Richard Swinburne The problem of evil Page 1/11

The Problem of Evil

RICHARD SWINBURNE

Introduction

[1] God is, by definition, omniscient, omnipotent, and perfectly good. By "omniscient" I
understand "one who knows all true propositions." By "omnipotent" I understand "able to do
anything logically possible." By "perfectly good" I understand "one who does no morally bad
action," and I include among actions omissions to perform some action. The problem of evil
is then often stated as the problem whether the existence of God is compatible with the
existence of evil. Against the suggestion of compatibility, an atheist often suggests that the
existence of evil entails the nonexistence of God. For, he argues, if God exists, then being
omniscient, he knows under what circumstances evil will occur if he does not act; and being
omnipotent, he is able to prevent its occurrence. Hence, being perfectly good, he will
prevent its occurrence and so evil will not exist. Hence the existence of God entails the
nonexistence of evil. Theists have usually attacked this argument by denying the claim that
necessarily a perfectly good being, foreseeing the occurrence of evil and able to prevent it,
will prevent it.

[2] And indeed, if evil is understood in the very wide way in which it normally is understood
in this context, to include physical pain of however slight a degree, the cited claim is
somewhat implausible. For it implies that if through my neglecting frequent warnings to go
to the dentist, I find myself one morning with a slight toothache, then necessarily, there
does not exist a perfectly good being who foresaw the evil and was able to have prevented
it. Yet it seems fairly obvious that such a being might well choose to allow me to suffer
some mild consequences of my folly — as a lesson for the future which would do me real
harm.

[3] The threat to theism seems to come, not from the existence of evil as such, but rather
from the existence of evil of certain kinds and degrees — of severe undeserved physical
pain or mental anguish, for example. I shall therefore list briefly the kinds of evil which are
evident in our world, and ask whether their existence in the degrees in which we find them
is compatible with the existence of God. I shall call the man who argues for compatibility
the theodicist, and his opponent the antitheodicist. The theodicist will claim that it is not
morally wrong for God to create or permit the various evils, normally on the grounds that
doing so is providing the logically necessary conditions of greater goods. The antitheodicist
denies these claims by putting forward moral principles which have as consequences that a
good God would not under any circumstances create or permit the evils in question. I shall
argue that these moral principles are not, when carefully examined, at all obvious, and
indeed that there is a lot to be said for their negations. Hence I shall conclude that it is
plausible to suppose that the existence of these evils is compatible with the existence of
God.

The Problem of Evil: Types

[4] What then is wrong with the world? First, there are painful sensations, felt both by men,
and, to a lesser extent, by animals. Second, there are painful emotions, which do not
involve pain in the literal sense of this word — for example, feelings of loss and failure and
frustration. Such suffering exists mainly among men, but also, I suppose, to some small
Richard Swinburne The problem of evil Page 2/11

extent among animals too. Third, there are evil and undesirable states of affairs, mainly
states of men's minds, which do not involve suffering. For example, there are the states of
mind of hatred and envy; and such states of the world as rubbish tipped over a beautiful
spot. And fourth, there are the evil actions of men, mainly actions having as foreseeable
consequences evils of the first three types, but perhaps other actions as well — such as
lying and promise breaking with no such foreseeable consequences. I include among
actions, omissions to perform some actions. If there are rational agents other than men and
God (if he exists), such as angels or devils or strange beings on distant planets, who suffer
and perform evil actions, then their evil feelings, states, and actions must be added to the
list of evils.

[5] I propose to call evil of the first type physical evil, evil of the second type mental evil,
evil of the third type state evil, and evil of the fourth type moral evil. Since there is a clear
contrast between evils of the first three types, which are evils that happen to men or
animals or the world, and evils of the fourth type, which are evils that men do, there is an
advantage in having one name for evils of any of the first three types — I shall call these
passive evils. I distinguish evil from mere absence of good. Pain is not simply the absence
of pleasure. A headache is a pain, whereas not having the sensation of drinking whiskey is,
for many people, mere absence of pleasure. Likewise, the feeling of loss in bereavement is
an evil involving suffering, to be contrasted with the mere absence of the pleasure of
companionship.

