Objection 1. It would seem that there is nothing voluntary in human acts. For that is voluntary "which has its principle within itself." as Gregory of Nyssa [Nemesius, De Natura Hom. xxxii.], Damascene (De Fide Orth. ii, 24), and Aristotle (Ethic. iii, 1) declare. But the principle of human acts is not in man himself, but outside him: since man's appetite is moved to act, by the appetible object which is outside him, and is as a "mover unmoved" (De Anima iii, 10). Therefore there is nothing voluntary in human acts.
Objection 2. Further, the Philosopher (Phys. viii, 2) proves that in animals no new movement arises that is not preceded by a motion from without. But all human acts are new, since none is eternal. Consequently, the principle of all human acts is from without: and therefore there is nothing voluntary in them.
Objection 3. Further, he that acts voluntarily, can act of himself. But this is not true of man; for it is written (John 15:5): "Without Me you can do nothing." Therefore there is nothing voluntary in human acts.
On the contrary, Damascene says (De Fide Orth. ii) that "the voluntary is an act consisting in a rational operation." Now such are human acts. Therefore there is something voluntary in human acts.
I answer that, There must needs be something voluntary in human acts. In order to make this clear, we must take note that the principle of some acts or movements is within the agent, or that which is moved; whereas the principle of some movements or acts is outside. For when a stone is moved upwards, the principle of this movement is outside the stone: whereas when it is moved downwards, the principle of this movement is in the stone. Now of those things that are moved by an intrinsic principle, some move themselves, some not. For since every agent or thing moved, acts or is moved for an end, as stated above (Question 1, Article 2); those are perfectly moved by an intrinsic principle, whose intrinsic principle is one not only of movement but of movement for an end. Now in order for a thing to be done for an end, some knowledge of the end is necessary. Therefore, whatever so acts or is moved by an intrinsic principle, that it has some knowledge of the end, has within itself the principle of its act, so that it not only acts, but acts for an end. On the other hand, if a thing has no knowledge of the end, even though it have an intrinsic principle of action or movement, nevertheless the principle of acting or being moved for an end is not in that thing, but in something else, by which the principle of its action towards an end is not in that thing, but in something else, by which the principle of its action towards an end is imprinted on it. Wherefore such like things are not said to move themselves, but to be moved by others. But those things which have a knowledge of the end are said to move themselves because there is in them a principle by which they not only act but also act for an end. And consequently, since both are from an intrinsic principle, to wit, that they act and that they act for an end, the movements of such things are said to be voluntary: for the word "voluntary" implies that their movements and acts are from their own inclination. Hence it is that, according to the definitions of Aristotle, Gregory of Nyssa, and Damascene [See Objection 1, the voluntary is defined not only as having "a principle within" the agent, but also as implying "knowledge." Therefore, since man especially knows the end of his work, and moves himself, in his acts especially is the voluntary to be found.
Reply to Objection 1. Not every principle is a first principle. Therefore, although it is essential to the voluntary act that its principle be within the agent, nevertheless it is not contrary to the nature of the voluntary act that this intrinsic principle be caused or moved by an extrinsic principle: because it is not essential to the voluntary act that its intrinsic principle be a first principle. Yet again it must be observed that a principle of movement may happen to be first in a genus, but not first simply: thus in the genus of things subject to alteration, the first principle of alteration is a heavenly body, which is nevertheless, is not the first mover simply, but is moved locally by a higher mover. And so the intrinsic principle of the voluntary act, i.e. the cognitive and appetitive power, is the first principle in the genus of appetitive movement, although it is moved by an extrinsic principle according to other species of movement.
Reply to Objection 2. New movements in animals are indeed preceded by a motion from without; and this in two respects. First, in so far as by means of an extrinsic motion an animal's senses are confronted with something sensible, which, on being apprehended, moves the appetite. Thus a lion, on seeing a stag in movement and coming towards him, begins to be moved towards the stag. Secondly, in so far as some extrinsic motion produces a physical change in an animal's body, as in the case of cold or heat; and through the body being affected by the motion of an outward body, the sensitive appetite which is the power of a bodily organ, is also moved indirectly; thus it happens that through some alteration in the body the appetite is roused to the desire of something. But this is not contrary to the nature of voluntariness, as stated above (ad 1), for such movements caused by an extrinsic principle are of another genus of movement.
Reply to Objection 3. God moves man to act, not only by proposing the appetible to the senses, or by effecting a change in his body, but also by moving the will itself; because every movement either of the will or of nature, proceeds from God as the First Mover. And just as it is not incompatible with nature that the natural movement be from God as the First Mover, inasmuch as nature is an instrument of God moving it: so it is not contrary to the essence of a voluntary act, that it proceed from God, inasmuch as the will is moved by God. Nevertheless both natural and voluntary movements have this in common, that it is essential that they should proceed from a principle within the agent.
Source: Summa Theologica (New Advent)