or the now-not-yet; the at-the-time is the now-no-longer or the no-longer-now. The then and the at-the-time both have a now-character, a now-reference. In one place Aristotle says quite concisely, without carrying out the analysis in this detailed way—but without which his whole interpretation of time is unintelligible—τῷ φερομένῳ ἀκολονθεῖ τὸ νῦν,²³ the now follows the moving thing, the object making the transition from one place to another; that is to say, the now is seen concomitantly in experiencing the motion. And to say that it is concomitantly seen means for Aristotle, in the broader sense, that it is concomitantly counted. What is thus concomitantly counted in following a motion, what is thus said, the nows—this is time. ᾟ δ᾿ ἀριθμητὸν τὸ πρότερον καὶ ὕστερον, τὸ νῦν ἔστιν.²⁴ As counted, the nows themselves count—they count the places, so far as these are traversed as places of the motion. Time as ἀριθμὸς phoras is the counted that counts. Aristotle's interpretation of time matches the phenomenon extremely well when he says that time is something counted connected with motion so far as I see this motion in the horizon ἐκ τινος εἰς τι, "from something to something."

In one place Aristotle says about πρότερον and ὕστερον: τὸ δὴ πρότερον καὶ ὕστερον ἐν τόπῳ πρῶτόν ἐστιν;²⁵ it is first of all in place, in the change and sequence of places. He is thinking of before and after here as still wholly without any time-determinateness. The Aristotelian definition of time can also be formulated at first in this way: time is what is counted in connection with motion which is experienced with respect to before and after. But what is thus counted is unveiled as the nows. The nows themselves, however, can be expressed and understood only in the horizon of earlier and later. The "with respect to the before and after" and the "in the horizon of the earlier and later" do not coincide; the second is the interpretation of the first.²⁶ If we take the πρότερον and ὕστερον provisionally as before and after, previous and subsequent, the genesis of Aristotle's definition of time becomes clearer. If we take it straight away as earlier and later, then the definition seems absurd at first, but this only indicates that a central problem is still present in it: the question about the origin of the now itself. The first translation gives the literal conception, but the second already includes a large element of interpretation.

We intentionally translated the Aristotelian definition of time as something counted in connection with motion so far as this motion is seen in the horizon of earlier and later. We have already taken the πρότερον-ὕστερον in

23. Cf. Physica, 5, 219b 22; see also 220a 6

24. Ibid, 219b 25.

25. Ibid, 219a 14f.

26. Cf. Sein und Zeit, pp. 420 ff.