[6] Some thinkers have, of course, claimed that a good God would create a "best of all
(logically) possible worlds" (i.e., a world than which no better is logically possible), and for
them the mere absence of good creates a problem since it looks as if a world would be a
better world if it had that good. For most of us, however, the mere absence of good seems
less of a threat to theism than the presence of evil, partly because it is not at all clear
whether any sense can be given to the concept of a best of all possible worlds (and if it
cannot then of logical necessity there will be a better world than any creatable world) and
partly because even if sense can be given to this concept it is not at all obvious that God
has an obligation to create such a world — to whom would he be doing an injustice if he did
not? My concern is with the threat to theism posed by the existence of evil.

Objection 1: God Ought Not to Create Evildoers

[7] Now much of the evil in the world consists of the evil actions of men and the passive
evils brought about by those actions. (These include the evils brought about intentionally by
men, and also the evils which result from long years of slackness by many generations of
men. Many of the evils of 1975 are in the latter category, and among them many state
evils. The hatred and jealousy which many men and groups feel today result from an
upbringing consequent on generations of neglected opportunities for reconciliations.) The
antitheodicist suggests as a moral principle (P1) that a creator able to do so ought to create
only creatures such that necessarily they do not do evil actions. From this it follows that
God would not have made men who do evil actions. Against this suggestion the theodicist
naturally deploys the free-will defense, elegantly expounded in recent years by Alvin
Plantinga. This runs roughly as follows: it is not logically possible for an agent to make
another agent such that necessarily he freely does only good actions. Hence if a being G
creates a free agent, he gives to the agent power of choice between alternative actions, and
how he will exercise that power is something which G cannot control while the agent
remains free. It is a good thing that there exist free agents, but a logically necessary
Richard Swinburne The problem of evil Page 3/11

consequence of their existence is that their power to choose to do evil actions may
sometimes be realized. The price is worth paying, however, for the existence of agents
performing free actions remains a good thing even if they sometimes do evil. Hence it is not
logically possible that a creator create free creatures "such that necessarily they do not do
evil actions." But it is not a morally bad thing that he create free creatures, even with the
possibility of their doing evil. Hence the cited moral principle is implausible.

[8] The free-will defense as stated needs a little filling out. For surely there could be free
agents who did not have the power of moral choice, agents whose only opportunities for
choice were between morally indifferent alternatives — between jam and marmalade for
breakfast, between watching the news on BBC 1 or the news on ITV. They might lack this
power either because they lacked the power of making moral judgments (i.e., lacked moral
discrimination); or because all their actions which were morally assessable were caused by
factors outside their control; or because they saw with complete clarity what was right and
wrong and had no temptation to do anything except the right. The free-will defense must
claim, however, that it is a good thing that there exist free agents with the power and
opportunity of choosing between morally good and morally evil actions, agents with
sufficient moral discrimination to have some idea of the difference and some (though not
overwhelming) temptation to do other than the morally good. Let us call such agents
humanly free agents. The defense must then go on to claim that it is not logically possible
to create humanly free agents such that necessarily they do not do morally evil actions.

[9] Unfortunately, this latter claim is highly debatable, and I have no space to debate it. I
propose therefore to circumvent this issue as follows. I shall add to the definition of
humanly free agents, that they are agents whose choices do not have fully deterministic
precedent causes. Clearly then it will not be logically possible to create humanly free agents
whose choices go one way rather than another, and so not logically possible to create
humanly free agents such that necessarily they do not do evil actions. Then the free-will
defense claims that (P1) is not universally true; it is not morally wrong to create humanly
free agents — despite the real possibility that they will do evil. Like many others who have
discussed this issue, I find this a highly plausible suggestion. Surely as parents we regard it
as a good thing that our children have power to do free actions of moral significance — even
if the consequence is that they sometimes do evil actions. This conviction is likely to be
stronger, not weaker, if we hold that the free actions with which we are concerned are ones
which do no have fully deterministic precedent causes. In this way we show the existence of
God to be compatible with the existence of moral evil — but only subject to a very big
assumption — that men are humanly free agents. If they are not, the compatibility shown
by the free-will defense is of little interest. For the agreed exception to (P1) would not then
justify a creator making men who did evil actions; we should need a different exception to
avoid incompatibility. The assumption seems to me not clearly false, and is also one which
most theists affirm for quite other reasons. Needless to say, there is no space to discuss the
assumption here.

