THIRD ARTICLE [I-II, Q. 111, Art. 3]

Whether Grace Is Fittingly Divided into Prevenient and Subsequent

Grace?

Objection 1: It would seem that grace is not fittingly divided into

prevenient and subsequent. For grace is an effect of the Divine love.

But God's love is never subsequent, but always prevenient, according

to 1 John 4:10: "Not as though we had loved God, but because He hath

first loved us." Therefore grace ought not to be divided into

prevenient and subsequent.

Obj. 2: Further, there is but one sanctifying grace in man, since it

is sufficient, according to 2 Cor. 12:9: "My grace is sufficient for

thee." But the same thing cannot be before and after. Therefore grace

is not fittingly divided into prevenient and subsequent.

Obj. 3: Further, grace is known by its effects. Now there are an

infinite number of effects--one preceding another. Hence if with

regard to these, grace must be divided into prevenient and

subsequent, it would seem that there are infinite species of grace.

Now no art takes note of the infinite in number. Hence grace is not

fittingly divided into prevenient and subsequent.

_On the contrary,_ God's grace is the outcome of His mercy. Now both

are said in Ps. 58:11: "His mercy shall prevent me," and again, Ps.

22:6: "Thy mercy will follow me." Therefore grace is fittingly

divided into prevenient and subsequent.

_I answer that,_ As grace is divided into operating and cooperating,

with regard to its diverse effects, so also is it divided into

prevenient and subsequent, howsoever we consider grace. Now there are

five effects of grace in us: of these, the first is, to heal the

soul; the second, to desire good; the third, to carry into effect the

good proposed; the fourth, to persevere in good; the fifth, to reach

glory. And hence grace, inasmuch as it causes the first effect in us,

is called prevenient with respect to the second, and inasmuch as it

causes the second, it is called subsequent with respect to the first

effect. And as one effect is posterior to this effect, and prior to

that, so may grace be called prevenient and subsequent on account of

the same effect viewed relatively to divers others. And this is what

Augustine says (De Natura et Gratia xxxi): "It is prevenient,

inasmuch as it heals, and subsequent, inasmuch as, being healed, we

are strengthened; it is prevenient, inasmuch as we are called, and

subsequent, inasmuch as we are glorified."

Reply Obj. 1: God's love signifies something eternal; and hence can

never be called anything but prevenient. But grace signifies a

temporal effect, which can precede and follow another; and thus grace

may be both prevenient and subsequent.

Reply Obj. 2: The division into prevenient and subsequent grace does

not divide grace in its essence, but only in its effects, as was

already said of operating and cooperating grace. For subsequent

grace, inasmuch as it pertains to glory, is not numerically distinct

from prevenient grace whereby we are at present justified. For even

as the charity of the earth is not voided in heaven, so must the same

be said of the light of grace, since the notion of neither implies

imperfection.

Reply Obj. 3: Although the effects of grace may be infinite in

number, even as human acts are infinite, nevertheless all are reduced

to some of a determinate species, and moreover all coincide in

this--that one precedes another.

________________________

Summa Theologica, Part I-II (Pars Prima Secundae) From the Complete American Edition · 第 716 / 746 题