Owen—Relations Expressed by the Passive Voice. Ill 
the active, putting—as the passive does in fact—its first term 
(subject, the actee) in the nominative, and—as it does not in 
fact—its last term (agent, actor) in the accusative—again with¬ 
out a preposition. 
But evolution of this sort was barred by the tenacity with 
which the accusative was held to indicate what suffers action— 
not what is last term in thought. This old conception of accu¬ 
sative function was continued naturally and without disaster 
even when the verb expressed an action altogether figurative, 
as in “Sight requires” or “demands an eye to see with,” which 
I sense as merely figurative paraphrasing of “No seeing can 
occur without an eye” or “Eyes are indispensable to seeing.” 
But the old conception was continued further quite unnaturally 
and disastrously. Actual and figurative actions so beset the 
mind that they produce a habit which we follow in their ab¬ 
sence. When the verb, although it names no action, is at¬ 
tended by a first term and a last, we often still proceed as if 
there were an action— e. g. in “A equals B” (Conf. exceeds, 
outweighs, overlies, antedates, etc.,) in which a well inflected 
language would exhibit B as object of the verb. 
But in “equaling” there plainly is no genuine action. Also 
any spurious action-—any unreal “affecting”—any “suffering 
the action”—is rather taken for granted than distinctly even 
imagined. Habit seems to be what leads to the continued use 
of the accusative as namer of what suffers. As habit strengthens, 
any possibility of sensing the accusative as used or usable for 
any other purpose weakens; and when action ceases to be even 
taken for granted, the possibility of using the accusative inflec¬ 
tion as a sign of last-term membership in thought is practi¬ 
cally certain to be overlooked. 
The cessation comes abruptly—when for instance bulk-equiva¬ 
lence—or, say, a single, incomplete identity of e. g. A and B— 
is superseded in the mind by multiple, complete identity. If 
this last identity comprises, as it might, a score of single ones, 
whatever rationality there may have been in using the accusa¬ 
tive with one identity would seem to be increased not less than 
twenty-fold; and yet in “A is B” which indicates complete iden- 