Objection 2: Against Passive Evil

[10] All that the free-will defense has shown so far, however (and all that Plantinga seems
to show), is grounds for supposing that the existence of moral evil is compatible with the
existence of God. It has not given grounds for supposing that the existence of evil
consequences of moral evils is compatible with the existence of God. In an attempt to show
an incompatibility, the antitheodicist may suggest instead of (P1), (P2) — that a creator
Richard Swinburne The problem of evil Page 4/11

able to do so ought always to ensure that any creature whom he creates does not cause
passive evils, or at any rate passive evils which hurt creatures other than himself. For could
not God have made a world where there are humanly free creatures, men with the power to
do evil actions, but where those actions do not have evil consequences, or at any rate evil
consequences which affect others — e.g., a world where men cannot cause pain and
distress to other men? Men might well do actions which are evil either because they were
actions which they believed would have evil consequences or because they were evil for
some other reason (e.g., actions which involved promise breaking) without them in fact
having any passive evils as consequences. Agents in such a world would be like men in a
simulator training to be pilots. They can make mistakes, but no one suffers through those
mistakes. Or men might do evil actions which did have the evil consequences which were
foreseen but which damaged only themselves.

[11] I do not find (P2) a very plausible moral principle. A world in which no one except the
agent was affected by his evil actions might be a world in which men had freedom but it
would not be a world in which men had responsibility. The theodicist claims that it would not
be wrong for God to create interdependent humanly free agents, a society of such agents
responsible for each other's well-being, able to make or mar each other.

[12] Fair enough, the antitheodicist may again say. It is not wrong to create a world where
creatures have responsibilities for each other. But might not those responsibilities simply be
that creatures had the opportunity to benefit or to withhold benefit from each other, not a
world in which they had also the opportunity to cause each other pain? One answer to this is
that if creatures have only the power to benefit and not the power to hurt each other, they
obviously lack any very strong responsibility for each other. To bring out the point by a
caricature — a world in which I could choose whether or not to give you sweets, but not
whether or not to break your leg or make you unpopular, is not a world in which I have a
very strong influence on your destiny, and so not a world in which I have a very full
responsibility for you. Further, however, there is a point which will depend on an argument
which I will give further on. In the actual world very often a man's withholding benefits from
another is correlated with the latter's suffering some passive evil, either physical or mental.
Thus if I withhold from you certain vitamins, you will suffer disease. Or if I deprive you of
your wife by persuading her to live with me instead, you will suffer grief at the loss. Now it
seems to me that a world in which such correlations did not hold would not necessarily be a
better world than the world in which they do. The appropriateness of pain to bodily disease
or deprivation, and of mental evils to various losses or lacks of a more spiritual kind, is
something for which I shall argue in detail a little later.

[13] So then the theodicist objects to (P2) on the grounds that the price of possible passive
evils for other creatures is a price worth paying for agents to have great responsibilities for
each other. It is a price which (logically) must be paid if they are to have those
responsibilities. Here again a reasonable antitheodicist may see the point. In bringing up our
own children, in order to give them responsibility, we try not to interfere too quickly in their
quarrels — even at the price, sometimes, of younger children getting hurt physically. We try
not to interfere, first, in order to train our children for responsibility in later life and second
because responsibility here and now is a good thing in itself. True, with respect to the first
reason, whatever the effects on character produced by training, God could produce without
training. But if he did so by imposing a full character on a humanly free creature, this would
be giving him a character which he had not in any way chosen or adopted for himself. Yet it
would seem a good thing that a creator should allow humanly free creatures to influence by
their own choices the sort of creatures they are to be, the kind of character they are to
have. That means that the creator must create them immature, and allow them gradually to
Richard Swinburne The problem of evil Page 5/11

make decisions which affect the sort of beings they will be. And one of the greatest
privileges which a creator can give to a creature is to allow him to help in the process of
education, in putting alternatives before his fellows.

Objection 3: The Quantity of Evil

[14] Yet though the antitheodicist may see the point, in theory, he may well react to it
rather like this. "Certainly some independence is a good thing. But surely a father ought to
interfere if his younger son is really getting badly hurt. The ideal of making men free and
responsible is a good one, but there are limits to the amount of responsibility which it is
good that men should have, and in our world men have too much responsibility. A good God
would certainly have intervened long ago to stop some of the things which happen in our
world." Here, I believe, lies the crux — it is simply a matter of quantity. The theodicist says
that a good God could allow men to do to each other the hurt they do, in order to allow
them to be free and responsible. But against him the antitheodicist puts forward as a moral
principle (P3) that a creator able to do so ought to ensure that any creature whom he
creates does not cause passive evils as many and as evil as those in our world. He says that
in our world freedom and responsibility have gone too far — produced too much physical
and mental hurt. God might well tolerate a boy hitting his younger brothers, but not Belsen.

