, not from the principle of causality, but from
the principle that we are products of hereditary
C. Ginet
and environmental factors. We may refer to this

position as: In their article "Decision, Intention and Cer?

6. Hard evolutionism. tainty" (1958), Hampshire and Hart tried to show,
common to the other that an cannot
7. Mild determinism: positions among things, agent both
referred to by this label is the contention that moral know what he is going to do and deliberate about
does not require that we are free in what to do. A similar line of thinking was later
responsibility
any sense which is incompatible with the assump? presented by Ginet (1962). He wanted to prove
tion that our decisions and actions are caused. that
Some of to this view hold
the adherents that (i) It is conceptually impossible that a decision
presupposes that our decisions are should be caused.
responsibility
caused, whereas others deny that it does. Ginet maintains that (i) follows from the follow?
Three variations may be distinguished, depend? ing two propositions:
ing upon the sense of "freedom" considered crucial. (ii) If it were conceptually possible for a decision
7.1. Freedom as modifiability : according to this to be caused, then it would be conceptually pos?
position, the only requirement is that the agent is sible for a person to know his own decision before
free in the sense that the use of blame will modify he had made it, and
his behavior in the future. Knowledge about his (Hi) It is conceptually impossible for a person
past, e.g., about whether or not he could have to know his own decision before he has made it.
decided otherwise, is of interest only in so far as it Since (in) seems to be his most important thesis,
helps us to judge his present modifiability. I shall here assume, for the sake of argument, that
7.2. Freedom as hypothetical behavior: moral respon (ii) is true (but see below).
10 I the dispute over fatalism and free will as a "how-do-you-handle-this-argument?"-game rather than as serious
regard
philosophy. (Indirectly it may have a certain philosophical value in pressing forward more precise analyses of "having some?
thing in one's power.") So, although it has had a renaissance during the last decade, I shall pay no attention to it in this work.
A bibliography including most of the recent articles can be found in Steven Cahn's "Fatalistic Arguments," The Journal of
Philosophy, vol. 61 (1964), p. 295.

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 RECENT WORK ON THE FREE-WILL PROBLEM 183

According to Ginet, the possibility of knowing S. Hampshire

one's own decision in advance involves an ab?
the statement "I already know Although Hampshire (1959), just as Ginet,
surdity, because stresses that a decision involves knowledge of the
that I shall later decide to do . . ., but I have not
future, his argument against the possibility of self
yet decided what I shall do . . . that is, I do not seem to on a
prediction does not be based cognitive
yet know what I shall do" (p. 51), makes two
inconsistent claims. theory of decisions, but is valid, if valid at all, even
logically if a decision is conceived as a normative
The main weakness of this argument is Ginet's standpoint
of a certain kind.
conception of a decision. He seems to conceive of a
His main thesis, as far as the present problem is
decision to do something as some kind of knowledge
is that we cannot our own
of something. Once concerned, predict
this position is taken, it is
decisions without making them. But he admits that
rather easy to prove (Hi), because a person cannot
"involuntary" decisions, wherein he probably in?
claim both that he has and does not have the
cludes decisions based on habits, may be predicted
knowledge in question. But we must distinguish
without making them. So what his thesis really
between deciding and knowing. To decide to kill
amounts to is that we cannot predict those of our
Olga on May 1 is not to know anything about the
decisions which we have it in our power to make
future, not even that one will try to kill her. To or not to make?without them.
make a decision is to take a normative standpoint
actually making
And perhaps we have no decisions in our power?
of a certain kind; it is to commit oneself to a certain
The argument which we are about to consider
alternative. Roxbee-Cox, one of the critics of
does not prove, but presupposes, that we have
Ginet's position, says aptly "It is in fact rather some decisions in our
to emphasize power.
misleading that making up one's He formulates his main argument in this way:
mind is a way of gaining knowledge, for this implies
"Either I take into consideration the reasons that
that discovering on empirical grounds that a cer?
will influence me, in which case I am already
tain decision will be made and actually making
that decision are competing methods of gaining the engaged in forming a plan; or I somehow contrive
to ignore the factors that will influence me, in
same knowledge" (p. 90).u which case I cannot honestly profess any con?
We may, however, interpret Ginet's article in a fidence in my own prediction"
different way, viz., as making the point that it is (p. 130). His main
to deliberate about what point, in other words, seems to be that the detailed
psychologically impossible of our reasons for and against the dif?
to do when one is convinced about what one knowledge
ferent alternatives necessary in order to predict the
actually will do.12 This may be acceptable, though decision presupposes that the reasons
is a very strong term. But this psycho? forthcoming
"impossible" are compared and considered in such a way that
logical difficulty is not a function of the agent's
the decision actually is being made.
knowledge, but of his beliefs. And even knowledge Does his argument prove his thesis? I think we
is all right, if, as pointed out by Roxbee-Cox, the
can that in order to predict own deci?
what he knows when he deliberates. accept my
agent forgets
the difficulty has nothing to do with sion, I must study some "factors that will influence
Moreover,
our
me." But this neither implies that it is necessary to
whether decisions are caused or not.
take into consideration my reasons for and against
Suppose Ginet had succeeded in proving that it the different alternatives :perhaps it will suffice to
is "conceptually impossible that a decision should or psychological
be caused." This would not imply that Smith's study certain neuro-physiological
factors, nor that it is sufficient to do so: perhaps
decision was not caused, only that if it was a
also other factors will influence me.
decision in Ginet's sense, then it was not caused
(Ginet is quite aware of this point). Moreover,
he gives no reasons to show that freedom in
Knowing about the Prediction
his sense is relevant to the moral questions of
guilt and blame, and this, I think, was a wise Implicit in the line of reasoning so far considered
omission. is the view that the self-predicting agent is not
11The
discussion elicited by Ginet's article includes contributions by, among others, Canfield (1961-62), Swiggart (1962-63),
Lehrer (1963), Thalberg (1963-64), Berofsky (1964), Taylor (1964.1).
12 Some of
the psychological problems rooted in the distinction between agent and spectator are discussed by Hampshire
(1959), O'Shaughnessy (1963), and by several authors in The Monist, vol. 49 (1965).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 AMERICAN PHILOSOPHICAL QUARTERLY
184

confronted with an outcome which is independent relevant, or as it has been said by Bernard Mayo
of his own predictive activity in the sense in which : "no . . . record can . . . include
(in Mayo [1958.2])
the occurrence of an eclipse of the sun is in? a record of the act of recording" (p. 229).
dependent of the activity of the predicting To assert that our predictions may influence our
astronomer. This view?the view of interaction? future decisions does not imply that they always
is and can the basis for an will do so. I see no reason to accept the latter view.
acceptable, give argu?
ment against the possibility of self-prediction which The motivations favoring the predicted decision
is different from those above. The argument is may be so strong that they check whatever counter
this: P's prediction at time tx of his decision at influences the prediction might elicit. It is even
time ?2maY influence his actual decision at t2 and possible that whereas prediction no. 1 influences
the of it. Also the decision so that a prediction no. 2 is
thereby possibility predicting predicted
another person's prediction may influence him if needed, no. 2 does not modify the predicted result.
he gets to know about it, but this other person Whether a decision will be influenced by one's own
may have a second prediction in reserve, whereas final prediction is an empirical question,14 but if
the chooser necessarily will know his own last such influence takes place, the prediction will be
prediction.13
falsified.
A short presentation of this difficulty is given in
Oldenquist (1963-64). He wants to show that it is
Conclusion
impossible to know one's decision in advance. He
presents his argument as follows: The authors presented have focused their atten?
In order to know at that I at will make tion on the possibility of self-prediction. With this
(i) t1 t2
the decision D, I must, at tl9 know the conditions, as a starting-point, Ginet tried to prove that our
concept of a decision a decision is
say A, B,C, which are causally sufficient for D. implies that
In order to know that A, B, C are causally that we cannot know our
uncaused, Hampshire
sufficient for D, I must know that those factors decisions in advance, and Oldenquist that pre?
which might lead to not-D, will not do so. But my dictive knowledge is impossible because of the
own activity in studying the conditions and pre? effect of knowing about the prediction. I have
the outcome. been more or less critical toward their whole
dicting my decision may influence
Hence, I must at tl9 know that it is false that approach. It may be of value from the point of
C are view of obtaining a better understanding of some
(ii) A, B, plus my predictive activity, pl9
sufficient for for the same I of our or of the agent and his situation,
not-D, and, reason, concepts
must knowthat it is false that but it is wasted from the point of view of solving
(Hi) A, B, C plus px plus my knowledge that (ii) the question of free-will. This is easily seen if we
is false are sufficient for not-D. assume that they had succeeded in proving that
we into an infinite This are impossible, and that this dis?
Hence, get regress. point, self-predictions
also made other authors, seems to me covery was communicated to Mr. Smith's judge,
by many
correct, but I do not consider Oldenquist's con? who then replied to Smith's excuse by saying "I am
clusion well formulated. He says that his argument sorry Smith, but you are to hang, for it has been
has shown that "I cannot possibly know that all discovered that your will was free in the sense that
circumstances besides A, B, C are causally irrele? itwas impossible for you to predict in advance that
is
vant to my decision" (p. 56). This formulation you would decide to kill Olga." A reply which
does not take us to the heart of the matter, because either nonsensical or cruel or both.
it refers to a type of uncertainty that we encounter As mentioned above, few authors deal directly
even in experimental science. The crucial thing is with the question whether or not our decisions and
that as we are one actions are caused.15 Nevertheless, some words
self-predictors always step
behind the mastering of the factors that may be about this issue seem
appropriate.

13As of knowing about the prediction, see also Lyon C. J. F. Williams (1960
regards the implications (1959), MacKay (i960),
61),14Canfield (1961) & (1961-62), Beck (1965), and Perry (1965).
Cp. Stenner (1964).
15But of different of "determinism" and different
(1958) and Ofstad (1961). A discussion
see, e.g., Farrer interpretations
to this issue will be found in Hook Pts. I & II. See also Brandt Hospers (1961), and
arguments pertaining (1958), (1959),
in B?hm Scriven (1957),
Pears (1963), ch. 4. The use of causal explanations in science is discussed (1957), Gr?nbaum (1957),
Bunge (1959), Waismann (1959), Munn (i960), Nagel (1961), and Boring (1963).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 RECENT WORK ON THE FREE-WILL PROBLEM
185

So far neither psychological nor physiological (i)What is, according toMelden, an action "more
research have succeeded in establishing any uni? than" a bodily movement? (ii) Why is the causal
versal law of relevance to the present discussion. model inapplicable to actions? (Hi) Does his line
Moreover, it seems very likely that at least some of thinking have any significance for the question
characteristics of some decisions and actions will of responsibility and freedom ?
remain statistically rather than universally related Melden tries to answer (i) by analyzing the
to other factors in the universe. One reason for
example of a driver signalling at a cross-road. The
this : the interaction between the predictor and the driver moves his arm in a certain way, but

phenomena to be predicted has already been men? "because of our familiarity with the rules of the
tioned. Another reason is the limited inter road, we
recognize
. . . that he is signalling"
subjective accessibility of the required data. In (p. 191). The report that the driver is signalling is
some cases it may be the morally most important not based on different sense-impressions from those
characteristics which are only statistically related we have when observing his bodily movements.
to other factors, in other cases these characteristics No telescope is needed to see that the driver is
may perhaps be universally related to other factors. signalling and not only moving his arm. Melden
Let us suppose that we knew that those charac? says, quite explicitly, that although "moving the
teristics of Smith's decision which made itmorally arm" and "signalling" are different descriptions, a
wrong had a low degree of statistical relation to case of the former is, in appropriate circumstances,
other factors in the universe. Would this give us a the "very same thing" as a case of the latter. So
basis for an answer to his excuse? We could say what is the difference? His point seems to be that
"Smith, you are to hang, for your will was free in the observer seeing the movement of the arm as a
the sense that those characteristics of your decision case of signalling has been trained through certain
which made it morally wrong had a low degree of social rules. He meets the stimuli with an acquired
statistical relation to other factors in the universe." ability to understand what he observes, just as he
But the answer is ridiculous. And it is ridiculous meets certain geometrical figures on paper with an
because it is just as irrelevant as ifwe said "Smith, ability to perceive them as letters. Melden's idea,
you are to
hang, because your hair is green." in other words, seems to be that P's bodily move?
ment M in the situation S in society X at time t, is
an instance of the action A if and only if there is
V. Two-Domainism
within X, at t, a rule according to which move?
According to Kant, our decisions and actions are ments such as M, should, in S, be recognized as
causally determined qua appearances, but neither actions of type A.
caused nor uncaused qua things in themselves. The But how shall we understand the term "action"
logic of this attempt of solving the alleged dilemma here? What is it that the social rules help us to
between determinism and free will: splitting the recognize the bodily movement as? As an action?
world into two domains, one subjected to causality, That is, as a bodily movement recognized
as ... ?
another beyond it, has also some modern fol? The of an action remains
concept unexplained.
lowers,16 of which Melden (1961) perhaps is the Moreover, social rules are not even infallible
most emphatic one.17 instruments for the identification of actions, for
Applied to our example, Melden's view may be whereas the driver according to the rules of the
expressed thus: Smith's bodily movements, e.g., road was signalling, actually he was perhaps
his pressing the trigger, can be described in pointing
to a church, or his arm was made to move
behavioral or neuro-physiological terms and are by an electric impulse transmitted to his brain by
causally determined (he takes the term "cause" in an evil
spirit. Furthermore, a person may act even
what he calls a "Humean sense"), but Smith's if he is a nonconventional actor doing something
action?his killing his aunt?although involving which we have not been trained to understand.
bodily movements, is not identical with a bodily He may have to explain the nature of his action to
movement, and neither caused nor uncaused. the puzzled observers. I find Melden's view un?
I shall limit my discussion to three questions: acceptable, and first of all because he outs the
16 See in this connection also MacKay (i960) and K?rner (1964), but their views, different from each are not based
other,
on the kind of arguments which here will be considered.
17 on Melden
Critical comments will be found, e.g., in Ewing and Hamlyn also G. J. Warnock's "Actions
(1963-64) (1964). Cp.
and Events" in Pears (1963).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 l86 AMERICAN PHILOSOPHICAL QUARTERLY

