Are you claiming the future defines the past? — Devans99

No, the question-begging claim is that time as an infinite succession of moments is impossible because it would have no first element.

If you had thought it through, you would appreciate that the impossibility of perpetual motion implies a start of motion. — Devans99

Which does not entail a start of time.

No, the question-begging claim is that time as an infinite succession of moments is impossible because it would have no first element. — aletheist

So we are talking about an infinite causal regress right? You'd agree they have no first moment? If we take an example of a finite causal regress:

1. The cue hits the white ball
2. The white ball hits the black ball
3. The black ball goes in the pocket

Note that if we remove the first element of the finite causal regress ([1] above) then the rest of the regress disappears.

Infinite causal regress have by definition no first element; so they do not exist / are not possible.

Which does not entail a start of time. — aletheist

What caused the start of motion? Call it A, what caused A, call it B, what caused B, call it C. So we are in an infinite causal regress. The only way out of such is to posit something uncaused as the base of the regress; IE something from beyond causality; IE something from beyond time. IE time has a start.

If we take an example of a finite causal regress:

1. The cue hits the white ball
2. The white ball hits the black ball
3. The black ball goes in the pocket

Note that if we remove the first element of the finite causal regress ([1] above) then the rest of the regress disappears. — Devans99

Where we mark the "start" of any finite series of events, and how we parse it out into discrete steps, is completely arbitrary. Before #1, presumably a person deliberately pushes the cue such that it hits the white ball. That involves a multitude of mental decisions, nerve signals, and muscle movements. We would have to go much farther back in time before that in order to account for all the causal factors. Moreover, none of the three listed events is instantaneous--each requires a finite interval of time, during which complex physical interactions occur--and in between, each ball presumably rolls across a frictional surface, slightly slowing its velocity.

What caused the start of motion? — Devans99

Again, not relevant; the issue is whether time has a start, not whether motion has a start.

So we are in an infinite causal regress. The only way out of such is to posit something uncaused as the base of the regress; IE something from beyond causality; IE something from beyond time. IE time has a start. — Devans99

Again, non sequitur; even if something outside of time created time (as we both apparently believe), that would not by itself entail that time had a start.

I think if we consider any system over a finite period of time; then clearly it has an initial state that defines all subsequent states and the finite period of time has an initial starting moment.

So it is but a small step to see that any system over an 'infinite' period of time has no initial moment or state and therefore all subsequent states are undefined.

Again, non sequitur; even if something outside of time created time (as we both apparently believe), that would not by itself entail that time had a start. — aletheist

Everything has a start. It does not matter what topology you think time has; it has a start. Name a topology for time that has no start? Circles have start points BTW.

One more argument for you:

1. Assume time has no start
2. The state of the universe is given by the precise positions and velocity vectors of all its particles (10^80 or so in the observable universe I read)
3. Call the current state of the universe X
4. How many times has the universe been in state X in the past?
5. A greater than any number of times *
6. Reductio ad absurdum. [1] is wrong. Time has a start.

* Which is impossible all by itself; you cannot use successive addition to arrive at a number greater than any number

So it is but a small step to see that any system over an 'infinite' period of time has no initial moment or state and therefore all subsequent states are undefined. — Devans99

Non sequitur; having no initial moment/state does not entail having no "defined" moments/states (whatever that means), unless we add the question-begging premiss that a first moment/state is required to "define" any other moments/states.

Everything has a start. — Devans99

More question-begging.

Name a topology for time that has no start? — Devans99

A straight line extending from the infinite past to the infinite future. A hyperbola for which the initial and final moments/states are ideal limits that never actually occur.

Circles have start points BTW. — Devans99

Only if we arbitrarily designate one; a circle in itself has no points of any kind. If I use an inked stamp, I can "create" an entire circle on a piece of paper all at once, with no start point.

The state of the universe is given by the precise positions and velocity vectors of all its particles (10^80 or so in the observable universe I read) — Devans99

Instantaneous states, positions, and velocity vectors are all abstractions that we artificially create to describe reality. They are not themselves real. Besides, our best current science indicates that it is impossible to determine both the position and the velocity of any particle at the same hypothetical instant, let alone all the particles in the universe.