[15] The theodicist is in no way committed to saying that a good God will not stop things
getting too bad. Indeed, if God made our world, he has clearly done so. There are limits to
the amount and degree of evil which are possible in our world. Thus there are limits to the
amount of pain which a person can suffer — persons live in our world only so many years
and the amount which they can suffer at any given time (if mental goings-on are in any way
correlated with bodily ones) is limited by their physiology. Further, theists often claim that
from time to time God intervenes in the natural order which he has made to prevent evil
which would otherwise occur. So the theodicist can certainly claim that a good God stops
too much sufferings — it is just that he and his opponent draw the line in different places.
The issue as regards the passive evils caused by men turns ultimately to the quantity of
evil. To this crucial matter I shall return toward the end of the paper.

The Interconnectedness of Good and Evil

[16] We shall have to turn next to the issue of passive evils not apparently caused by men.
But, first, I must consider a further argument by the theodicist in support of the free-will
defense and also an argument of the antitheodicist against it. The first is the argument that
various evils are logically necessary conditions for the occurrence of actions of certain
especially good kinds. Thus for a man to bear his suffering cheerfully there has to be
suffering for him to bear. There have to be acts which irritate for another to show tolerance
of them. Likewise, it is often said, acts of forgiveness, courage, self-sacrifice, compassion,
overcoming temptation, etc., can be performed only if there are evils of various kinds. Here,
however, we must be careful. One might reasonably claim that all that is necessary for
some of these good acts (or acts as good as these) to be performed is belief in the
existence of certain evils, not their actual existence. You can show compassion toward
someone who appears to be suffering, but is not really; you can forgive someone who only
appeared to insult you, but did not really. But if the world is to be populated with imaginary
evils of the kind needed to enable creatures to perform acts of the above specially good
Richard Swinburne The problem of evil Page 6/11

kinds, it would have to be a world in which creatures are generally and systematically
deceived about the feelings of their fellows — in which the behavior of creatures generally
and unavoidably belies their feelings and intentions. I suggest, in the tradition of Descartes
(Meditations 4, 5 and 6), that it would be a morally wrong act of a creator to create such a
deceptive world. In that case, given a creator, then, without an immoral act on his part, for
acts of courage, compassion, etc., to be acts open to men to perform, there have to be
various evils. Evils give men the opportunity to perform those acts which show men at their
best. A world without evils would be a world in which men could show no forgiveness, no
compassion, no self-sacrifice. And men without that opportunity are deprived of the
opportunity to show themselves at their noblest. For this reason God might well allow some
of his creatures to perform evil acts with passive evils as consequences, since these provide
the opportunity for especially noble acts.

[17] Against the suggestion of the developed free-will defense that it would be justifiable for
God to permit a creature to hurt another for the good of his or the other's soul, there is one
natural objection which will surely be made. This is that it is generally supposed to be the
duty of men to stop other men hurting each other badly. So why is it not God's duty to stop
men hurting each other badly? Now the theodicist does not have to maintain that it is never
God's duty to stop men hurting each other; but he does have to maintain that it is not God's
duty in circumstances where it clearly is our duty to stop such hurt if we can — e.g., when
men are torturing each other in mind or body in some of the ways in which they do this in
our world and when, if God exists, he does not step in.

