I have been musing upon time and temporality, particularly the proto-theory adopted by my book manuscript. Below are some straight-forward thoughts about time assuming a broadly pragmatist perspective. Those unfamiliar with that perspective might find the thoughts anything but straight-forward, since they presume basic premises that are radically at odds with mainstream philosophy in America. Yay, Peirce!
Time is change by definition.
Time is synonymous with, but not identical to, change. "Change" denotes a difference in what exists, and connotes a local change in a determinate existence. If a thing can change, it is temporal, which is synonymous with existing and vice-versa (assuming the real vs. existing distinction). If a thing can not change, then it is atemporal and does not exist. It may still be real, as in the case of numbers or generals/universals (assuming neo-scholastic realism).
Time as such does not exist.
If time is change, then time exists only insomuch as change exists, which further depends upon some other thing in flux.
Is time the same thing or identical to change?
In other words, why is time synonymous with but not identical to change? In a commonsense meaning, "change" requires an alteration in existence, e.g., in predication logically-speaking, whereas there is more to time than actuality or change in actuality. Time also denotes the structure of the possible as well as the actual, and this point qualifies or nuances the prior definition.
Time is change insomuch as change is an alteration in the structure of possibility.
Any change is time, since a difference in the actual connotes a difference in the possible, but the reverse is not necessarily true. There can be an alteration in possibility without an alteration in actuality.
Time is relative.
Since time depends on actual change, it is relative. Since a temporal event is relative to existence, then discussions of possibility must include both the logical and probabilistic senses of "possibility." Hence, an alteration in the "structure of possibility" should be conceived in terms of probability functions or some such, which are not binary true/false functions, but n-dimensional.
Time is local.
Since time depends on a change in existence, and existence is always spatial, then time is relative to a locality.
Time is asymmetric.
Time flows in one way: there is a strict order of events. By implication, causality is neither symmetric nor necessarily reversible. That is, if time is asymmetric, then change is so, and since causality is a kind of change, then causality must be asymmetric. Hence, there may be cosmic epochs that might never repeat.
Since time is relative, local, asymmetric, etc. we should distinguish between the ontological and ontic.
The ontology of time, or descriptions of the structure of possibility as such given a logos of nature, should be distinguished from the onticology of time, or descriptions of the structure of real possibility given a physics of nature. The latter is dependent upon the former.
Time is continuous and teleological.
I have already assumed the continuity of time, and I refer readers to my links and comments on Peirce's arguments. As for "teological," I just take that to be an implication of the combined realities of chance and law: if there is change, then it must be in some degree determinate, and since time is asymmetric, then we can describe this asymmetric determinate change in terms of what might be given this development, i.e., in terms of a non-deterministic emergent teleology.
Why not use "possible worlds" talk to describe this?
Possible worlds talk was not constructed to support my basic ontological principles, and it is not worth the time to attempt to translate it. Consider one immediate obstacle. Typical "possible worlds" is not temporal: discussion of the worlds is static or atemporal. One obvious way to temporalize such discussion would be to provide a temporal index, e.g., World P at time t1 or World Q at time t2. However, the indexical concept of time quantifies and universalizes time. In contrast, I am beginning from a process metaphysics where time as such is indeterminate because it is relative. Moreover, since time is continuous, it cannot be unequivocally indexed or universally measured in any meaningful way befitting ontology. Ontology and metaphysics, unlike science, cannot tolerate measurement imprecision. Relative indices such as "sooner" or "later could obviously be managed, but that is not very productive. Moreover, the act of indexing time, e.g., speaking of "at time t1," implies that a universal signification that cannot be had. That is, indexing implies that it would be meaningful to talk about World P and World Q at time t1, but that is not meaningful and therefore it would be best to avoid that terminology.
Trust me, this is a lot slower and clearer than those 100 pages of my dissertation, let along my revised manuscript.