|
Objection 1: It would seem that habit does not imply order to an
act. For everything acts according as it is in act. But the
Philosopher says (De Anima iii, text 8), that "when one is
become knowing by habit, one is still in a state of potentiality, but
otherwise than before learning." Therefore habit does not imply the
relation of a principle to an act.
Objection 2: Further, that which is put in the definition of a
thing, belongs to it essentially. But to be a principle of action,
is put in the definition of power, as we read in Metaph. v, text.
17. Therefore to be the principle of an act belongs to power
essentially. Now that which is essential is first in every genus. If
therefore, habit also is a principle of act, it follows that it is
posterior to power. And so habit and disposition will not be the first
species of quality.
Objection 3: Further, health is sometimes a habit, and so are
leanness and beauty. But these do not indicate relation to an act.
Therefore it is not essential to habit to be a principle of act.
On the contrary, Augustine says (De Bono Conjug. xxi) that
"habit is that whereby something is done when necessary." And the
Commentator says (De Anima iii) that "habit is that whereby we act
when we will."
I answer that, To have relation to an act may belong to habit, both
in regard to the nature of habit, and in regard to the subject in which
the habit is. In regard to the nature of habit, it belongs to every
habit to have relation to an act. For it is essential to habit to
imply some relation to a thing's nature, in so far as it is suitable
or unsuitable thereto. But a thing's nature, which is the end of
generation, is further ordained to another end, which is either an
operation, or the product of an operation, to which one attains by
means of operation. Wherefore habit implies relation not only to the
very nature of a thing, but also, consequently, to operation,
inasmuch as this is the end of nature, or conducive to the end.
Whence also it is stated (Metaph. v, text. 25) in the
definition of habit, that it is a disposition whereby that which is
disposed, is well or ill disposed either in regard to itself, that is
to its nature, or in regard to something else, that is to the end.
But there are some habits, which even on the part of the subject in
which they are, imply primarily and principally relation to an act.
For, as we have said, habit primarily and of itself implies a
relation to the thing's nature. If therefore the nature of a thing,
in which the habit is, consists in this very relation to an act, it
follows that the habit principally implies relation to an act. Now it
is clear that the nature and the notion of power is that it should be a
principle of act. Wherefore every habit is subjected in a power,
implies principally relation to an act.
Reply to Objection 1: Habit is an act, in so far as it is a
quality: and in this respect it can be a principle of operation. It
is, however, in a state of potentiality in respect to operation.
Wherefore habit is called first act, and operation, second act; as
it is explained in De Anima ii, text. 5.
Reply to Objection 2: It is not the essence of habit to be related
to power, but to be related to nature. And as nature precedes
action, to which power is related, therefore habit is put before power
as a species of quality.
Reply to Objection 3: Health is said to be a habit, or a habitual
disposition, in relation to nature, as stated above. But in so far
as nature is a principle of act, it consequently implies a relation to
act. Wherefore the Philosopher says (De Hist. Animal. x,
1), that man, or one of his members, is called healthy, "when he
can perform the operation of a healthy man." And the same applies to
other habits.
|
|