[18] Now different views might be taken about the extent of our duty to interfere in the
quarrels of others. But the most which could reasonably be claimed is surely this — that we
have a duty to in-terfere in three kinds of circumstances — (1) if an oppressed person asks
us to interfere and it is probable that he will suffer considerably if we do not, (2) if the
participants are children or not of sane mind and it is probable that one or other will suffer
considerably if we do not interfere, or (3) if it is probable that considerable harm will be
done to others if we do not interfere. It is not very plausible to suppose that we have any
duty to interfere in the quarrels of grown sane men who do not wish us to do so, unless it is
probable that the harm will spread. Now note that in the characterization of each of the
circumstances in which we would have a duty to interfere there occurs the word "probable,"
and it is being used in the "epistemic" sense — as "made probable by the total available
evidence." But then the "probability" of an occurrence varies crucially with which community
or individual is assessing it, and the amount of evidence which they have at the time in
question. What is probable relative to your knowledge at t1 may not be at all probable
relative to my knowledge at t2. Hence a person's duty to interfere in quarrels will depend on
their probable consequences relative to that person's knowledge. Hence it follows that one
who knows much more about the probable consequences of a quarrel may have no duty to
interfere where another with less knowledge does have such a duty — and conversely.
Hence a God who sees far more clearly than we do the consequences of quarrels may have
duties very different from ours with respect to particular such quarrels. He may know that
the suffering that A will cause B is not nearly as great as B's screams might suggest to us
and will provide (unknown to us) an opportunity to C to help B recover and will thus give C
a deep responsibility which he would not otherwise have. God may very well have reason
for allowing particular evils which it is our bounden duty to attempt to stop at all costs
simply because he knows so much more about them than we do. And this is no ad hoc
hypothesis — it follows directly from the characterization of the kind of circumstances in
which persons have a duty to interfere in quarrels.
Richard Swinburne The problem of evil Page 7/11

[19] We may have a duty to interfere in quarrels when God does not for a very different
kind of reason. God, being our creator, the source of our beginning and continuation of
existence, has rights over us which we do not have over our fellowmen. To allow a man to
suffer for the good of his or someone else's soul one has to stand in some kind of parental
relationship toward him. I don't have the right to let some stranger Joe Bloggs suffer for the
good of his soul or of the soul of Bill Snoggs, but I do have some right of this kind in respect
of my own children. I may let the younger son suffer somewhat for the good of his and his
brother's soul. I have this right because in small part I am responsible for his existence, its
beginning and continuance. If this is correct, then a fortiori, God who is, ex hypothesi, so
much more the author of our being than are our parents, has so many more rights in this
respect. God has rights to allow others to suffer, while I do not have those rights and hence
have a duty to interfere instead. In these two ways the theodicist can rebut the objection
that if we have a duty to stop certain particular evils which men do to others, God must
have this duty too.

Objection 4: Passive Evil Not Due to Human Action

[20] In the free-will defense, as elaborated above, the theist seems to me to have an
adequate answer to the suggestion that necessarily a good God would prevent the
occurrence of the evil which men cause — if we ignore the question of the quantity of evil,
to which I will return at the end of my paper. But what of the passive evil apparently not
due to human action? What of the pain caused to men by disease or earthquake or cyclone,
and what too of animal pain which existed before there were men? There are two additional
assumptions, each of which has been put forward to allow the free-will defense to show the
compatibility of the existence of God and the existence of such evil. The first is that, despite
appearances, men are ultimately responsible for disease, earthquake, cyclone, and much
animal pain. There seem to be traces of this view in Genesis 3:16-20. One might claim that
God ties the goodness of man to the well-being of the world and that a failure of one leads
to a failure of the other. Lack of prayer, concern, and simple goodness lead to the evils in
nature. This assump-tion, though it may do some service for the free-will defense, would
seem unable to account for the animal pain which existed before there were men. The other
assumption is that there exist humanly free creatures other than men, which we may call
fallen angels, who have chosen to do evil, and have brought about the passive evils not
brought about by men. These were given the care of much of the material world and have
abused that care. For reasons already given, however, it is not God's moral duty to interfere
to prevent the passive evils caused by such creatures.

[21] This defense has recently been used by, among others, Plantinga. This assumption, it
seems to me, will do the job, and is not clearly false. It is also an assumption which was
part of the Christian tradition long before the free-will defense was put forward in any
logically rigorous form. I believe that this assumption may indeed be indispensable if the
theist is to reconcile with the existence of God the existence of passive evils of certain
kinds, e.g., certain animal pain. But I do not think that the theodicist need deploy it to deal
with the central cases of passive evils not caused by men — mental evils and the human
pain that is a sign of bodily malfunctioning. Note, however, that if he does not attribute
such passive evils to the free choice of some other agent, the theodicist must attribute them
to the direct action of God himself, or rather, what he must say is that God created a
universe in which passive evils must necessarily occur in certain circumstances, the
occurrence of which is necessary or at any rate not within the power of a humanly free
agent to prevent. The antitheodicist then naturally claims, that although a creator might be
Richard Swinburne The problem of evil Page 8/11

justified in allowing free creatures to produce various evils, nevertheless (P4) a creator is
never justified in creating a world in which evil results except by the action of a humanly
free agent. Against this the theodicist tries to sketch reasons which a good creator might
have for creating a world in which there is evil not brought about by humanly free agents.

