We have further to determine what motions are contrary to each
other, and to determine similarly how it is with rest. And we have
first to decide whether contrary motions are motions respectively from
and to the same thing, e.g. a motion from health and a motion to
health (where the opposition, it would seem, is of the same kind as
that between coming to be and ceasing to be); or motions
respectively from contraries, e.g. a motion from health and a motion
from disease; or motions respectively to contraries, e.g. a motion
to health and a motion to disease; or motions respectively from a
contrary and to the opposite contrary, e.g. a motion from health and a
motion to disease; or motions respectively from a contrary to the
opposite contrary and from the latter to the former, e.g. a motion
from health to disease and a motion from disease to health: for
motions must be contrary to one another in one or more of these
ways, as there is no other way in which they can be opposed.

    Now motions respectively from a contrary and to the opposite
contrary, e.g. a motion from health and a motion to disease, are not
contrary motions: for they are one and the same. (Yet their essence is
not the same, just as changing from health is different from
changing to disease.) Nor are motion respectively from a contrary
and from the opposite contrary contrary motions, for a motion from a
contrary is at the same time a motion to a contrary or to an
intermediate (of this, however, we shall speak later), but changing to
a contrary rather than changing from a contrary would seem to be the
cause of the contrariety of motions, the latter being the loss, the
former the gain, of contrariness. Moreover, each several motion
takes its name rather from the goal than from the starting-point of
change, e.g. motion to health we call convalescence, motion to disease
sickening. Thus we are left with motions respectively to contraries,
and motions respectively to contraries from the opposite contraries.
Now it would seem that motions to contraries are at the same time
motions from contraries (though their essence may not be the same; "to
health" is distinct, I mean, from "from disease", and "from health"
from "to disease").

    Since then change differs from motion (motion being change from a
particular subject to a particular subject), it follows that
contrary motions are motions respectively from a contrary to the
opposite contrary and from the latter to the former, e.g. a motion
from health to disease and a motion from disease to health.
Moreover, the consideration of particular examples will also show what
kinds of processes are generally recognized as contrary: thus
falling ill is regarded as contrary to recovering one's health,
these processes having contrary goals, and being taught as contrary to
being led into error by another, it being possible to acquire error,
like knowledge, either by one's own agency or by that of another.
Similarly we have upward locomotion and downward locomotion, which are
contrary lengthwise, locomotion to the right and locomotion to the
left, which are contrary breadthwise, and forward locomotion and
backward locomotion, which too are contraries. On the other hand, a
process simply to a contrary, e.g. that denoted by the expression
"becoming white", where no starting-point is specified, is a change
but not a motion. And in all cases of a thing that has no contrary
we have as contraries change from and change to the same thing. Thus
coming to be is contrary to ceasing to be, and losing to gaining.
But these are changes and not motions. And wherever a pair of
contraries admit of an intermediate, motions to that intermediate must
be held to be in a sense motions to one or other of the contraries:
for the intermediate serves as a contrary for the purposes of the
motion, in whichever direction the change may be, e.g. grey in a
motion from grey to white takes the place of black as
starting-point, in a motion from white to grey it takes the place of
black as goal, and in a motion from black to grey it takes the place
of white as goal: for the middle is opposed in a sense to either of
the extremes, as has been said above. Thus we see that two motions are
contrary to each other only when one is a motion from a contrary to
the opposite contrary and the other is a motion from the latter to the
former.