rules of society where he should have put the action, etc., are
parts of a
general revival of so
intention of the individual.18 called philosophical psychology. Let us look at two
Melden tries to establish (ii), not by proving it, of the points which often are made by recent
but by trying to show that our usual explanations writers within this tradition, and let us formulate
of actions are of non-causal nature. For instance, to them so that we can use them as
vaguely starting
explain Smith's killing by saying that he had points for different interpretations: (a) a decision
decided to kill, is not a causal statement, for his or action is not a definite entity, and (b) we decide
decision is intelligible only as a decision to kill. and act on the basis of reasons. Let us limit our
Very seldom, however, do we explain actions discussion to two questions : (i) Does (a) or (b) have
with reference to decisions, but forgetting about any consequences for the question whether or not
that, the important thing is that his argument is our decisions or actions are caused ? (ii) Does (a) or
fallacious. Even if Smith's decision is described as (b) have any consequences for the question
a decision to kill Olga, his decision may or may not whether or not we have it in our power to decide
be followed by action. Perhaps he died just after or act otherwise ?
his decision, or someone held his hand, and even if I start with (a) and distinguish three inter?
he acted, he may not do what he decided to do. pretations, formulating them in terms of our
And even if he did what he decided to do, the example.
empirical relation between his decision and action (a?) :even themost minute observations of Smith's
may exemplify a universal relation between behavior is compatible with different hypotheses
decisions of a certain type and actions of a certain about what he was doing. His action may have been
type. Finally, even if Melden had been right on killing Olga, but it may also have been trying-to
- out - - - - - - or -
only some
this point, it would have revealed find whether the is loaded not, or
gun
features of our common ways of explaining actions. trying-to-find-out-whether-or-not-the-bullet-would
It would not have shown that a causal model is -go-through-her-heart,
or
getting-rid-of-an-ob
inapplicable to actions. stacle-in-order-to-get-hold-of-Olga's-cat.
Let us assume, for the sake of argument, that (at) seems acceptable, although observations of
Melden had succeeded in proving that actions are a of Smith's behavior may cer?
larger segment
this have any significance tainly limit the class of plausible classifications. But
beyond causality. Would
for the free-will issue ?Melden does not deny that since I cannot see that (a^) has any consequences
Smith's bodily movements are causally determined. of the type indicated in (i) and (ii), I shall say no
Hence, if the judge says (basing himself on more about it.

Melden) "I hold you morally because (a2) : the statement that Smith killed Olga is not
responsible
action was neither caused nor uncaused," an statement.
your empirical
Smith can answer (basing himself on Melden)
: (a2) is false. Whether Smith killed Olga or not
"Yes, that is true, but my killing involved a set of depends, among other things, upon his behavior
bodily movements and they were causally deter? and intentions, and to decide what he intended is
mined. They could not have been different from an empirical task.19 But even if (a2) were true, it
what they
were. Hence, your condemnation is un? would be irrelevant to the question of Smith's
justified." power to have acted otherwise. On the other hand,
Melden, I am happy to say, has not tried to if we take the term "caused" in such a way that
show that the statement that an action is neither only empirical events can be caused, and interpret
an action is not an empirical
caused nor uncaused implies that the agent had it (a2) to imply that
in his power to act otherwise. So even if the thesis event, then it follows that actions are not caused.
of his book were true, he would not have shown (a3) : Smith's action can be described in different
that we are free in the sense required by moral ways. We can say "Smith killed Olga" or "Smith
guilt or blame. shot Olga" or "Smith shot Olga with his red gun,"
etc.

(?3) is acceptable, but has no implications for the

VI. Reason Versus Cause
question whether or not his action was caused. But
Hampshire's and Melden's studies of intention, could we say that the possibility of alternative des?
choice, action, voluntary action, reason for an
criptions gives a wonderful solution of the question
18
Gp. Yolton (1962).
19
Cp. Geach (i960).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 RECENT WORK ON THE FREE-WILL PROBLEM
187

whether or not Smith had it in his power to act VII. Moral and Neutral Libertarianism
otherwise? He did shoot her, but he could have
C. A. Campbell
acted otherwise, for he did have it in his power to
do the action describable by the sentence "Smith Campbell is one of the most energetic defenders
shot Olga with his red gun." But this must be a of the so-called libertarian position both before and
joke. within the period selected for this article. His
Let us turn to (b). Does the statement "Smith position can be summarized as follows:
had a reason for killing Olga" imply or exclude (i) The crucial question is not whether there is
that "Smith's killing Olga was caused"? That some sense in which we are free, but whether we
depends, first of all, upon what we mean by are free in the sense required in order that we shall
"reason." If by Smith's "reason" we refer, inter "deem it proper to attribute moral praise or blame
alia, to his pro-attitude toward killing Olga (cp. to agents" (1957, p. 159).
Davidson, p. 687), it is rather obvious that to (ii) In order to find this sense, we must analyze
explain Smith's action in terms of his reasons may what Campbell calls the "moral consciousness."
be an example of a causal explanation. On the (Hi) Such analysis shows (a) that the crucial
other hand, if by Smith's "reason" we refer to question is whether our choices and decisions (our
Olga's death, saying that the fact that she would "inner acts")
are free or not, because even if
die if he pulled the trigger, was the reason why he external circumstances prevent a
morally wrong
pulled the trigger (cp. Shwayder, p. 86), then a act, we may be responsible for having decided to
reason is not a cause, although Smith's anticipation do this act, and, on the other hand, we are not
of the fact in question may have been a causal responsible for an action unless it is rooted in an
factor. Similarly, if by Smith's "reason" we refer to "inner act," for we do not consider robots respon?
what he thinks justifies his action ("She always sible. It also shows (b) that in order that we shall
was so angry with her cat"), his reason is not a be proper objects of moral praise or blame, our
cause, although his desire to improve the cat's decisions must be free in the sense that the two
conditions may have been a causal factor. Finally, following conditions are fulfilled:
if "reason" refers to the goal toward which his (1) the sole cause (author) of the decision is the
action was a means
(e.g.,
to become a
cat-owner), agent himself, and not his character as formed by
heredity and environment
then, a reason is not a causal
again, necessarily (p. 160).22
factor, but Smith's belief that shooting would (2) the agent could have decided otherwise than
help him to realize his goal might be. Hence, we he in fact did, in a contra-causal sense of "could"
sometimes use "reason" in such a way that a
(p. 164).
reason is a causal factor, but even when we do not
Campbell does not maintain that every decision
use it that way, the statement that an action was is free in this sense. Our decisions to follow this
done for a reason does not that it was not or that desire are caused and hence not free in the
imply
caused.20 sense indicated. It is only the choice between fol?
Does the fact that we act on the basis of reasons lowing one's strongest desire and doing what one
have any consequences for the question whether believes one ought to do, which is causally unde?
we have it in our power to act otherwise than we termined and has the self as sole author. In his
in fact do ?A plausible answer seems to be that we own words: "In the act of deciding whether to put
do have a certain degree of power both over the forth or withhold the moral effort required to resist
amount of positive or negative value that we temptation and rise to duty,23 is to be found an act
attribute to a certain anticipated consequence as which is free in the sense required for moral
well as over the weight we give this value as com?
responsibility" (p. 168).
pared with other values. By modifying our Campbell's main reason for maintaining that we
evaluations, or by trying to see our decision or are free in the sense indicated is that an analysis of
action within a wider perspective of norms and our "moral consciousness" shows that it is impos?
values, we may increase our power to decide and sible for us to disbelieve that we are free in that
act in different ways.21 sense. I find it hard to take this line of argumenta
20
Gp. Ayer (1963), pp. 266 ff.
21
Gp. Hampshire (1959), p. 129 and Ofstad (1961), pp. 196-203.
22
Also Chisholm stresses, in his lecture the role of the self in human freedom.
23 (1964),
Campbell says that this means "doing what one believes we ought to do." See (1957), p. 173.

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 l88 AMERICAN PHILOSOPHICAL QUARTERLY

tion seriously, but since its weaknesses have been indicates, for this sense does not exclude that it is a
pointed
out
by many authors,24 I shall say no more matter of luck whether the effort will turn out to
about it, but rather consider two other questions: be of sufficient strength. He probably also means
Does freedom in the sense indicated suffice for that the agent is free to mobilize the amount of
moral responsibility? Has Campbell formulated effort required in order to resist any temptations
the sense of "free" which he himself seems to (p. 221). But if this is what he means, then he
consider crucial? should have argued that we have it in our power
First of all, his position seems to imply that a to put forth this amount of effort, and not that we
wrongdoer who did not think about how he ought have contra-causal freedom, for freedom in the
to act cannot be morally responsible (p. 221)?a latter sense (no matter its exact meaning) does not
exclude that the was it
very strange consequence. Secondly, Campbell agent powerless. Moreover,
seems to postulate a very odd principle of causality, is hard to reconcile his contra-causal sense of
for it withdraws in the moment the agent thinks "could" with his belief that it may be more or less
about his duty and reappears as soon as he forgets difficult to "rise to duty" (p. 220). Also this belief
about it. Thirdly, even a person who thought about suggests that he has inmind a comparative concept
what he ought to do is not responsible if he did not of power to decide. My guess, therefore, is that a
decide to put forth or withhold effort. For if he definition of "freedom" in terms of power to
did not, then he did not do the act which Campbell decide would have been a more
adequate expres?
considers free. Fourthly, let us suppose that sion of his intentions.
Eichmann deliberated about whether to have the
shot, we assume, was the solution
Jews Ax, (which, R. Taylor
he desired) or killed by gas, A2, (which, we assume,
he considered his duty), and that he decided to do Contrary to Campbell, Taylor in his contri?
what he believed to be his duty. Let us also assume bution toHook (1958), and in his book (1963), does
that his decision was free in Campbell's sense. Does not limit our freedom to situations where we choose
this imply that it was free in the sense required in between following our strongest desire or doing
order that he should be morally responsible? It what we think we ought to do. But he shares
does not, because it does not imply that Eichmann Campbell's view that causal indeterminacy is
had it in his power to do something which would necessary but not sufficient for freedom in the sense
have been less evil than Ax or A2. Fifthly, let us required by moral responsibility. It is also necessary
suppose that if Eichmann had decided to save that the act was an act of our self. By "self" Taylor
the Jews, they would have been saved. Even this seems to refer to a
transempirical entity, and the

does not show that he was for statement that we have such a self cannot be sub?
morally responsible,
perhaps he did not have it in his power to believe jected to empirical tests. But why should a person
that it could be right to act contrary to Hitler's be morally responsible if his act was rooted in this
order, and even if he did have this in his power, entity, and without responsibility if
transempirical
perhaps he did not have it in his power to act on it was caused by his very substantial desire to kill ?
the basis of this belief. Finally, even if "Eichmann Let us assume, for the sake of argument, that

himself" was the "sole cause" of his decision, how Smith is a self of the kind Taylor seems to have in
can we avoid the consequence that also that part mind, and that this self "originated" the act of
of Eichmann which was formed by heredity and killing (Ai). This does not imply that Smith had it
environment will experience
our reactions?25 in his power not to kill Olga. Perhaps Smith's self
Campbell's philosophy of freedom seems to have originated Ax with such a strength that he was
been formulated by a person living within an unable to act otherwise. Moreover, if Smith was
unproblematic moral universe where the paradigm powerless in so far as Ax was caused by Smith's
case of potential wrong-doing is: Shall I steal a "self," defined, inter alia, in terms of his thoughts
little bit of mother's jam or shall I "rise to duty"? and desires, how can we exclude the possibility that
Let us turn to the second question.
I do not he is just as powerless ifwe define him in terms of a
think Campbell really means that a decision to put certain transempirical or substantial self? Must we
forth or withhold effort is free only in the sense he not
postulate
a
super-substantial
self to prevent
24 can be found
Campbell's way of thinking was defended by Lehrer (i960). Criticism in Bradley (1958), Franklin (1962),
Nowell-Smith (1954). Campbell's answers will be found in (1958), (1962) and to NowellSmith in (1957), pp. 214-228.
25 does not pretend to correspond to the historical facts.
Obviously, my use of this example

