Thomas Aquinas, Commentary on Aristotle's Physics

First he states his intention and the order in which one must proceed. Secondly he follows this out, where he says, 'To start, then . . .' (217 b 33).

He says, therefore, first that the treatment of time follows upon what has gone before. By this he points out the problem under consideration. For as with the foregoing, likewise with time, it is first necessary to proceed by proposing extraneous arguments, that is, sophistical arguments, or arguments given by others, as to whether time exists or not, and if it exists, what its nature is.

Next where he says, 'To start, then . . .' (217 b 33), he begins to treat time, first by raising objections, and secondly by determining the truth, where he says, 'We must take this . . .' (219 a 2).

Concerning the first part he makes two points. By raising objections he inquires first whether time is, and secondly, what time is, where he says, 'As to what time is . . .' (218 a 32).

Concerning the first part he makes two points. First, he gives two arguments to show that time does not exist. Secondly, where he says, 'Again, the "now" . . .' (218 a 9), he inquires whether in the whole of time there is one or many 'nows'.

559. He says, therefore, first that from the following two arguments one can conceive either that time does not exist at all, or that it is something which can be perceived only obscurely and with difficulty.

The first argument is as follows. Anything that is composed of things which do not exist cannot itself exist or have any substance. But time is composed of things which do not exist. For one part of time is the past which does not now exist, and the other part is the future which does not yet exist. And from these two things the whole of time, given as infinite and perpetual, is composed. Therefore it is impossible that time is something.

560. He gives the second argument where he says, 'Further, if a divisible thing . . .' (218 a 4). The argument is as follows.

In regard to each divisible thing it is necessary, while it exists, that one or more of its parts exist. But this is not true of time. For certain parts of time are now past, and other parts are in the future, and no divisible part of time exists in act. The 'now', which exists in act, is not a part of time. For a part is either that which measures a whole, as two is a part of six, or at least that from which the whole is composed, as four is a part of six, not as measuring it, but because six is composed of four and two. However, time is not composed of 'nows', as will be proven below. Therefore, time is not something.

561. Next where he says, 'Again, the "now" . . .' (218 a 9), he inquires whether the same 'now' is in all time.

Concerning this he makes three points. First he raises the question. Secondly, where he says, 'If it is always different . . .' (218 a 11), he objects to one part of the question. Thirdly, where he says, 'Yes, but neither is it possible . . .' (218 a 21), he objects to the other part of the question.

He says, therefore, first that it is not easy to know whether the 'now', which seems to distinguish past from future, always remains the same in the whole of time, or whether it is different.

562. Next where he says, 'If it is always different . . .' (218 a 11), he shows with the following argument that the 'now' is not different.

Two different parts of time cannot exist simultaneously unless one contains the other. Thus a greater time contains a smaller time. For example, a year contains a month, and a month contains a day. (For a day and a month and a year exist simultaneously.) But one 'now', since it is indivisible, does not contain another. Therefore, if we take two 'nows' in time, it is necessary that the 'now', which formerly was and in a way is not, at some time was corrupted, and that two 'nows' never exist simultaneously. But everything that is corrupted must be corrupted in some 'now'. However, it cannot be said that that former 'now' was corrupted in that former 'now' itself. For then that 'now' was, and nothing is corrupted while it exists. Likewise, it cannot be said that that former 'now' was corrupted in a later 'now'. For it is impossible for two 'nows' to be so related to each other that they are immediately consecutive, just as this is also impossible regarding two points. This is assumed here, but will be proven in Book VI. Therefore between any two 'nows' there is an infinity of 'nows'. If, therefore, a former 'now' is corrupted in some later 'now', then it follows that that 'now' which is first exists simultaneously with all the intermediate 'nows'. This is impossible, as was said. Therefore it is impossible that the 'now' be different.

563. Next where he says, 'Yes, but neither is it possible . . .' (218 a 21), he shows with two arguments that the 'now' cannot be one and the same.

The first argument is as follows. There cannot be only one terminus of a finite, divisible thing, whether it be a continuum in only one dimension, as a line, or in many dimensions, as a surface and a body. For two points are the termini of one finite line, and many lines of a surface, and many surfaces of a body. But the 'now' is the terminus of time. Therefore, when we take some finite time, there must be many 'nows'.

564. He gives the second argument where he says, 'Further, if coincidence . . .' (218 a 25). The argument is as follows. Those things are said to be simultaneous, and neither before nor after, which exist in the same 'now'. Therefore, if the same 'now' remains in the whole of time, it follows that things which existed a thousand years ago are simultaneous with things which exist today.

Lastly he concludes that such are the objections concerning the 'nows' of time.