[22] One reason which he produces is one which we have already considered earlier in the
development of the free-will defense. This is the reason that various evils are logically
necessary conditions for the occurrence of actions of certain especially noble kinds. This was
adduced earlier as a reason why a creator might allow creatures to perform evil acts with
passive evils as consequences. It can also be adduced as a reason why he might himself
bring about passive evils — to give further opportunities for courage, patience, and
tolerance. I shall consider here one further reason that, the theodicist may suggest, a good
creator might have for creating a world in which various passive evils were implanted, which
is another reason for rejecting (P4). It is, I think, a reason which is closely connected with
some of the other reasons which we have been considering why a good creator might
permit the existence of evil.

[23] A creator who is going to create humanly free agents and place them in a universe has
a choice of the kind of universe to create. First, he can create a finished universe in which
nothing needs improving. Humanly free agents know what is right, and pursue it; and they
achieve their purposes without hindrance. Second, he can create a basically evil universe, in
which everything needs improving, and nothing can be improved. Or, third, he can create a
basically good but half-finished universe — one in which many things need improving,
humanly free agents do not altogether know what is right, and their purposes are often
frustrated; but one in which agents can come to know what is right and can overcome the
obstacles to the achievement of their purposes. In such a universe the bodies of creatures
may work imperfectly and last only a short time; and creatures may be morally ill-educated,
and set their affections on things and persons which are taken from them. The universe
might be such that it requires long generations of cooperative effort between creatures to
make perfect. While not wishing to deny the goodness of a universe of the first kind, I
suggest that to create a universe of the third kind would be no bad thing, for it gives to
creatures the privilege of making their own universe. Genesis 1 in telling of a God who tells
men to "subdue" the earth pictures the creator as creating a universe of this third kind; and
fairly evidently — given that men are humanly free agents — our universe is of this kind.

[24] Now a creator who creates a half-finished universe of this third kind has a further
choice as to how he molds the humanly free agents which it contains. Clearly he will have to
give them a nature of some kind, that is, certain narrow purposes which they have a natural
inclination to pursue until they choose or are forced to pursue others— e.g., the immediate
attainment of food, sleep, and sex. There could hardly be humanly free agents without
some such initial purposes. But what is he to do about their knowledge of their duty to
improve the world — e.g., to repair their bodies when they go wrong, so that they can
realize long-term purposes, to help others who cannot get food to do so, etc.? He could just
give them a formal hazy knowledge that they had such reasons for action without giving
them any strong inclination to pursue them. Such a policy might well seem an excessively
laissez-faire one. We tend to think that parents who give their children no help toward
taking the right path are less than perfect parents. So a good creator might well help agents
toward taking steps to improve the universe. We shall see that he can do this in one of two
ways.

[25] An action is something done for a reason. A good creator, we supposed, will give to
agents some reasons for doing right actions — e.g., that they are right, that they will
Richard Swinburne The problem of evil Page 9/11

improve the universe. These reasons are ones of which men can be aware and then either
act on or not act on. The creator could help agents toward doing right actions by making
these reasons more effective causally; that is, he could make agents so that by nature they
were inclined (though not perhaps compelled) to pursue what is good. But this would be to
impose a moral character on agents, to give them wide general purposes which they
naturally pursue, to make them naturally altruistic, tenacious of purpose, or strong-willed.
But to impose a character on creatures might well seem to take away from creatures the
privilege of developing their own characters and those of their fellows. We tend to think that
parents who try too forcibly to impose a character, however good a character, on their
children, are less than perfect parents.