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 RECENT WORK ON THE FREE-WILL PROBLEM l8g

that Smith becomes a slave of his substantial self? I must be permitted to say, therefore, that this self
And a self to ..., and so on ? or seems to live a rather life."
super-super-substantial power-center quiet
And if Taylor answers that the objection is illegi? My hypothesis is that the libertarian will try to
= his substantial meet this objection by denying that the scientist
timate, because Smith self, why
not avoid the whole mystery by setting Smith = his can give perfect predictions. Such predictions are
empirical self? impossible, he may maintain, because the trans

empirical entity manifests itself in the empirical

domain in a way which for the scientist must appear
Indeterminism, Libertarianism, and theBelief in
as partly random. In other words, my hypothesis is
Transempirical Entities that the libertarian's indeterminism is intended as
The best working hypothesis when dealing with a wall of defense around his belief in certain
an author that moral entities.26
maintaining responsibility transempirical
presupposes freedom in the sense that the agent's
decision was uncaused or unpredictable is that his
VIII. Hard Determinism and Hard
thinking is confused. Probably he has failed to see
Evolutionism
the difference between the following two inter?
pretations of "P, in S, could have decided D2 The position of the hard determinist consists of
instead of D": (i) It was equally probable that P, three standpoints: (i) he asserts that our actions
in S, should make the decision D2 as Dx, and (ii) and decisions are parts of causal sequences (ances
P, in S, had it in his power to decide D2 instead of torially caused), (ii) he has a certain view on the
Dx. The latter statement tells us something about morally correct or
"proper"
sense of "moral
the agent. The former tells us something about responsibility," and (Hi) he maintains that moral
Dx and D2's relations to other events in the responsibility in this sense requires freedom in a
universe. Failing to distinguish between them, the sense which is incompatible with (i).
author may try to prove (?), intending to prove (ii). I have earlier distinguished two variations of
But there is also another and deeper reason why a hard determinism. According to the first, the
philosopher may defend indeterminism: he may principle of causality is incompatible with the
be dissatisfied with our empirical self or our moral aspect of guilt and blame, but we may be
limited power and postulate the existence of a justified in holding people guilty and blaming
transempirical self or power-center, defending them in a non-moral sense. According to the
indeterminism in order to exclude the applica? second, the principle of causality is incompatible
bility of Occam's razor. with the notions of desert and merit, but we may be
Campbell does not use the expression "trans? justified in making use of moral as well as non
empirical power-center," but his to moral blame on utilitarian
unwillingness grounds.
accept empirical disconfirrnation of the statement Recently hard determinism of the first kind has
"F had it in his power to decide D2 instead of D^9 been defended, among others, by Paul Edwards in
suggests that he postulates something like such a his contribution to Hook He does not
(1958).27
power-center. Taylor does not use the expression deny that the principle of causality is compatible
"transempirical self," but it is rather clear that he with freedom of action, if by calling an action
refers to a
transempirical entity. He seems un? "free" we mean only that it may be influenced
willing to evaluate the empirical self highly enough the use to
through of blame (limiting ourselves
to identify it with one's real self. Given this situa? He does not even deny that we use
wrongdoings).
tion, libertarians, such as Campbell and Taylor, the expression "moral responsibility" in such a
may come to defend indeterminism to meet the way that it freedom in this sense only. But
requires
following objection of the scientist: "I don't know he maintains that this is an improper sense, and
whether we have a
transempirical self or power that if we take "moral responsibility" in its
center, or not. I don't even understand what these "proper" sense, then the principle of causality
expressions mean. I know only that I don't have implies that we are not morally responsible. The
to worry about such things. I can the crucial are what Edwards means
predict questions, then,
characteristics of our decisions without paying any by "moral responsibility in the proper sense," why
attention to the operation of any such entities. And he holds this to be the proper sense, and why the
26
Cp. at this point Ofstad (1955), (1961), pp. 232-234, and (1963).
27
Baylis (1958), pp. 37-38, seems to take the second position, but I shall not go into his reasoning here.

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 IQO AMERICAN PHILOSOPHICAL QUARTERLY

principle of causality excludes responsibility in this in his power not to have killed her; and we have
sense.
thereby reached the crucial question: Whether
Edwards gives no explicit answers to the first and Smith on May i, 1965, had it in his power to act
second question, but he seems to mean that a otherwise. No power over his character or over

judgment about guilt or blame, to be moral in the anything else can help if he lacked the required
proper sense, must be independent of violent amount of power in that situation.
emotions and based on general principles.28 Let Hence, the critical thing for the hard determin
us accept this and let us turn to the problem ist is to prove that there is a moral-responsibility
which is most relevant from the point of view of relevant sense of "having something in one's
this article: Why does the principle of causality power" which is such that the agent's power to
exclude that we can be morally responsible in the have abstained from that for which he is held
"proper"
sense.
responsible (a decision, an action, etc.) is reduced or
Edwards' answer, applied
to our
example,
can annihilated ifwe assume that his decision or action
perhaps be thus expressed: In order that we shall was
part of a causal sequence. Edwards' attempt
be justified in morally blaming Smith for killing to show that we have not chosen our own charac?
can
Olga, he must have had it in his power to act ter?even forgetting about what this really
otherwise. To have had such power, he must have mean?misses the point.
chosen his own character (p. 123). But the prin? Whereas Edwards tried to reach his conclusions
ciple of causality implies that our character is by arguing from so-called deterministic premisses,
formed by heredity and environment. And since other philosophers, like Hospers, have tried to
this principle must be accepted, it follows that we reach similar conclusions from premisses referring
are not justified in morally blaming Smith. to hereditary and psychological factors (cp. Hospers
This reasoning seems to me confused.29 First of (1961) and his contribution to Hook (1958) ).31
all, the principle of causality implies at most that Hospers distinguishes between what he calls an
Smith's character is caused, not that it is caused "upper" and a "deeper" level of moral discourse.
by heredity and environment. Secondly, even if it If we conduct our discussion on the former level, we
did imply this, it does not follow that Smith was are justified in considering people free in so far
unable to abstain from killing. Thirdly, even if it as they would have acted otherwise if they had
followed that Smith was powerless in this respect, wanted to, and to make use of moral blame in so
it does not follow that he would be less powerless far as this helps to make them do what they ought
if he had chosen his own character. For the fact to do. But to talk about deserving moral blame, we
that Smith in the year 1900 chose his character must move on to a
"deeper"
level of discourse.
does not guarantee that he on May 1, 1965, had it Hospers' position seems to be that we would
in his power to abstain from killing. Moreover, deserve blame for our wrongdoings only if we had
what does it mean to "choose one's own charac? caused our own character. But neither Smith nor

ter" ?30And who chose the character of that Mr. anyone else is free in this sense, for our characters

Smith who in 1900 chose the character of Mr. are formed by hereditary and environmental factors.
Smith? The arguments which showed that Edwards'
Let us make the plausible assumption that view is untenable show also that Hospers' view is
Edwards requires that Smith in 1900 not only untenable. Rather than repeating my arguments,
chose his character, but that he also had it in his I shall comment on an important difference
power to choose another character than the one between their views. Whereas Edwards means that
he actually chose. Then this gives us a sense of we are never free and responsible in what he calls
"having something in one's power," and if we the "proper sense," Hospers,
at some
places
in his

apply this sense to the situation where Smith killed article, seems to mean that we may be free in such
Olga, it is conceivable that in this sense he had it a sense that we deserve moral blame.

28 must be based on emotions?not reasons?but he gives no reasons

He seems to mean that theories stressing modifiability
for this.
29Pitcher that it does not distinguish between the theses that some causes
(1961) criticizes hard determinism, arguing
excuse and that all causes excuse.
30 and Sentimentalism" to Hook to this unclarity, see
Hook, in his contribution "Necessity, Indeterminism, (1958), points
that it is just as meaningless to talk about "choosing my own character" as "eating my own
pp. 187-188. And Pap maintains
character." See "Determinism, Freedom, Moral Responsibility, and Causal Talk" in Hook (1958), p. 215.
31A
somewhat similar point of view has been developed by Matson (1956).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 RECENT WORK ON THE FREE-WILL PROBLEM 191

Let us assume that Smith on May i, 1965, had it principle of causality is compatible with freedom
in his power to overcome his urge to kill. According in the sense required by moral responsibility, then
to Hospers, he does not deserve praise for this, for an indeterminist can be a mild determinist, and
it is, just as his character, a product of hereditary this is a confusing terminology. Hence, I shall
and environmental factors. But once he has this require that a philosopher, in order to be called a
power, what shall we say if he does not make use "mild determinist," accepts that our decisions and
of it ? In that case, Hospers says, moral exhortations actions are
parts of causal sequences.
may be useful. But I think he means more than In what sense of "morally responsible" does mild
that, for the utility-point-of-view belongs to the determinism maintain that we can be morally
"upper level," whereas Hospers, when he makes responsible? Some of them seem to take this
this point, is discussing at the "deeper level." He expression as meaning that behavior is modifiable
probably means that in this case Smith deserves through the use of praise and blame. But according
moral blame. Does this imply that he means that to this usage, hard determinism implies mild deter?
Smith in this case has caused his own character? minism. That mild determinism shall be different
I think it rather indicates that he has moved on to from hard determinism in an interesting way, the
a
different, and, morally speaking,
more relevant mild determinist must maintain that the principle
meaning of "free," viz., whether or not one was of causality is compatible even with freedom in
able to?had it in one's power to?overcome one's the sense required in order to be morally respon?
urge to kill. If we interpret his view in this way, his sible in the sense accepted as the "proper" sense
stress on the traumatic effect of early childhood by the hard determinist, e.g., in the sense that one
may be seen, not as an can be morally guilty or But not all
experiences alleged proof blameworthy.
that we are never free in the sense required by mild determinists make this contention. One reason
moral responsibility in the so-called deeper sense, for this may be that they consider such notions as
but as emphasizing some of the factors which may moral guilt or blameworthiness as logically con?
lead to a reduction of our power to decide and act tradictory. Another reason may be that the mild
as we
ought
to. determinists agree with the hard determinists that
Some hard determinists consider their premisses causality is incompatible with freedom in the sense
as true a priori, and maintain, therefore, that the required by moral guilt or blameworthiness, but
conclusion that we are not free in the sense required conclude, in opposition to the hard determinists,
by moral responsibility, is also true a priori. The that this shows that these notions are unacceptable,
hard evolutionist, on the other hand, argues on the and that the correct notion is modifiability. The
basis of empirical statements and can prove at most difference between them would then be that the
that it is very unlikely that we are free in the mild determinist asserts, and the hard determinist
required sense.32 However, if he takes as his denies, that modifiability can be a morally relevant
the statement that we have not sense of
premiss empirical "responsible."33
caused or chosen our character, or that it is caused In the following I shall limit my discussion to the
by early childhood experiences, then he can prove reasoning of those who at least appear to maintain
his thesis only if he can establish an empirical rela? that the principle of causality is compatible with
tion between, e.g., some childhood experiences freedom in the sense required in order to be
and, on the other hand, the adult's lack of power morally guilty or blameworthy.
to decide otherwise. Empirical research of relevance
to the establishment of such relations has barely P. H. Nowell-Smith
started.
Nowell-Smith, one of the ablest defenders of mild
determinism of the last decade,34 does not define
IX. Mild Determinism
a notion of moral
guilt, but looking at the list of
If "mild determinism" refers to the view that the sentences whose presuppositions he intends to dis
32
Hook, it seems to me, is quite right in criticizing some hard determinists for oscillating between that it is self
holding
to assert that we are free in the sense
contradictory required by moral responsibility, and that we seldom, if ever, are free in
the required sense. Op. cit., pp. 187-188.
33
Gp. Smart (1961).
34 in favor of this view can be found also in, e.g., Blanshard,
Arguments op. cit., Ducasse "Determinism, Freedom and
in Hook op. cit., Pap, op. cit., Brandt
Responsibility" (1958), Hook, (1959), Mandelbaum (1959-60), Ofstad (1961), ch. VI,
sect. 18, Zink (1962), and Vivian (1964). Gp. also
Hampshire (1965).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 AMERICAN PHILOSOPHICAL QUARTERLY
ig2

cuss (1954, p. 270), we seem justified in interpreting him ?Nowell-Smith's view seems to be that positive
his view in the following way : as well as
negative evidence for this statement must

(i) Smith may be morally responsible for killing be based upon observations of Smith's behavior in
Olga both in the sense that he was morally bad in other situations. For instance, we must to
try
doing what he did (cp. my sense of "moral guilt"), establish that Smith in earlier situations has been
and in the sense that he was
blameworthy. able to overcome impulses to kill. Positive evidence
That someone shall be justified in holding Smith of this kind is always open to the objection: "How
responsible in these two senses, two conditions can we be certain that this shows that he "could
must be fulfilled : have abstained on May i, 1965?" and negative
(ii) In killing Olga, Smith did something evidence is open to a similar objection: "How can
we exclude the that he might have been
morally wrong. possibility
(Hi) Smith does not have an acceptable excuse able to abstain on May 1 ?", but,
according to
for what he did. the author, it would be absurd to base responsi?
Smith may excuse himself in different ways, but bility, as well as the absence of responsibility, on
I shall limit myself to Nowell-Smith's discussion of the logical possibility of such contingencies.
the excuse that Smith did not act freely. The Let us now return to the question of how
morally relevant sense of "free" is, according
to Nowell-Smith distinguishes between those if-substi
Nowell-Smith, that the agent could have acted tutions which exclude guilt and blameworthiness
otherwise. If Smith could not have acted other? and those which do not. His view on this point may
wise, then he is not guilty and not blame? be interpreted in different ways. The most plausible
worthy. If he could have acted otherwise, then he interpretation seems to me to be that
is guilty and blameworthy, unless he has another (v) If Tx is true only for such if-substitutions
acceptable
excuse. The questions
are then: How which the moral rules accept as excuses, then Smith
does Nowell-Smith interpret "Smith could have can excuse himself.
acted otherwise" (T)? Is it true that freedom in For instance, moral rules accept unconsciousness
this sense is compatible with the principle of as an excuse, and if Smith was unconscious when

causality? Does freedom in this sense suffice for he killed Olga, so that Tx is true only if he had not
guaranteeing that Smith was free in the sense been unconscious, then he has an
acceptable
excuse. But if he would have acted otherwise
required by guilt and blameworthiness? if he
According to Nowell-Smith, had tried to, then he is responsible, for the moral
(iv) Tis equivalent to "Smith would have acted rules do not accept not-trying
as an excuse.

