FIRST ARTICLE [I, Q. 49, Art. 1]

Whether Good Can Be the Cause of Evil?

Objection 1: It would seem that good cannot be the cause of evil. For

it is said (Matt. 7:18): "A good tree cannot bring forth evil fruit."

Obj. 2: Further, one contrary cannot be the cause of another. But

evil is the contrary to good. Therefore good cannot be the cause of

evil.

Obj. 3: Further, a deficient effect can proceed only from a deficient

cause. But evil is a deficient effect. Therefore its cause, if it has

one, is deficient. But everything deficient is an evil. Therefore the

cause of evil can only be evil.

Obj. 4: Further, Dionysius says (Div. Nom. iv) that evil has no

cause. Therefore good is not the cause of evil.

_On the contrary,_ Augustine says (Contra Julian. i, 9): "There is no

possible source of evil except good."

_I answer that,_ It must be said that every evil in some way has a

cause. For evil is the absence of the good, which is natural and due

to a thing. But that anything fail from its natural and due

disposition can come only from some cause drawing it out of its proper

disposition. For a heavy thing is not moved upwards except by some

impelling force; nor does an agent fail in its action except from some

impediment. But only good can be a cause; because nothing can be a

cause except inasmuch as it is a being, and every being, as such, is

good.

And if we consider the special kinds of causes, we see that the

agent, the form, and the end, import some kind of perfection which

belongs to the notion of good. Even matter, as a potentiality to

good, has the nature of good. Now that good is the cause of evil by

way of the material cause was shown above (Q. 48, A. 3). For it was

shown that good is the subject of evil. But evil has no formal cause,

rather is it a privation of form; likewise, neither has it a final

cause, but rather is it a privation of order to the proper end; since

not only the end has the nature of good, but also the useful, which

is ordered to the end. Evil, however, has a cause by way of an agent,

not directly, but accidentally.

In proof of this, we must know that evil is caused in the action

otherwise than in the effect. In the action evil is caused by reason

of the defect of some principle of action, either of the principal or

the instrumental agent; thus the defect in the movement of an animal

may happen by reason of the weakness of the motive power, as in the

case of children, or by reason only of the ineptitude of the

instrument, as in the lame. On the other hand, evil is caused in a

thing, but not in the proper effect of the agent, sometimes by the

power of the agent, sometimes by reason of a defect, either of the

agent or of the matter. It is caused by reason of the power or

perfection of the agent when there necessarily follows on the form

intended by the agent the privation of another form; as, for instance,

when on the form of fire there follows the privation of the form of

air or of water. Therefore, as the more perfect the fire is in

strength, so much the more perfectly does it impress its own form, so

also the more perfectly does it corrupt the contrary. Hence that evil

and corruption befall air and water comes from the perfection of the

fire: but this is accidental; because fire does not aim at the

privation of the form of water, but at the bringing in of its own

form, though by doing this it also accidentally causes the other.

But if there is a defect in the proper effect of the fire--as, for

instance, that it fails to heat--this comes either by defect of the

action, which implies the defect of some principle, as was said above,

or by the indisposition of the matter, which does not receive the

action of the fire, the agent. But this very fact that it is a

deficient being is accidental to good to which of itself it belongs to

act. Hence it is true that evil in no way has any but an accidental

cause; and thus is good the cause of evil.

Reply Obj. 1: As Augustine says (Contra Julian. i): "The Lord calls

an evil will the evil tree, and a good will a good tree." Now, a good

will does not produce a morally bad act, since it is from the good

will itself that a moral act is judged to be good. Nevertheless the

movement itself of an evil will is caused by the rational creature,

which is good; and thus good is the cause of evil.

Reply Obj. 2: Good does not cause that evil which is contrary to

itself, but some other evil: thus the goodness of the fire causes

evil to the water, and man, good as to his nature, causes an act

morally evil. And, as explained above (Q. 19, A. 9), this is by

accident. Moreover, it does happen sometimes that one contrary causes

another by accident: for instance, the exterior surrounding cold

heats (the body) through the concentration of the inward heat.

Reply Obj. 3: Evil has a deficient cause in voluntary things

otherwise than in natural things. For the natural agent produces the

same kind of effect as it is itself, unless it is impeded by some

exterior thing; and this amounts to some defect belonging to it.

Hence evil never follows in the effect, unless some other evil

pre-exists in the agent or in the matter, as was said above. But in

voluntary things the defect of the action comes from the will

actually deficient, inasmuch as it does not actually subject itself

to its proper rule. This defect, however, is not a fault, but fault

follows upon it from the fact that the will acts with this defect.

Reply Obj. 4: Evil has no direct cause, but only an accidental cause,

as was said above.

_______________________

Summa Theologica, Part I (Prima Pars) From the Complete American Edition · 第 327 / 722 题