[26] The alternative way in which a creator could help creatures to perform right actions is
by sometimes providing additional reasons for creatures to do what is right, reasons which
by their very nature have a strong causal influence. Reasons such as improving the universe
or doing one's duty do not necessarily have a strong causal influence, for as we have seen
creatures may be little influenced by them. Giving a creature reasons which by their nature
were strongly causally influential on a particular occasion on any creature whatever his
character, would not impose a particular character on a creature. It would, however, incline
him to do what is right on that occasion and maybe subsequently too. Now if a reason is by
its nature to be strongly causally influential it must be something of which the agent is
aware which causally inclines him (whatever his character) to perform some action, to bring
about some kind of change. What kind of reason could this be except the existence of an
unpleasant feeling, either a sensation such as a pain or an emotion such as a feeling of loss
or deprivation? Such feelings are things of which agents are conscious, which cause them to
do whatever action will get rid of those feelings, and which provide reason for performing
such action. An itch causally inclines a man to do whatever will cause the itch to cease, e.g.,
scratch, and provides a reason for doing that action. Its causal influence is quite
independent of the agent — saint or sinner, strong-willed or weak-willed, will all be strongly
inclined to get rid of their pains (though some may learn to resist the inclination). Hence a
creator who wished to give agents some inclination to improve the world without giving
them a character, a wide set of general purposes which they naturally pursue, would tie
some of the imperfections of the world to physical or mental evils.

[27] To tie desirable states of affairs to pleasant feelings would not have the same effect.
Only an existing feeling can be causally efficacious. An agent could be moved to action by a
pleasant feeling only when he had it, and the only action to which he could be moved would
be to keep the world as it is, not to improve it. For men to have reasons which move men of
any character to actions of perfecting the world, a creator needs to tie its imperfections to
unpleasant feelings, that is, physical and mental evils.

[28] There is to some considerable extent such tie-up in our universe. Pain normally occurs
when something goes wrong with the working of our body which is going to lead to further
limitation on the purposes which we can achieve; and the pain ends when the body is
repaired. The existence of the pain spurs the sufferer, and others through the sympathetic
suffering which arises when they learn of the sufferer's pain, to do something about the
bodily malfunctioning. Yet giving men such feelings which they are inclined to end involves
the imposition of no character. A man who is inclined to end his toothache by a visit to the
dentist may be saint or sinner, strong-willed or weak-willed, rational or irrational. Any other
way of which I can conceive of giving men an inclination to correct what goes wrong, and
generally to improve the universe, would seem to involve imposing a character. A creator
could, for example, have operated exclusively by threats and promises, whispering in men's
ears, "unless you go to the dentist, you are going to suffer terribly," or "if you go to the
Richard Swinburne The problem of evil Page 10/11

dentist, you are going to feel wonderful." And if the order of nature is God's creation, he
does indeed often provide us with such threats and promises — not by whispering in our
ears but by providing inductive evidence. There is plenty of inductive evidence that
unattended cuts and sores will lead to pain; that eating and drinking will lead to pleasure.
Still, men do not always respond to threats and promises or take the trouble to notice
inductive evidence (e.g., statistics showing the correlation between smoking and cancer). A
creator could have made men so that they naturally took more account of inductive
evidence. But to do so would be to impose character. It would be to make men, apart from
any choice of theirs, rational and strong-willed.

[29] Many mental evils too are caused by things going wrong in a man's life or in the life of
his fellows and often serve as a spur to a man to put things right, either to put right the
cause of the particular mental evil or to put similar things right. A man's feeling of
frustration at the failure of his plans spurs him either to fulfill those plans despite their initial
failure or to curtail his ambitions. A man's sadness at the failure of the plans of his child will
incline him to help the child more in the future. A man's grief at the absence of a loved one
inclines him to do whatever will get the loved one back. As with physical pain, the spur
inclines a man to do what is right but does without imposing a character — without, say,
making a man responsive to duty, or strong-willed.

[30] Physical and mental evils may serve as spurs to long-term cooperative research
leading to improvement of the universe. A feeling of sympathy to the actual and prospective
suffering of many from tuberculosis or cancer leads to acquisition of knowledge and
provision of cure for future sufferers. Cooperative and long-term research and cure is a very
good thing, the kind of thing toward which men need a spur. A man's suffering is never in
vain if it leads through sympathy to the work of others which eventually provides a long-
term cure. True, there could be sympathy without a sufferer for whom the sympathy is felt.
Yet in a world made by a creator, there cannot be sympathy on the large scale without a
sufferer, for whom the sympathy is felt, unless the creator planned for creatures generally
to be deceived about the feelings of their fellows; and that, we have claimed, would be
morally wrong.

[31] So generally many evils have a biological and psychological utility in producing spurs to
right action without imposition of character, a goal which it is hard to conceive of being
realized in any other way. This point provides a reason for the rejection of (P4).