otherwise, if this or that condition had been ful? Nowell-Smith mentions "acted under physical
filled" (Tx). compulsion," "acted under ignorance of fact" as
But it is always possible to find a substitution examples of excuses which the moral rules accept,
which will make Tx true. Hence, we need some and "not wanting to do what is right," "being too
criterion to distinguish between permitted and for? dishonest,"as typical examples of causes which the
bidden substitutions, otherwise Smith can never rules do not accept as excuses.
excuse himself by saying that he was not free. He tries to explain why moral rules accept only
Before looking at Nowell-Smith's criterion, let us, certain special conditions as excuses. His explana?
however, present his view as regards what kind of tions seems to be that
statement T is and how it can be verified and (vi) Moral rules accept as excuses only condi?
falsified. tions of such a kind that if wrongdoers acted under
His view seems to be that T cannot be directly such conditions, then, as a general rule, it will be
verified or falsified by observing what Smith did, for pointless to blame them.
the fact that he killed Olga neither proves nor dis? If Smith would have acted otherwise only if he
proves that he could have acted otherwise. To estab? had not been unconscious or subjected to physical
lish jTwe must show that Smith had the capacity to compulsion, then he has an acceptable excuse
abstain from killing and that nothing in the situa? because it is, as a general rule, pointless to blame
tion prevented him from making use of his capacity. persons who acted wrongly for such reasons. And
To disprove T it suffices to show that he either it is pointless because the purpose of blaming is to
lacked the capacity or that something prevented strengthen
or weaken certain traits of character.
him from making use of it. How can we prove that Simplifying his view as much as I can, it seems
Smith had the capacity and that nothing prevented to me to boil down to this :

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 RECENT WORK ON THE FREE-WILL PROBLEM I93

(vii) Smith's act in S ismorally wrong and free Using this terminology, my view can be expressed
if and only if Smith's act is in conflict with a moral by saying that it is the particular and not the
rule, and it is, in general, possible to discourage general use of "power" which is relevant to
people from acting in the way Smith did by questions of moral guilt and blame. The general
blaming them. (Let us express this by saying that use of "power" is of significance only indirectly,
the act is modifiable in general). as evidence for or against the possession of power
If (vii) is fulfilled, then Smith acted freely and is in the particular sense.

guilty and blameworthy. And he is guilty and In his answer (i960) to J. L. Austin's criticism
blameworthy even if blaming him will increase the of his view (1956), Nowell-Smith seems more aware
likelihood that he will kill again. On the other of the distinction between the particular and the
hand, if Smith's act is not modifiable in general, general use of "could" than in Ethics, and he raises
then he was not free and is neither guilty nor the question whether we can exclude the possibility
blameworthy, even if blaming Smith would have that a person who usually is able to do something
had good effects on his character. may lack the ability "just then and there" (p. 97),
I find these consequences unacceptable. The but he does not discuss the modifications which
emphasis on general modifiability may be con? would have to be made in his earlier view if we
venient within legal systems, but in moral contexts accept the particular use of "could" as the morally
we tend to hold that a person is not guilty unless relevant one.
he?the specific person in question?had it in his This brings me to another point. Both in Ethics
power to act otherwise. Nowell-Smith thinks about and in his answer to Austin, Nowell-Smith is too
moral problems in a too legalistic way. He talks much influenced by the idea that the wrongdoer
about moral rules as if they constituted a definite must have had an ability to act otherwise. But to
group of rules roughly as legal rules. And he does say that he must have been able to act otherwise
not clearly distinguish between the conditions is not the same as to say that he must have had an
which certain rules recognize
as valid excuses, and ability to do so.36 Smith may have had all the
the conditions which ought to be accepted as valid abilities pertaining to Ax and A2 that a person can
excuses.
have, but nevertheless have been powerless, because
Nowell-Smith would probably object to this it was a question of resisting a temptation, and to
criticism by saying that his requirement that the resist a temptation is not an ability. It may be
agent could have acted otherwise refers to what the granted that Smith's situation may be similar to
person in question was able to do in a particular the situation of the frustrated golfer in the sense
situation. But this line of defense is excluded, that both tried to do something and failed. But
because it is the analysis of this requirement which their situations are very different. It makes good
has led to (vii). sense to say "I am
taking
a course in how to hole
At this point it may be clarifying to distinguish six-foot putts. I am doing quite well. Very
soon I
between a
general and a
particular
use of "could" shall usually succeed when I try, and then I shall
and "power"-sentences.35 If we take "could" or obtain my diploma for having acquired an ability,"
"power" in a
particular sense, saying that a person but the following would be quite strange: "I am
could or had it in his power to do A 2 instead of taking a course in how to resist temptations to kill
Ax, in S, means that he in that particular situation aunts. I am doing quite well. Last week I resisted
was fully able to do A2. The fact that he never six out of ten As soon as I can resist
temptations.
before has been able to do so is not logically incom? eight out of ten (if there are any aunts left at that
patible with the assertion. The fact that he always time!), I shall get my diploma for having acquired
has been able to do so does not logically imply it. an ability." It ismisleading to say that moral guilt
We are considering one isolated moment. If we or blame presupposes the ability to act otherwise.
take "could" or
"power" in a general sense, saying This is one of the reasons why I prefer to say that
that a person could or had it in his power to do A 2 guilt presupposes that the agent had it in his power
instead of Ax, in S, means that he generally is able to act otherwise.
to do such actions as A2. In this sense the agent As indicated above, Nowell-Smith points out
could have acted otherwise in S even if he was that T can be proved or disproved only through
unable to act otherwise in the particular sense. inductive evidence, which, of course, is true. But
36
Cp. Honor? (1964).
38
Cp., e.g., Locke (1962) and Thalberg (1961-62).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 AMERICAN PHILOSOPHICAL QUARTERLY
194

when he classifies persons emphasizing the un? his requirement in such a way that it suffices that
certainty of such evidence as
ordinary sceptics of Smith's future behavior will be modified. Saying
induction, he misses an
important point, viz., that that freedom in this sense can distinguish between
the uncertainty of the evidence primarily is due to exculpatory and non-exculpatory cases, is, how?
the difficulty of getting hold of relevant evidence. ever, not very interesting, since it becomes analyti?
That Smith's record of previous not-killing shall cally true. And ifwe interpret his requirement to be
count as evidence for the hypothesis that he had it that the agent would have acted otherwise, if he
in his power to resist his urge to kill Olga, we must had tried to act otherwise, his justification dis?
know that the previous cases are similar in all appears, for it may be useful, both in general and
relevant to the one. And we do in the case, to blame a person even if
respects present particular
not know this, and this lack of knowledge is not a he would not have acted otherwise if he had tried
consequence of the inductive character of the to, and it may be useless even if he would have
evidence. acted otherwise if he had tried to.37 The require?
The answer to the second question I raised in ment that Smith had it in his power to act other?
the beginning of my discussion of Nowell-Smith's wise, may, however, come in as an
empirical
view, whether or not freedom in his sense is com? requirement: the likelihood of producing good
patible with the principle of causality, obviously effects through the use of blame may be higher if
is in the affirmative. Let us turn to the third, and we blame only persons who in some sense had it
more interesting question: whether freedom in in their power to act otherwise. But once we have
Nowell-Smith's sense gives the agent the freedom raised this empirical question, we see that also
required by moral guilt and blame. other senses of "free" may be relevant, for the
He does not give
a clear answer to this question.
attitudes of the wrongdoer may be such that he is
He begins his discussion by distinguishing between more likely to be modified if he experiences the
guilt, blameworthiness, and blame, but he does blame as justified than if he experiences it as
not discuss the presuppositions of guilt and blame unjustified, and perhaps he will not experience it
as justified unless he acted freely in the following
separately, and in the end the concept of individual
moral guilt disappears completely. His view seems sense. . . .
Hence, in the same sense in which it, on
to be that a person is guilty if and only if he has Nowell-Smith's view, may be an empirical require?
disobeyed a moral rule and none of the excuses ment that a wrongdoer could have acted otherwise,
admitted by the rule applies to him, and this it may also be an empirical requirement that he
means that a person is guilty if and only if he has acted freely in other senses.
acted contrary to the moral rules. And this is
legalistic thinking about moral problems.
What about free will as a condition of blame? In J. L. Austin
several of his publications Nowell-Smith has A lively debate over the correctness, especially
charged philosophers for introducing concepts of the linguistic, but to some extent also the moral
freedom whose moral relevance is not clear. As adequacy of interpreting "P could have acted
opposed to them, he maintains that his own sense : otherwise" as "P would have acted otherwise, if
"P would have acted otherwise, if this or that con? he had chosen or decided or tried to" (T2), has
dition had been fulfilled" can be justified as taken place within the decade, especially as a
morally relevant, because it is pointless to blame result of Austin's paper (1956).38 Mild deter?
unless the agent was free in some such sense. But, minists have been attacked by a united front of
as shown above, his requirement boils down to a indeterminists, libertarians, and hard determinists.
about And The main lines of attack may be summarized as
requirement general modifiability.
blaming Smith may discourage him from future follows :
wrongdoings even if the act in question is not (i) T2 does not express what we in ordinary
modifiable in general, and, even if it is, blame may language
mean when we say that someone could
have no effects on Smith. Suppose we then interpret have acted otherwise.

37
Cp. Hart (1962), p. 25.
38The discussion elicited by Austin's includes
contributions
paper by Lehrer (1959-60), Nowell-Smith (i960), O'Connor (i960),
Goldberg & Heidelberger Lehrer
(1960-61),
(1961-62), Thalberg (1961-62), Bradley (1962), Kaufman (1962),
Taylor (i960),
Locke (1962), Dore (1962-63), Scarrow (1962-63), Whiteley (1962-63), Aune (1962-63), Baier (1963), Henschen-Dahlquist (1963),
Dore (1963-64), Ewing (1963-64), and Honor? (1964).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 RECENT WORK ON THE FREE-WILL PROBLEM 195

(ii) Freedom in the sense formulated in T2 is not since all he wants to know is whether human
equivalent to freedom in the sense of having it in beings are modifiable through the use of praise and
one's power to act otherwise. blame. On the other hand, a philosopher accepting
(Hi) Freedom in sense T2 is not equivalent to that human beings can be morally guilty may main?
freedom in the sense required by moral guilt and tain that guilt presupposes that the agent had it in
blame. his power, in a sense, to have acted
categorical
One of Austin's central contentions is that we otherwise, but this philosopher certainly would
use such sentences as "Smith could have acted have to stand for his requirement even if no one
otherwise" with indeterministic implications, and actually used "could"-sentences so as to
correspond
that if-analyses therefore are inconsistent with with it.
actual usage. Moreover, even if all of us always used "could"
I see no reason to deny that so-called ordinary sentences to express categorical statements, this

usage is flexible enough to include a usage of the does not imply that it would be incorrect to
kind indicated by Austin. But the question is analyze "could"-sentences partly by the use of
whether this observation has any consequences for "would-have-if"-sentences.39 Finally,
even if it
the free-will problem. It seems to me that it has no were completely misleading, from a linguistic
such implications. Even if everyone always used point of view, to interpret "could"-sentences by
such sentences as "Smith could have acted other? the use of "would-have-if"-sentences, it might
wise" in an indeterministic sense, this would not nevertheless be freedom in the sense of whether or
imply that it is freedom in this sense which is not Smith would have acted otherwise if he had
required by moral responsibility. To find the tried to, which is relevant from the point of view
morally relevant senses of "freedom," we must raise of moral guilt and blame. Many of the articles
the question of freedom in relation to different about how to
analyze "can"- and "could"- sen?

types of moral systems, e.g., teleological and tences are uninteresting from the point of view of
deontological systems. If we, e.g., presuppose act the free-will debate because the authors fail to see
utilitarianism, it certainly is not freedom in an that the free-will problem partly is of moral nature.
indeterministic sense which is relevant.
The most fruitful part of Austin's paper is his
The Role of the Self
attempt to show that such sentences as "Smith
could have acted otherwise" express categorical To declare Smith morally guilty, we require that
statements about the past, they say that Smith was the wrong act, Al9 in some sense can be attributed
able to do something, and not that he would have to him.40 Furthermore, he must have had it in his
been able to do something, if this or that condition power to act otherwise, i.e., to do A2 instead of
had been fulfilled. Aj, and if A2 had occurred, we
require that it was
I consider it very likely that such sentences at his action. How shall we conceive of the relation
least sometimes are used in this categorical way. between an action and the self?
But Austin's great mistake is to believe that this Taylor (1963), and in his article in Hook (1958),
categorical sense must be an indeterministic sense, Campbell (1957) and Chisholm (1964), among others,
whereas it quite often refers to the power seem to the existence of a
agent's postulate transempirical
to decide or act otherwise. But the crucial or substantial self. Forgetting
point, of about the difficulties
course, is that linguistic hypotheses of this kind are of interpreting the meaning of this use of "self"
not relevant to the free-will issue, for even if we or what kind of evidence would be relevant for
assume that "could"-sentences always express and against it, it is hard to understand the nature
categorical statements about our power to decide of the relation between the substantial or trans
and act, this would neither imply that we have empirical self and such empirical things as deciding
such power, nor that this sense of "freedom" is and acting. As long as this problem has not been
morally relevant. An act-utilitarian, e.g., will say clarified, it seems reasonable to remain sceptical
that this sense is of no crucial to him, toward this way of connecting action and self.
importance
39
See, e.g., Nowell-Smith (i960), p. 93, and cp. also Locke (1962), Whiteley (1962-63), and Ewing (1963-64).
40
Gp. in this connection Wilson (1958). See also Rails' discussion of the relation between responsibility and personal identity
(1963). Philippa Foot defends the tough line that "The argument that it will be a different me who will be beaten tomorrow
carries no weight, for 'different' or not, the back which will be beaten is the one about which I am concerned today," (1962),
p. 78.
B

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 AMERICAN PHILOSOPHICAL QUARTERLY
I96