How many times has the universe been in state X in the past? — Devans99

Zero. Besides wrongly treating an instantaneous state as a reality, the latest argumentation wrongly presupposes that the universe can be in the same state more than once.

unless we add the question-begging premiss that a first moment/state is required to "define" any other moments/states. — aletheist

It is not question begging it is just the way reality works:

- Does today define what happens tomorrow?
- Or does tomorrow define what happens today?

Zero. Besides wrongly treating an instantaneous state as a reality, the latest argumentation wrongly presupposes that the universe can be in the same state more than once. — aletheist

- The probability of being in state X must be greater than 0% (because we have been in that state)
- Leading to the number of times in state X as ∞ * non-zero = ∞ times

It is not question begging it is just the way reality works: — Devans99

You cannot prove that time has a start by assuming that time has a start. Besides, if every moment has a preceding moment, then time cannot have a start, because that would require a (first) moment that does not have a preceding moment.

- The probability of being in state X must be greater than 0% (because we have been in that state)
- Leading to the number of times in state X as ∞ * non-zero = ∞ times — Devans99

That is not how probability works in the mathematics of infinity.

You cannot prove that time has a start by assuming that time has a start. — aletheist

Where exactly am I assuming that time has a start?

Do you believe that a greater than any finite number of days has passed?

Besides, if every moment has a preceding moment, then time cannot have a start, because that would require a (first) moment that does not have a preceding moment. — aletheist

Every moment has another moment before it and there is a start of time if time is a circle.

That is not how probability works in the mathematics of infinity. — aletheist

How does it work?

Where exactly am I assuming that time has a start? — Devans99

By insisting that moments/states are "undefined" otherwise.

Do you believe that a greater than any finite number of days has passed? — Devans99

Time is not composed of days. A day is an arbitrary unit of duration that we use to mark and measure the passage of time.

Every moment has another moment before it and there is a start of time if time is a circle. — Devans99

If time is a circle, then every moment has another moment before it, but there is no start of time unless we arbitrarily designate one. Remember, I can use an ink stamp to "create" an entire circle on a piece of paper all at once.

That is not how probability works in the mathematics of infinity. — aletheist

How does it work? — Devans99

Again, in reality there are no instantaneous states, so the "probability" of any such state occurring is meaningless. Besides, in infinite time there would be infinitely many such states, and no reason in principle to assume that any two of them are identical.

By insisting that moments/states are "undefined" otherwise. — aletheist

But time is indubitably linear in nature; earlier times precede and define later times. The argument I am using works for any infinite regress where earlier elements define later elements:

1. An infinite causal regress has no first element
2. If it has no nth element, it has no nth+1 element
3. Therefore no infinite causal regresses exist

Infinite past time would be an infinite causal regress; earlier periods define later periods. So it is just not possible.

Time is not composed of days. A day is an arbitrary unit of duration that we use to mark and measure the passage of time. — aletheist

It does not matter what unit of time we use. Same question: do you think that a greater than any finite number of Planck time units has passed?

Obviously the point here is that time passing is a sequential process and there is no way for a sequential process to ever construct actual infinity. Hence time must be finite.

If time is a circle, then every moment has another moment before it, but there is no start of time unless we arbitrarily designate one. Remember, I can use an ink stamp to "create" an entire circle on a piece of paper all at once. — aletheist

Well the start of time would be at the Big Bang / Big Crunch point of the circle.

Again, in reality there are no instantaneous states, so the "probability" of any such state occurring is meaningless. Besides, in infinite time there would be infinitely many such states, and no reason in principle to assume that any two of them are identical. — aletheist

"In physics, the Poincaré recurrence theorem states that certain systems will, after a sufficiently long but finite time, return to a state arbitrarily close to (for continuous state systems), or exactly the same as (for discrete state systems), their initial state."

https://en.wikipedia.org/wiki/Poincaré_recurrence_theorem

So the argument I've used is valid for continuous systems too. Eternal past time leads to the absurd conclusion that the universe has been in the same identical state a greater than any finite number of times.

Now we really are going in circles. I see no point in continuing (pun intended). Cheers!

OK. Thanks for the conversation.