[32] So, I have argued, there seem to be kinds of justification for the evils which exist in
the world, available to the theodicist. Although a good creator might have very different
kinds of justification for producing, or allowing others to produce, various different evils,
there is a central thread running through the kind of theodicy which I have made my
theodicist put forward. This is that it is a good thing that a creator should make a half-
finished universe and create immature creatures, who are humanly free agents, to inhabit
it; and that he should allow them to exercise some choice over what kind of creatures they
are to become and what sort of universe is to be (while at the same time giving them a
slight push in the direction of doing what is right); and that the creatures should have
power to affect not only the development of the inanimate universe but the well-being and
moral character of their fellows, and that there should be opportunities for creatures to
develop noble characters and do especially noble actions. My theodicist has argued that if a
creator is to make a universe of this kind, then evils of various kinds may inevitably — at
any rate temporarily— belong to such a universe; and that it is not a morally bad thing to
create such a universe despite the evils.
Richard Swinburne The problem of evil Page 11/11

The Quantity of Evil

[33] Now a morally sensitive antitheodicist might well in principle accept some of the above
arguments. He may agree that in principle it is not wrong to create humanly free agents,
despite the possible evils which might result, or to create pains as biological warnings. But
where the crunch comes, it seems to me, is in the amount of evil which exists in our world.
The arititheodicist says, all right, it would not be wrong to create men able to harm each
other, but it would be wrong to create men able to put each other in Belsen. It would not be
wrong to create backaches and headaches, even severe ones, as biological warnings, but
not the long severe incurable pain of some diseases. In reply the theodicist must argue that
a creator who allowed men to do little evil would be a creator who gave them little
responsibility; and a creator who gave them only coughs and colds, and not cancer and
cholera would be a creator who treated men as children instead of giving them real
encouragement to subdue the world. The argument must go on with regard to particular
cases. The antitheodicist must sketch in detail and show his adversary the horrors of
particular wars and diseases. The theodicist in reply must sketch in detail and show his
adversary the good which such disasters make possible. He must show to his opponent men
working together for good, men helping each other to overcome disease and famine; the
heroism of men who choose the good in spite of temptation, who help others not merely by
giving them food but who teach them right and wrong, give them something to live for and
something to die for. A world in which this is possible can only be a world in which there is
much evil as well as great good. Interfere to stop the evil and you cut off the good.

[34] The theodicist's God is a god who thinks the higher goods so worthwhile that he is
prepared to ask a lot of man in the way of enduring evil. Creatures determining in
cooperation their own character and future, and that of the universe in which they live,
coming in the process to show charity, forgiveness, faith, and self-sacrifice is such a
worthwhile thing that a creator would not be unjustified in making or permitting a certain
amount of evil in order that they should be realized. No doubt a good creator would put a
limit on the amount of evil in the world and perhaps an end to the struggle with it after a
number of years. But if he allowed creatures to struggle with evil, he would allow them a
real struggle with a real enemy, not a parlor game. The antitheodicist's mistake lies in
extrapolating too quickly from our duties when faced with evil to the duties of a creator,
while ignoring the enormous differences in the circumstances of each. Each of us at one
time can make the existing universe better or worse only in a few particulars. A creator can
choose the kind of universe and the kind of creatures there are to be. It seldom becomes us
in our ignorance and weakness to do anything more than remove the evident evils — war,
disease, and famine. We seldom have the power or the knowledge or the right to use such
evils to forward deeper and longer-term goods. To make an analogy, the duty of the weak
and ignorant is to eliminate cowpox and not spread it, while the doctor has a duty to spread
(under carefully controlled conditions). But a creator who made or permitted his creatures
to suffer much evil and asked them to suffer more is a very demanding creator, one with
high ideals who expects a lot. For myself I can say that I would not be too happy to worship
a creator who expected too little of his creatures. Nevertheless such a God does ask a lot of
creatures. A theodicist is in a better position to defend a theodicy such as I have outlined if
he is prepared also to make the further additional claim — that God knowing the worth-
whileness of the conquest of evil and the perfecting of the universe by men, shared with
them this task by subjecting himself as man to the evil in the world. A creator is more
justified in creating or permitting evils to be overcome by his creatures if he is prepared to
share with them the burden of the suffering and effort.