R. B. Edwards and D. H. Whittier, defending an of Smith's mental deliberation. And no such proof
empirical notion of the self, have argued that "the has been given.
self is its activities as well as its experiences" (see Suppose it was not Smith's deliberation, but
e.g., Edwards [1965], p. 274). If we accept this certain electric impulses transmitted to his brain,
line of thinking, Smith's trying, deliberating, which led to Ax and which would have led to A2.
choosing, and deciding are activities of his self, and How would this affect his responsibility ?Whether
it may be empirically possible to connect his act or not we can attribute the action to Smith will
to some of these activities and thereby to his self. depend upon his amount of control over these
But perhaps the relation between such activities electric impulses. For instance, if he knew that if
and the self here becomes too intimate? It may be he decided not to kill Olga, then electric impulses
useful to distinguish, e.g., between "trying to do A would be transmitted to him in such a way that he
occurs in Smith" and "Smith tries to do A.99 In would succeed in refraining from killing her, then
order that the trying shall be an activity of his he cannot excuse himself by saying that he did not
self, we must require, it seems to me, that it also have it in his power to refrain from killing her. He
in some ways corresponds
to his norm- and value can excuse himself only if he had decisive reasons
system. But I shall not go into this question here, to believe that he was so completely powerless that
because, from the point of view of moral responsi? there was no point in trying, and then he can
bility, the decisive thing is that the agent had con? excuse himself even if he actually would have suc?
trol over those activities which led to his action, ceeded in doing A2, if he had tried to.
not that he considered them as parts of his self.
An example: Suppose we had asked Eichmann Power toDo What?
whether or not he had it in his power to refrain
from sending the Jews to the gas-chambers, and The requirement that the agent, to be respon?
that he had answered: "Yes, if I had tried to stop sible, must have had it in his power to act other?
the transportations, I would have succeeded, but wise, is not satisfied by the fact that the agent had
such a trying would not have been a trying of the it in his power to do something else than what he
good and obedient, but of the bad and disobedient did. A 2must be different from Ax in a morally
Eichmann, and hence it would not really have relevant respect. To console Smith, driven by
been my trying at all." This would not be a irresistible impulses toward killing Olga with his
relevant excuse. Eichmann might be guilty
even red gun, that he is free to kill her with his black
if the trying would not have been his trying in this gun, would be like saying to a man whose body has
sense of the term. been trapped under a tree: "Why do you com?
Let us turn to another point pertaining to the plain? You can still move your ears" (09Shaugnessy
was to
relation between action and self. Let us grant that [1956]). But even if his only alternative
Smith's deliberation was part of his self. How can poison her, he may be guilty of having shot her
we know that his action was brought about through instead of having poisoned her?even if he is not
his deliberation and not through something else, morally responsible for having killed her.
e.g., a certain neuro-physiological activity
or a To say that the agent must have had it in his
certain electric impulse from outside ?Let us accept, power to act otherwise does not imply that he must
for the sake of discussion, that Smith's deliberation, have had it in his power to succeed in doing that
conceived as a mental or psychological process, in which he had it in his power to do instead of the
some way corresponds
to a certain neuro-physio? thing he did. Smith is responsible for killing Olga
logical process. Even if we then assume that this even if the only thing he could have done instead
neuro-physiological process was caused by some would have been to try to comfort her, for trying to
previous neuro-physiological processes, and related comfort her is succeeding in doing something else
by a law of nature to his action, this does not imply than killing her.41
that his mental deliberation was unrelated to his
action. To prove that it was, one would have to to Try?
Do We Have It in Our Power
prove that the correlations between the act and the
neuro-physiological processes in question would Let us accept that "P, in S, had it in his power
remain the same no matter what the characteristics to do A 2 instead of A" implies "P, in S, would

41
Cp. Baier (1963).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 RECENT WORK ON THE FREE-WILL PROBLEM I97

have done A2 instead of Al9 if he had chosen or impossible for him to do something, provided that
decided or tried to do A 2." (Let us select "if he had he knows what would be involved in doing it. He
tried to do A2" as representative of such condi? can
always set out on the course of action, even if

tions.) The latter sentence, however, does not imply he knows that he will encounter difficulties at the
that P had it in his power to try to do A2. Suppose very first stage of effective action, and even if he
he did not have it in his power to try, in what sense believes that he will in fact fail before any effective
did he then have it in his power to do A2? Appar? action has been taken" (pp. 182-83).
ently only in the sense in which a beheaded Whether or not we shall accept the thesis that in
person had it in his power to talk: "he would normal situations one can
always try, depends upon
have talked if he had kept his head on." This is what we mean by "trying." If we are willing to say
a serious objection
to the would-have-if-analyses.
that Smith tried not to kill Olga in so far as he
The suggested solutions are: whispered certain sounds to himself, then we may
(i) "P had it in his power to try" must itself probably assert, on empirical grounds, that we
be analyzed by the use of an
if-analysis, e.g., always have it in our power to try,43 for those who
as "P would have tried, if he had tried to try or cannot
whisper certain sounds to themselves are
decided to try." not rational animals. But if we take "trying" in a
(ii) "P had it in his power to try" must be stronger sense referring to the making of an effort,
analyzed indeterministically, viz., as "P's trying or perhaps even a strong and energetic effort like
not-trying to do A2 in S are uncaused." thinking about the consequences, studying the case
or meaningless
(Hi) It is misleading to ask such quite closely, etc., then it is probably false that we
questions as "Does he have it in his power to try?" can
always try.
"Does he have it in his power to choose?" Those defending that we always have it in our
(iv) We always have it in our power to try. power to try tend to interpret "try" as referring to
(v) "P had it in his power to try" can be what we can call "low-integrated" activities, where

analyzed by the use of a categorical statement. trying becomes something like talking to oneself.
(i) does not work, because it leads to an endless But thereby they decrease the likelihood that the
regress, (ii) does not work because saying that statement "If he had tried, he would have done it"
Smith's trying was not caused does not imply that will be true unless in very unimportant cases, like
he had it in his power to try. Both things have often taking one tie instead of another.
been pointed out, and I shall say nothing more Suppose Smith had it in his power to try not to
about them. kill Olga, but that he did not think about the
Thalberg (1964) argues that such concepts as possibility of trying to do so. Ler us even suppose
capacity, being able to, and probably also? that given his intelligence, upbringing, and imagin?
although he does not say so explicitly?having ation, he could not even be expected to think about
in one's power, have no to this. maintains that a. person, in such
something application Hampshire
normal cases of deciding and trying. Such concepts a situation, can excuse himself (p. t86), and I agree
apply
to actions, not to the will.42 He tries to with him, if "could not be expected to" implies
show this by an example: ". . . if we take a high that the person was
powerless. It is a virtue of
school graduate who is faced with a choice of two Hampshire's exposition that he discusses such cases
or three universities, it would be puzzling to ask, and tries to combine limitations of power with
"Can he make up his mind?has he the patience, limitations of intellect. But if he really wanted to
the skill, the means for reaching a decision?" develop a philosophy of the freedom of the mind,
(p. 413). I am unable to find the question puzzling. not only of the will, he should also have discussed
Let us turn to the fourth suggestion. According the influence of a
person's moral convictions.
to Hampshire, Smith can excuse himself by saying Suppose Smith would have done A2 if he had tried
that he tried with all his power and failed, or that to, suppose he did think about trying and that he had
he abstained from trying because he knew that he it in his power to try, but was firmly convinced
would fail (1959, p. 183), but he cannot excuse that it would be morally wrong to try. Did he then
himself by saying that he did not have it in his have freedom of mind ? A person's moral convic?
power to try, for "there is no sense in which it is tions may be one of the most important limitations
42
Cp. also Pap, op. cit.
48
Skinner (1963), defending the view that we have it in our power to try and to decide, seems to take "try" and "decide"
in very weak senses.

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 AMERICAN PHILOSOPHICAL QUARTERLY
198

of his power to try?for bad or for good. I may be action, this gives a nice "solution" of our problem:
aware of the possibility of killing my child. It may unless a person is abnormal, he is
always
able to
be true that if I try to do it, I shall do it, and I may try. The difficult task of finding out whether or
know that this is so, but nevertheless be unable to not Smith had it in his power to try is reduced to
do it, because I am unable to set out on a course of the simpler task of finding out whether or not he
action which I so strongly feel to be morally wrong. was crazy. But this is
unacceptable, for a person
Let us suppose that Smith did think about the who in the strong sense of "try" could try to do
possibility of trying not to kill Olga, that he did any action, far from being normal, would be a
not consider it morally outrageous to let her live, very strange creature.
and that he was not subject to an irresistible urge Even if we assume that a normal person can try
to kill. Given these suppositions, we conclude that to do any action in a weak sense of "try," and at
Smith, so far as we can see, had it in his power to least some actions in a stronger sense of "try," it
even in a sense of But what can does not follow that a
try, strong "try." particular person, e.g.,
this mean, if we reject indeterministic or hypo? Smith, had it in his power to try in a sense of "try"
thetical analyses ? strong enough tomake it likely that if he had tried,
One possible answer is given by Locke (1962). He he would have done A2 instead of Av
suggests that "P could have tried" means "P had The above discussion suggests that the difficulty
the opportunity to try" which means that nothing of formulating a categorical sense of "power" has
prevented him from trying (p. 253). Locke's sug? nothing to do with the principle of causality, but
gestion is valuable in so far as it interprets "P could is a result of our insufficient of the
knowledge
try" as an empirical statement about P, but is, in processes involved in trying, deciding, and acting.
the end, not very helpful. A person who is empty But no matter its wording, the relevant sense will
inside is not prevented from trying to do A2, but make "P had it in his power to try" to an empirical
does not for that reason have it in his power to try. statement, for to assert that a person had it in his
On the other hand, a person may have it in his power to try is to advance an empirical hypothesis
power to try even if his attempts will meet with about him. We may be able to show that the
strong inner resistance. statement is false, because we can show that the
A number of authors have tried to show that to person was
prevented in some way. But even when
have something in one's power cannot always everything looks perfect, and we are inclined to
mean that one by doing so and so brings about this conclude that he had the power in question, we
or that.44 There must be something which one has must reckon with the possibility that something
in one's power in a more direct sense. If we had did prevent him. Since such assertions are always
things in our power only in the former sense, we uncertain in this respect, it seems a wise policy to
would have nothing in our power, because this admit that also our assertions about people's
sense, as indicated above, leads to an endless moral guilt do not rest on rocks.
regress. Danto's concept of a basic action (1964),
though not very clear, seems to be intended to
delimit those activities which one has in one's power
in a direct sense. According to his view, trying is a Causality and Power
basic action; it is not something we bring about If the principle of causality applies to human
by first doing something else. But Danto seems to behavior, e.g., to our decisions and actions, can
believe that if a certain activity, e.g., trying, is we then have it in our power to decide or act
basic, then one can never bring it about, and this otherwise than we in fact do? Given the two
seems to me false. Any kind of activity, trying, statements :

deciding, thinking, etc., may be an example both (ii) Smith's killing of Olga (Ax) in S was caused,
of basic and not-basic activity. Even if trying and

usually is something that one does not bring about, (Hi) Smith, in S, had it in his power to do A2
one may bring it about, e.g., by deciding to try. instead of Av Does (ii) imply that (Hi) is false?
Danto tries to define a normal person in terms (Hi),
we assume, can be taken as
equivalent
to
of a certain repertoire
of basic actions, saying that (iv) Smith, in S, would have done A2 instead of
persons who are unable to do these actions are
Ax if he had tried to, and he had it in his power to
abnormal. If we assume that trying is a basic try to do A2.
44
See, e.g., Pears (1963), p. 24.

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 RECENT WORK ON THE FREE-WILL PROBLEM 199

Obviously (ii) does not imply that (iv)9s first tautologous, convertible to counter-factual, and

conjunct is false. The important question seems to simple (see, p. 364). He concludes that it is, and
be whether (ii) implies the falsity of that it therefore is logically possible to give a
(v) Smith, in S, had it in his power to try to causal explanation of being in State Two, which
do A2. implies that it is logically possible to hold both that
To get the problem as explicit as possible, we Smith's not-trying to refrain from killing Olga was
shall assume that Smith did not try to do A2, and caused and that he at the same time had the ability
that his
not-trying
was caused. Let us reformulate to try to refrain from killing her.
(ii) thus: Smith, in S, killed Olga (A-?), and did not Canfield tries to explain what he means by "free
try to do A 2 (NTA2), and both A1 and NTA2 were will," but says very little about what he means by
caused.
"ability," and does not try to show that his inter?
And for the purpose of discussing the relation pretation of this term (whatever it is) is relevant to
to (v), we can shorten this reformulation: questions of guilt and blame. He illustrates what
(i) Smith's NTA2 in S was caused. he means by the example of a person who is
The standard argument for showing that (i) tempted to laugh, but keeps himself from laughing.
implies that (v) is false, can be formulated thus : If This person, he says, does not only have the ability
NTA2 was caused, then certain conditions were to refrain from laughing, but also the ability to
causally sufficient for NTA2. If certain conditions laugh.
were causally sufficient for NTA2, then it was I find this view very strange. Applied to my own
causally impossible for Smith, in S, to try to do A2. example, it implies that from the statement that
If it was causally impossible for Smith to try to Smith refrained from letting Olga live, we can infer
do A 2, then he could not try to do A2. And if he that he was able (or even: had the ability) to let
could not try to do A2, then he did not have it in her live ! It is this strange thinking about abilities
his power to try to do A2.Ab which is the really weak point in Canfield's
An interesting feature of the recent free-will article.
debate is the attempt to show that causality is In his answer to Dore's
criticism of his article
compatible with power even if "power" is not (1963),46 Canfield (1963) argues as if he means that
analyzed in hypothetical terms. Let us first look at saying that a person has an ability to do something
an article to our is tomake a categorical, empirical statement about
by Canfield (1962). Applied
example, the main lines of his reasoning seem to be that person, for he mentions that a physical
as follows: examination might tell us whether or not a certain
(a) let us assume that Smith, in S (or, let us say, person had certain abilities. But the important thing
at time t2), when he killed is to develop
Olga, had the ability to and defend this interpretation of
refrain from killing her as well as the ability to kill "ability" (or better: of "was able to"). Once that
her. has been done, the assertion that is com?
causality
Let us refer to the state of Smith at t2 as "State patible with ability in such a sense is not more
Two," and by "State One" to that state, whatever astonishing than the assertion that causality is
it may be, whose description is a detailed descrip? compatible with being healthy.
tion of Smith and his environment at some time Another, and perhaps more promising line of
prior to t2: tv Let us now formulate the true thinking is found in an article by Aune (1962-63).
statement : He follows up Moore's well-known suggestion that
(b) At tx Smith was in State One and at t2 in we must distinguish between two different senses
State Two. Let us also presuppose that the follow? of "could-not," and develops it in a fruitful way.
ing general statement is true : Since NTA2 was caused, certain conditions, say
(c) Whenever someone is in State One at time C, were causally sufficient for JVTA2. From this it
tn, at time tn+i he will be in State Two. follows that
Since (b) is an instantiation of (c), (c) is consistent (a) it would be inconsistent with certain laws of
with (b). The important question, according to nature to assert that C was the case and NTA2 did
Canfield, is whether (b) is consistent with (c) even not occur.
if we assume that (c) is not only true, but a so This inference is not uncontroversial, for it is
called law of nature, e.g., universal in form, non relative to a certain of "was caused"
interpretation
45
See, e.g., Taylor (i960), p. 89; and cp. Ayer (1963), p. 265.
46 are criticized
Canfield's arguments also by Lehrer (1964.1).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 200 AMERICAN PHILOSOPHICAL QUARTERLY

and makes use of such expressions

as "laws of probably accepted by quite a few philosophers, but
nature," but we shall assume, for the sake of it can be denied without making logical or
argument, that the inference must be errors, for it is a normative
accepted. empirical standpoint
We can formulate (a) also in the following way : as regards one of the conditions of moral guilt. As
(b) It could not have been the case that C long
as we have not shown that this normative
occurred and Smith, in S, tried to do ^fa standpoint is more correct than alternative stand?

According to Aune, it is important to see that points, the free-will problem is not solved. A
(b) uses "could-not" as referring to a conjunction philosopher might hold an optimistic opinion con?
of factors. It says that certain conditions and a cerning our power to decide and act otherwise and
certain trying to do A 2 "add up to an impossible nevertheless maintain that we have no free will,
totality, a totality that could not have existed" because he holds that moral guilt does not pre?
(p. 411). The "could-not" does not say or imply suppose freedom in this sense, but in a sense in
that there was something Smith could not do. which we are not free. On the other hand, a
Those maintaining that (i) implies that (u) is false philosopher maintaining that we do not have it in
confuses two different senses of "could-not," one our to decide and act
power otherwise, may
referring to the total lack of likelihood that a nevertheless hold that we have a free will because
certain system, Smith-in-S, shall give
a certain he, e.g., believes that the notion of moral guilt is
output (an attempt to do A2), the other referring nonsensical, and that the question
of moral
to the powerlessness of a certain person, (i) implies responsibility is the question of the usefulness of
the first sense of "could not", (v) denies "could blame. If he could show that this is the morally
not" in the second sense.47 correct concept of responsibility, the positions of
the indeterminist, the libertarian, and the hard
determinist or evolutionist would be without
Power and Empirical Conditions interest to the moral philosopher.

The fact that the principle of causality does not

Austin's "A Plea for Excuses" (1956-57) was one
of the first publications within the recent debate to
annihilate or reduce our to decide or act
power
stimulate interest in the moral context of the free?
otherwise than we in fact do, does not imply that
will problem, but judging from his later "Ifs and
we have it in our power to decide and act other?
wise. Our power may be limited by lack of imagi? Cans," he did not clearly understand the implica?
tions of his own
nation and inner conflicts and ethical approach.
knowledge, to show that judg?
etc. Hence, what one man can do is
Beardsley (1960-61) attempts
convictions, ments of moral and blame may be made
not what another can do. What we can praise
necessarily from what she called different "moral perspec?
do in one situation may be different from what we
tives," e.g., the of "moral worth,"
can in another. And even if our perspectives
accomplish power moral
"moral credit," and of "ultimate equality."
to make a certain decision now is
well-integrated She also tried to show that the sense of "freedom"
quite limited, it may increase through patient
endeavor. required by praise and blame depends upon the
perspective from which the judgment is made.
Since our power to decide and act is related to
This is a fruitful insight, but the author should have
empirical conditions, itmay be fruitful to ask what
related the different perspectives to different types
kind of variations in these conditions will lead to
of moral systems, and analyzed the problems
this or that degree of power, or what kind of
without assuming that judgments of praise and
a person make use of to increase his
strategies may can true or she does not do
blame be false. Because
power in a certain respect ??Very few philosophers she sees the three as
this, perspectives supple?
have dealt with questions of this kind.48
menting each other, and does not pay enough
attention to the conflicts among them.
The moral relativity of the free-will problem is a
Ethics and Freedom as Power
central theme in Ofstad (1961), and it is stressed,
The thesis that an agent cannot be morally among others, by Kaufman (1962), Whiteley
guilty if he was powerless to act otherwise is (1962-63), and Bronaugh (1963-64).
47 "
Cp. also Ofstad (1961), pp. 91, 241. Hampshire, in his latest book (1965), seems to take a similar position. He says: 'There
"
is no possibility of his acting differently' does not entail 'He cannot act differently' (p. 32).
48 But see
Hampshire (i960), e.g., pp. 190 ff. and Ofstad (1961), pp. 185-203.

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 RECENT WORK ON THE FREE-WILL PROBLEM 201

Moral Guilt and Freedom power to act rightly, but believed that he did,

Power to act otherwise is not the only condition nevertheless, refraining from even trying to do so ?
It seems to me that such a person may be just as
of moral guilt. Knowledge of facts and the accep?
tance of moral norms are but it is
morally bad as one who actually had the power in
necessary, too, to a
Hence, power act otherwise is not
us question.
the only condition which interests in this
necessary condition of moral guilt, but the absence
article.
of such power may be a sufficient condition of
Moral guilt, just as criminal guilt, should be moral innocence.
conceived of in such a way that it can vary in
degrees. 49An agent may be guilty to a high degree
Moral Blame and Freedom
if he intentionally did evil and had a high degree
of power to act otherwise. He may be guilty to a as we may have to catch a tiger and lock
Just
low degree if he either did not expect any harmful it up, we have to catch a man wild.
may running
consequences of his behavior or had a very low And just as we may blame a dog, we may blame a
degree of power to act otherwise. A maximum of man just because it is useful. But in such cases we
guilt presupposes power to act otherwise in its should talk about non-moral, rather than moral,
highest possible degree: no effort at all was blame.
needed to act rightly. Seldom, if ever, do we have Moral blame presupposes moral guilt, but guilt
this amount of power to act otherwise, especially is not a sufficient condition. To blame a person is
not in situations where we acted wrongly. It may to act toward him in a certain and an act is
way,
be emotionally convenient to believe that the great if it makes the world worse than if we
unjustified
evildoers are maximally guilty, but quite often acted otherwise. moral blame
Hence, presupposes
seem to be closer to evil things than evil as utility.50 And we must ask : Is blame
they guilt as well
agents, petrified by their limited power to act a useful instrument of reformation, prevention,
otherwise as well as by their limited knowledge or moral
education? We do not know,
deterrence,
and moral understanding. but a large number of studies on the effects of
Does moral guilt presuppose power to decide punishment suggest that the use of blame is a very
not only power to act otherwise ?
otherwise, ineffective way of modifying people's behavior.51
I have already taken a standpoint on this issue, fulfills a function for
Perhaps blaming primarily
for I have interpreted "P, in S, had it in his power the moral whereas knowl?
judge, argumentation,
to do A 2 instead of Al9" not as "P, in S, would have or a nicer wife are more effective instruments
edge,
done A 2 instead of A1 if he had chosen or decided of change. Hence, though it is not logically contra?
or tried to do so," but as "P in S, would have done to assert that a wrongdoer should be
dictory
A2 instead of Ax if he had chosen or decided or morally blamed, the conditions of moral blame :
tried to do so, and he had it in his power to choose are seldom, if ever, fulfilled, at
guilt and utility,
or decide or try to do so." Hence, power to fulfill least not to a high degree.
the if-clause is a condition of moral guilt. More?
over, a person may be responsible for his choices, Conclusion
X.
decisions, and attempts. And in those cases his power
at the action-level is irrelevant. Let me conclude with a short survey of the posi?
Power to decide and act otherwise is a condition tive and negative traits of the different positions.
of moral guilt because a wrongdoer who was The indeterminists' notion of freedom is irrele?
powerless to act rightly is not a morally bad man. vant from the point of view of moral guilt and
An agent who acted wrongly in spite of the fact blame.
that he tried with all his energy to act rightly is The libertarians have made a contribution
through their emphasis on such factors as trying,
a So is also a who
morally good person. wrongdoer
on sufficiently good grounds held the false an effort, and through their
opinion deciding, making
that he was so completely powerless to act rightly insight that moral responsibility, at least in one
that he had no reason to try. But what should we sense, presupposes the freedom of the agent. But
say about a wrongdoer who did not have it in his by conceiving of this agent as a transempirical

49
Cp. Hart (1962), p. 17.
50 For a similar
point of view applied to legal responsibility, see Hart (1962), p. 30.
61
See, e.g., W. K. Estes "An Experimental Study of Punishment," Psychological Monographs, vol. 57 (1944).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 202 AMERICAN PHILOSOPHICAL QUARTERLY

substance, they have been led on to strange and responsibility in the senses discussed by the liber?
unclear theories of "doing," and have found it tarians and the hard determinists and evolutionists,
necessary to defend indeterminism to try to safe? and they have, just as these, paid insufficient
guard the possibility of the operation of such a attention to the notion of moral guilt. They are
transempirical substance. right in maintaining that modifiability is a neces?
Both the hard determinists and the libertarians sary condition of moral blame, but it is not a
are right in insisting that the question of moral sufficient condition, and it is neither necessary nor
responsibility is not only the question of the useful? sufficient for moral guilt.
ness of praise or blame, but also of attributing right
* * *
or wrong actions to agents. Hence, are
they right
in denying that the free-will issue can be reduced to Solving the free-will problem implies solving
the question of the modifiability of behavior, and in three different kinds of problems: (i) In what
asserting that responsibility presupposes freedom senses of "moral responsibility," if any, is it correct
in a categorical sense. But by failing to see either to assert that we may be morally responsible?
that "freedom" here must be interpreted as power (ii) In what senses of "freedom," if any, is freedom
to decide and act otherwise, or that such power is a condition of moral responsibility as clarified in
compatible with the principle of causality, they ?
(i) (Hi) Under what conditions, if any, are we free
have arrived at the conclusion that since the in the senses clarified in (ii) ? Hence, the solution
are not
principle of causality must be accepted, we involves the clarification of moral, analytical, and
free in the sense required by moral responsibility. scientific problems. The recent debate has been
The hard evolutionists understand, or ought to concerned mostly with the analytical problems and
understand, that our freedom is not threatened by has been too optimistic about the possibility that
the principle of causality, but by certain types of linguistic analysis would lead to the solution of the
causal explanations, but they have failed to make whole problem. Fruitful research presupposes that
use of this insight in order to explore the empirical the analytical instruments are used within a well
conditions of varying degrees of freedom. defined moral context and in close contact with
The mild determinists have tended to neglect scientific research on man's behavior.

University of Stockholm Received May 25, ig66

BIBLIOGRAPHY*
ADLER, M. J. (ed.) The Idea of Freedom, 2 vols. (New (reprinted in Philosophical Papers, [Oxford, 1961]).
York; -"A Plea for Excuses," Proceedings of the Aristotelian
1958, 1961).
ANSGOMBE, G. E. M. Intention (Oxford, 1957). Society, vol. 57 (1956-1957), pp. 1-30 (reprinted in
-"The Two Kinds of Error inAction" (Symposium: Philosophical Papers, [Oxford, 1961]).
Human Action), The Journal of Philosophy, vol. 60 AYER, A. J. "Fatalism" in The Concept of a Person
(I963), pp. 393-401. (London, 1963).
AUNE, B. "Abilities, Modalities, and Free Will," BAIER, K. "Could and Would," Analysis Supplement,
and Research, vol. 23 vol. 23 (1963)5 PP. 20-29.
Philosophy Phenomenological
pp. 397-413 -"Action and Agent," The Monist, vol. 49 (1965),
(1962-1963),
AUSTIN, J. L. "Ifs and Cans," Proceedings of theBritish pp. 183-195.
Academy, London, vol. 42 (1956), pp. 109-132 BAYLIS, C. A. Ethics (New York, 1958), Ch. 2.

* This covers works written in English and printed Articles have been checked in the following
bibliography 1954-1965.
American Quarterly, Analysis, The Australasian Journal of Philosophy, The British Journal for the Philosophy
periodicals: Philosophical
of Science, Ethics, The Journal of Philosophy, Mind, The Monist, Proceedings of the British Academy, The Philosophical Quarterly, The
Philosophical Review, Philosophy, Philosophy and Phenomenological Research, Proceedings of the Aristotelian Society (including supple?
mentary Theoria, and Inquiry. Some minor articles and notes have been left out.
volumes),
Useful will be found in Adler (1958/61), and Ofstad (1961). See also the short bibliography by Roland
bibliographies
and especially the annotated in P. Edwards
Hall in The Philosophical Quarterly, vol. 15 (1965), pp. 179-181, bibliography
and A. Pap (eds.), A Modern Introduction to Philosophy (revised edition, New York, 1965).

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 RECENT WORK ON THE FREE-WILL PROBLEM
203

BEARDSLEY, E. L. "Determinism and Moral Perspec? -"Knowing about Future Decisions," Analysis, vol.
tives," Philosophy and Phenomenological Research, vol. 22 (1961-1962), pp. 127-129.
21 (1960-1961), pp. 1-20. -"The Compatibility of Free Will and Deter?
BECK, L.
"Agent, Actor, Spectator, and Critic," The minism," The Philosophical Review, vol. 71 (1962),
Monist, vol. 49 (1965), pp. 167-182. PP. 352-368.
BENNETT, J. "The Status of Determinism," The British -"Free Will and Determinism: A Reply," The
Journal for the Philosophy of Science, vol. 14 (1963 Philosophical Review, vol. 72 (1963), pp. 502-504.
1964)5 pp. 106-119. CHISHOLM, R. M. "The Descriptive Element in the
BERLIN, I. "From Hope and Fear Set Free," Proceedings Concept of Action" (Symposium: Action, Motive,
of the Aristotelian Society, vol. 64 (1963-1964), and Desire), The Journal of Philosophy, vol. 61
pp. 1-30. pp. 613-625.
(1964.1),
BEROFSKY, B. "Determinism and the Concept of a -Human Freedom and the Self, The Lindley Lecture of
Person," The Journal of Philosophy, vol. 61 (1964), the University of Kansas (Kansas, 1964).
pp. 461-475. COHEN, M. F. "Motives, Causal Necessity and Moral
B?HM, D. Causality andChance inModern Physics (London, Accountability," The Australasian Journal of
1957) Philosophy, vol. 42 (1964), pp. 322-334.
BORING, E. G. "When Is Human Behavior Pre? DANTO, A. C. "The Paradigm Case Argument and the
in D. E. Dulany Free Will
determined?" (ed.), Contributions
Problem," Ethics, vol. 69 (1958-1959),
toModern Psychology (Oxford, 1963). pp. 120-124.
BRANDT, R. "Blameworthiness and Obligation," in -"What We Can Do" (Symposium: Human
A. I. Melden (ed.), Essays in Moral Philosophy Action), The Journal of Philosophy, vol. 60 (1963),
(Seattle, 1958). PP. 435-445
-Ethical Theory (Englewood Cliffs, 1959), Ch. XX. -"Basic Actions," American Philosophical Quarterly,
-and KIM, J. "Wants as Explanations of Action" vol. 2 PP-
(1965), i44~I48.
(Symposium: Human Action), The Journal of-and MORGENBESSER, S. "Character and Free
Philosophy, vol. 60 (1963), pp. 425-435. Will," The Journal of Philosophy, vol. 54 (1957),
BRADLEY, M. C. "A Note on Mr. Maclntyre's PP- 493-505.
Determinism,99 Mind, vol. 68 (1959), pp. 521-526. D'ARCY, E. Human Acts (Oxford, 1963), Ch. 3.
BRADLEY, R. D. "Free Will: Problem or Pseudo DAVENEY, T. F. "Choosing," Mind, vol. 73 (1964),
problem?" The Australasian Journal of Philosophy, PP- 515-526.
vol. 36 (1958), pp. 33-45 DAVIDSON, D. "Actions, Reasons, and Causes," The
-" and Determinism," The Australasian
'Ifs,' 'Cans,' Journal of Philosophy, vol. 60 (1963), pp. 685-700.
Journal of Philosophy, vol. 40 (1962), pp. 146-158. I. "The Freedom of Man,"
DILMAN, Proceedings of the
-"Determinism or Indeterminism inMicrophysics," Aristotelian Society, vol. 62 (1961-1962), pp. 39-62.
The British Journal for thePhilosophy of Science,vol. 13
DIODORUS CROMUS "Time, Truth and Ability,"
(1962-1963), pp. 193-215 vol. 26
Analysis, (1965-1966), pp. 137-147.
-"Causality, Fatalism,
Morality," and
Mind, vol.
DONNELLAN, K. S. "Knowing What I am Doing"
72 pp. 591-594.
(1963),
(Symposium: Human Action), The Journal of
BRAITHWAITE, R. B. (ed.) "Analysis Problem No. 7,"
Philosophy, vol. 60 (1963), pp. 401-409.
Analysis, vol. 16 (1955), pp. 1-5. (Contributions by
DORE, C. "On the Meaning of'Could Have'," Analysis,
Brian Ellis, "Candidus," and Nicholas Rescher.)
vol. 23 (1962-1963), pp. 179-181.
BRETHERTON, D. "'Ought' Implies 'Can Say',"
-"More on the Meaning of'Could Have',"
the Aristotelian vol. Analysis,
Proceedings of Society, 63 (1962
vol. 24 (1963-1964)? PP- 4I~43
1963), PP- i45-l66.
-"Is Free Will Compatible with Determinism?"
BRONAUGH, R. "Freedom as the Absence of an
The Philosophical Review, vol. 72 (1963), pp.
Excuse," Ethics, vol. 74 (1964), pp. 161-173.
500-501.
BUNGE, M. Causality: The Place of theCausal Principle in
Modern Science EDWARDS, R. "Agency Without a Substantive Self,"
(Cambridge, 1959).
The Monist, vol.
CAMPBELL, C. A. On Selfhood and Godhood (London, 49 (1965), pp. 273-289.

1957), Lecture IX. EVANS, J. L. "Choice," The Philosophical Quarterly,

-"Free Will: A Reply to Mr. R. D. The vol. 5 (i955)5 PP- 303-315.
Bradley,"
Australasian Journal of Philosophy, vol. 36 (1958), EWING, A. C. Second Thoughts in Moral Philosophy
pp. 46-56. (London, 1959).
-"Moral Libertarianism :
A toMr. Franklin," -"May Can-Statements be Analysed Deterministic
Reply
The Philosophical Quarterly, vol. 12 (1962), pp. 337 ally?" Proceedings of the Aristotelian Society, vol. 64
347 (1963-1964)* PP- 157-176.
CANFIELD, J. "Determinism, Free-Will and the Ace FAIN, H. "Prediction and Constraint," Mind, vol. 67
Predictor," Mind, vol. 70 (1961), pp. 412-416. (i958)> PP- 366-378.

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 AMERICAN PHILOSOPHICAL QUARTERLY
204

FALK, W. D. "Action-guiding Reasons" (Symposium: (Symposium :Determinism in the Light of Recent

Action-guiding Reasons), The Journal of Philosophy, Physics), The Journal of Philosophy, vol. 54 (1957),
vol. 60 (1963), pp. 702-718. pp. 7I3-727.
FARRER, A. The Freedom of theWill (London, 1958). GUSTAFSON, D. F. "Voluntary and Involuntary,"
FEINBERG, J. "On Being 'Morally Speaking' a Philosophy and Phenomenological Research, vol. 24
Murderer," The Journal ofPhilosophy, vol. 61 (1964), (1963-1964), PP- 493-501.
pp. 158-171. HAINES, N. "Responsibility and Accountability,"
-"Action and Responsibility" in Max Black (ed.), Philosophy, vol. 30 pp. 141-163.
(1955),
Philosophy in America (Ithaca, 1965). HAKSAR, V. "Aristotle and the Punishment of Psycho?
FINGARETTE, H. Perspectives on vol. 39 pp. 323-340.
"Psychoanalytic paths," Philosophy, (1964),
Moral Guilt and Responsibility: A Re-Evalua?
HALVERSON, W. H. "The Bogy of Chance," Mind,
tion," Philosophy and Phenomenological Research, vol. vol. 73 (1964), pp. 567-570.
x6 (i9b5-i9&), pp. 18-36. D. W. and Human
HAMLYN, "Causality Behaviour,"
A. "Divine and Human Freedom"
FLEW, Omnipotence Proceedings of the Aristotelian Society, supplementary
in A. Flew and A. Maclntyre (eds.), New Essays in vol. 38 (1964), pp. 125-142.
Philosophical Theology (London, 1955). HAMPSHIRE, S. and HART, H. L. A. "Decision,
-"Determinism and Rational Behaviour," Mind, Intention and vol.
Certainty," Mind, 67 (1958),
vol. 68 (1959), pp. 377-382. pp. 1-12.
FOOT, P. "Free Will as Involving Determinism," The S. Thought and Action (London,
HAMPSHIRE, 1959).
Philosophical Review, vol. 66 (1957), pp. 439-450 ?-"Reply to Walsh on Thought and Action"
(reprinted in Morgenbesser & Walsh, Free Will (Symposium: Human Action), The Journal of
[1962]). Philosophy, vol. 60(1963), pp. 410-424.
FRANKLIN, R. L. "Dissolving the Problem of Free? -Freedom the Individual
of (London, 1965).
will," The Australasian Journal of Philosophy, vol. 39 R. and the
HANDY, "Determinism, Responsibility,
(1961), pp. 111-124. Social Setting," Philosophy and Phenomenological
-"Moral Libertarianism," The Philosophical Quarterly vol. 20
Research, (1959-1960), pp. 469-476.
vol. 12 (1962), pp. 24-35. HANSON, N. R. "Causal Chains," Mind, vol. 64 (1955).
FRIEDRICH, C. J. (ed.) "Responsibility" in Nomos pp. 289-311.
III (New York, i960). W. F. R. "My Own Free Will,"
HARDIE, Philosophy,
GALE, R. M. "Endorsing Predictions," The Philosophical vol. 32 (i957)3 PP- 21-38.
Review, vol. 70 (1961), pp. 376-385. R. M. Freedom and Reason
HARE, (Oxford, 1963), Ch. 4.
GALLAGHER, K. T. "On Choosing to Choose," Mind,
HART, H. L. A. "Prolegomenon to the Principles of
vol. 73 (1964), pp. 480-495. the Aristotelian
Punishment," Proceedings of Society,
GALLIE, W. B. Free Will and Determinism Yet Again, An vol. 60 (I959-?960), 1-26.
pp.
Inaugural lecture, (The Queens University of Mens and Criminal
-"Negligence, Rea, Responsi?
Belfast, 1957). bility" in A. G. Guest (ed.), Oxford Essays in
GALLOP, D. "On Being Determined," Mind, vol. 71 in Morgen
Jurisprudence (Oxford, 1961; reprinted
(1962), pp. 181-196. besser and Walsh, Free Will [1962]).
GASKING, E. "Causation and Recipes," Mind, vol. 64 -Punishment and the Elimination of Responsibility, L. T.
-Hobhouse Memorial Trust of
(i955)? PP. 479-487. Lecture, University
GEACH, P. T. "Ascriptivism," The Philosophical Review, London, 1962.
vol. pp. 221-225. -see S. and Hart, H. L. A.
69 (i960), Hampshire, (1958).
GINET, C. "Can theWill be Caused?" The Philosophical HARTSHORNE, C. "Freedom Requires Indetermin?

Review, vol. 71 pp. 49-55. ism and Universal Causality," The Journal of
(1962),
GLASGOW, W. D. "On Choosing," Analysis, vol. 17 Philosophy, vol. 55 (1958), pp. 793~8n.
PP- 135-139. HEDENIUS, I. "Broad's Treatment of Determinism
(i956-i957)>
-"The Concept of Choosing," Analysis, vol. 20 and Free Will" in P. A. Schlipp (ed.), The Philo?
pp. 63-67. sophy ofC.D. Broad (New York, 1959).
(1959-1960),
GLASS, B. "The Relation of the Physical Sciences to HEIDELBERGER, H. see Goldberg, B., and Heidel?

Causality" and in B. berger, H. (i960-1961).

Biology-Indeterminacy
Baumrin (ed.), Philosophy of Science, The Delaware HENSCHEN-DAHLQUIST, Ann-Mari, "Remarks to
Seminar Austin's Criticism of Moore's Analysis of 'can',"
(New York, 1963).
GOLDBERG, B. and HEIDELBERGER, H. "Mr. Theoria, vol. 29 (1963)? PP- SOS^^
Lehrer on the Constitution of Cans," Analysis, vol. HERBST, P. "Freedom and Prediction," Mind, vol. 66
21 (1960-1961), p. 96. (i957)5 PP- 1-27.
GR?NBAUM, A. "Complementarity in Quantum HONOR?, A. M. "Can and Can't," Mind, vol. 73
Physics and its Philosophical Generalization" (1964), pp. 463-479.

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 RECENT WORK ON THE FREE-WILL PROBLEM
205

HOOK, S. (ed.) Determinism and Freedom in the Age of MACKIE, J. L. "Responsibility and Language," The
Modern Science (New York, 1958). (Contributions Australasian Journal of Philosophy, vol. 33 (1965),
by: W. Barrett, E. L. Beardsley, M. Black, pp. 143-159.
B. Blanshard, R. Brandt, P. W. Bridgman, R. M. -"Evil and vol.
Omnipotence," Mind, 64 (1955),
Chisholm, C. J. Ducasse, P. Edwards, H. L. A. 200-212.
pp.
Hart, C. G. Hempel, H. W. Hintz, S. Hook, D. M. A Study in Ethical Theory (New
MACKINNON,
J. Hospers, A. Lande, A. Lerner, M. K. Munitz, Collier Ch.
York, Book, 1962), 4.
E. Nagel, F. S. C. Northrop, A. Pap, A. Schultz,
MACMURRAY, J. The Self as Agent (London, 1957).
D. W. Sciama, R. Taylor, P. Weiss, and H.
MADDEN, E. H. "Psychoanalysis and Moral Judge
Wilson.) and Phenomenological
ability," Philosophy Research,
HOSPERS, J. Human Conduct (New York, i96i),Ch. io
vol. 18 (1957-1958), PP- 68-79.
KAUFMAN, A. S. "Moral Responsibility and the Use M. "Determinism and Moral
MANDELBAUM, Res?
of 'Could Have'," The Philosophical Quarterly, vol.
ponsibility," Ethics, vol. 70 (1959-1960), pp. 204
12 (1962), pp. 120-128.
219.
-"Ability," The Journal of Philosophy, vol. 60 (1963),
MATSON, W. I. "On the Irrelevance of Free-Will to
PP-537-551 Moral and
Responsibility, the Vacuity of the
KENNER, L. "Causality, Determinism and Freedom of
Latter," vol. 65 (1956),
Mind, pp. 489-497.
theWill," Philosophy, vol. 39 (1964), pp. 233-248.
MAYO, B. A."A Logical Limitation on Determinism,"
KENNY, A. Action, Emotion andWill (London, 1963). vol. 33 (1958.1), pp. 5?-55
Philosophy,
KIM, see Brandt, R. and Kim,
J. J. (1963). -Ethics and theMoral Ch. XII.
Life (London, 1958.2),
K?RNER, S. "Science and Moral Responsibility," -"The Open Future," Mind, vol. 71 (1962), pp.
Mind, vol. 73 (1964), pp. 161-172.
1-14.
LACEY, A. R. "Free Will and Responsibility," Pro?
McGILL, V. J. "Conflicting Theories of Freedom,"
ceedings of the Aristotelian Society, vol. 58 (1957-1958), and Phenomenological vol. 20
Philosophy Research,
PP. 15-32.
(1959-1960), pp. 437-452.
LEHRER, K. "Can We Know That We Have Free Will A. I. Free Action (London,
MELDEN, 1961).
by Introspection?" The Journal of Philosophy,
MILLER, G. "An Examination of Ofstad's Radical
vol. 57 (i960), pp. I45-?57
Contextualism," Inquiry, vol. 7 (1964), pp. 209
-"Ifs, Cans, and Causes," Analysis, vol. 20 (1959 218.
1960), pp. 122-124.
MONTEFIORE, A. "Determinism and Causal Order,"
-"Cans and Conditionals: A Rejoinder," Analysis,
vol. 22 Proceedings of the Aristotelian Society, vol. 58 (1957
(1961-1962), pp. 23-24.
-"Decisions and Causes," The Philosophical !958),pp. 125-142.
Review, -"
vol. 'Ought' and 'Can'," The Philosophical Quarterly,
72 (1963), PP. 224-227.
-" vol. 8 (1958), pp. 24-40.
'Could' and Determinism," Analysis, vol. 24
(1963-1964)5 PP. i59-l6? MORGENBESSER, S., and WALSH, J. (eds.), Free
-"Doing the Impossible," The Australasian Journal
Will (Englewood Cliffs, 1962). (See articles by
vol. 42 H. L. A. Hart, P. Foot, and R. Taylor).
of Philosophy, (1964.1), pp. 86-97.
-"Doing the Impossible; A Second Try," The MORGENBESSER, S. see Danto, A. C, and Morgen?
Australasian vol. 42 besser, S.
Journal of Philosophy, (1964.2),
pp. 249-251. MORRIS, H. (ed.) Freedom and Responsibility (Stanford,
-and TAYLOR, R. "Time, Truth, and Modali? 1961).
ties," Mind, vol. 74 (1965), PP- 390-398. A.
"The Self as Agent and The
MUNK, Spectator,"
LEWIS, H. D. "Freedom and Responsibility" in idem, Monist, vol. 49 (1965), pp. 262-272.
Freedom and History, (London, 1962). MUNN, A. M. Free-Will andDeterminism (London, i960).
LOCKE, D. "Ifs and Cans Revisited," Philosophy, vol. MURPHY, A. "Jonathan Edwards on Free Will and
37 (1962), pp. 245-256. Moral Agency," The Philosophical Review, vol. 68

LYON, A. "The Prediction Paradox," Mind, vol. 68 (J959)5 PP- 181-202.

(i959)? PP- 510-517 NAGEL, E. The Structure of Science (New York, 1961),
MABBOTT, J. D. "Freewill and Punishment" in Ch. 10.
H. D. Lewis (ed.), Contemporary British Philosophy NOWELL-SMITH, P. H. Ethics (New York and
(London, 1956). London, 1954), Chs. 19-20.
-"Free Will" in Encyclopaedia Britannica (1957). -"Determinists and vol.
Libertarians," Mind, 63
A. C. vol. 66
MACINTYRE, "Determinism," Mind, (i954)> PP- 3I77337.
(1957), PP- 28-41. -"Choosing, Deciding and Doing," Analysis, vol. 18
MACKAY, D. M. "On the Logical Indeterminacy of a pp. 63-69.
(1957-1958),
Free Choice," Mind, vol. 69 (i960), pp. 31-40. -"Ifs and Cans," Theoria, vol. 26 (i960), pp. 85-101.

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 206 AMERICAN PHILOSOPHICAL QUARTERLY

O'CONNOR, D. J. "Determinism and Predictability," RALLS, A. "The Ascription of Personal Responsibility

The British Journal for the Philosophy of Science, vol. 7 and Identity," The Australasian Journal of Philosophy,
(I956-I957)> PP- 310-315. vol. 61 (1963), pp. 346-358.
-"Possibility and Choice," Proceedings of the Aris? K. W. "Doer and vol.
RANKIN, Doing," Mind, 69
totelian Society, Supplementary Volume 34 (i960), (i960), pp. 361-371.
pp. 1-24. -Choice and Chance (Oxford, 1961).
OLDENQUIST, A. "Causes, Predictions, and De? -"A Deterministic Windmill," The Australasian
vol.
cisions," Analysis, 24 (1963-1964), pp. 55-58. Journal of Philosophy, vol. 41 (1963), pp. 233-245.
H. "Broad on and Can," D. D. Moral Judgment (London, 1955),
OFSTAD, Ought Theoria, RAPHAEL,
vol. 21 (i955)> PP- 105-116. Ch. X.
-"Can We
Produce Decisions?" The Journal of ROBERTS, M. Responsibility andPractical Freedom (Cam?
Philosophy, vol. 56 (1959), pp. 89-94. bridge, 1965).
-"Frankena on and vol. 68
Ought Can," Mind, ROXBEE-COX, J. W. "Can I Know Beforehand What
(i959)> PP- 73-79 I am Going to Decide?" The Philosophical Review,
-An Inquiry into the Freedom of Decision (Oslo-Stock? vol. 72 (1963), pp. 88-92.
holm-London, 1961). RYAN, A. "Freedom," Philosophy, vol. 40 (1965),
-"Libertarianism and the Beliefin Transempirical pp.93-112.
Entities, L. J. Russell on Causation and Agency" D. S. "On an Analysis of 'Could
SCARROW, Have',"
in Philosophical Essays Dedicated to Gunnar Aspelin Analysis, vol. 23 (1962-1963), pp. 118-120.
(Lund, 1963). SCRIVEN, M. "The Present Status of Determinism in
OPPENHEIM, F. Dimensions of Freedom (New York, Physics" (Symposium: Determinism in the Light
1961). of Recent Physics), The Journal of Philosophy, vol.
O'SHAUGHNESSY, B. "The Limits of theWill," The 54 (1957), PP- 727-741.
Philosophical Review, vol. 65 (1956), pp. 443-490. SELLARS, R. W. "Guided Causality, Using Reason,
-"Observation and the Will" (Symposium: Human and "Free-WhT," The Journal of Philosophy, vol. 54
Action), The Journal of Philosophy, vol. 60 (1963), (1957), pp. 485-493.
pp. 367-392. SHUTE, C. "The Dilemma of Determinism after
PAPANOUTSOS, E. P. "Freedom and Causality," Seventy-five Years," Mind, vol. 70 (1961), 331
Philosophy, vol. 34 (i959)> PP- 193-203. 350.
PEARS, D. F. (ed.) Freedom and theWill (London, 1963). SHWAYDER, D. S. The Stratification of Behaviour
(Contributions by: P. L. Gardiner, S. Hampshire, (London, 1965).
H. L. A. Hart, I. Murdoch, D. F. Pears, P. F. SILBER, J. "Human
Action and the Language of

Strawson, J. F. Thomson, G. J. Warnock, M. Volitions," Proceedings of the Aristotelian Society,

and B. Williams.) vol. 64 (1963-1964)5 PP? 199-220.
Warnock,
PERRY, D. "Prediction, Explanation and Freedom," SKINNER, R. C. "Freedom of Choice," Mind, vol. 72
The Monist, vol. 49 (1965), pp. 234-247. O963)> PP? 463-480.
PETERS, R. S. The Concept ofMotivation (London, 1958). SMART, J. J. C. "Free-Will, Praise, and Blame," Mind,
PIKE, N. "Divine Omniscience and Voluntary Action," vol. 70 (1961), pp. 291-306.
The Philosophical Review, vol. 74 (1965), pp. 27-46. -"Causality, and Human Behaviour," Proceedings of
PITCHER, G. "Hart on Action and Responsibility," the Aristotelian Society, supplementary vol. 38 (1964),
The Philosophical Review, vol. 69 (i960), pp. 226 pp. 143-148.
235 SMITH, J. W. "Impossibility and Morals," Mind, vol.
-Necessitarianism," The Philosophical Quarterly, vol. 70 (1961), pp. 362-375
11 (1961), pp. 201-212. SPAKOVSKY von Anatol, Freedom, Determinism, Indeter?

PLANTINGA, "The Free Will Defence" in Max

A. minism (The Hague, 1963).
Black (ed.), Philosophy inAmerica (Ithaca, 1965). STENNER, A. J. "On Predicting our Future," The
PLAUT, H. C. "Condition, Cause, Free Will, and the Journal of Philosophy, vol. 61 (1964), pp. 415-428.
Direction of Time," The British Journal for the STRAWSON, P. F. "Freedom and Resentment,"
vol. 11 ( 1960-1961 ), pp. 212 the British vol.
Philosophy of Science, Proceedings of Academy, London, 48
221. (1962), pp. 187-211.
POWELL, B. "Uncharacteristic Actions," Mind, vol. 68 SWIGGART, P. "Doing and Deciding to Do," Analysis,
(i959)> PP- 492-509. vol. 23 (1962-1963), pp. 17-19.
RAAB, F. V. "Free Will and the Ambiguity of'Could'," TAYLOR, R. "I Can," The Philosophical Review, vol. 69
The Philosophical Review, vol. 64 (1955), pp. 60-77. (i960), pp. 78-89 (reprinted in Morgenbesser &
-"The Relevance of Morals to our Denials of Walsh, Free Will [1962]).
Responsibility" inH. Casta?eda & C. Nakhnikian -Metaphysics (Englewood Cliffs, 1963), Chs. 4-5.
(eds.), Morality and the Language of Conduct-"Deliberation and Foreknowledge," American

(Detroit, 1963). Philosophical Quarterly, vol. I (1964.1), pp. 73-80.

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions
 REGENT WORK ON THE FREE-WILL PROBLEM 207

-"Not Trying to do the Impossible," The Austral? Research, vol. 19 (1958-1959)? PP- 34I~353
asian Journal of Philosophy, vol. 42 (1964.2), pp. 98 WHITELEY, C. H. "Can," Analysis, vol. 23 (1962
100. I9^3)> PP- 9I_93
-see Lehrer, K. and Taylor, R. (1965). WHITTIER, D. "Causality and the Self," The Monist,
THALBERG, I. "Abilities and Ifs," Analysis, vol. 22 vol. 49 (1965), pp. 290-303.
(1961-1962), pp. 121-126. WILD, J. Existence and theWorld of Freedom (Englewood
-"Foreknowledge and Decisions in Advance," Cliffs, 1963), Part III.
Analysis, vol. 24 (1963-1964)5 PP- 49~54 WILL, F. L. "Intention, Error, and Responsibility,"
-"Freedom of Action and Freedom of Will," The The Journal of Philosophy, vol. 61 (1964), pp. 171
Journal of Philosophy, vol. 61 (1964), pp. 405-415. 179.
THOMAS, G. "Comments: Abilities and Physiology," WILLIAMS, C. J. F. "Logical Indeterminacy and Free?
The Journal of Philosophy, vol. 61 (1964), pp. will," Analysis, vol. 21 (1960-1961), pp. 12-13.
321-328. WILLIAMS, G. "The Natural Causation of Human

VESEY, G. N. A., "Volition," Philosophy, vol. 36 (1961), Freedom," Philosophy and Phenomenological Research,
PP. 352-365. vol. 19 (1958-1959), pp. 529-531.
VIVIAN, F. Human Freedom and Responsibility (London, WILSON, J. "Freedom and Compulsion," Mind, vol.

1964). 67 (1958), pp. 60-69.

WAISMANN, F. "The Decline and Fall of Causality" WOOTTON, B. Social Science and Social Pathology
in A. Crombie (ed.), Turning Points in Physics (London, 1959), Ch. VIII.
(Amsterdam, 1959). YOLTON, J. W. "Act and Circumstance," The Journal

WALSH, J., see Morgenbesser, S. and Walsh, J. (1962). of Philosophy, vol. 59 (1962), pp. 337"350.
WEISSMANN, H. A. "Freedom in Ethics?A Trans ZINK, S. The Concepts of Ethics (London, 1962), Chs.
causal Concept," Philosophy and Phenomenological VI-VIII.

This content downloaded on Mon, 17 Dec 2012 10:29:08 AM

All use subject to JSTOR Terms and Conditions