I was talking about the time before the first time travel event; before you've ever time travelled. You're talking about what happens if (or after) you time travel, so you're not talking about the time before you've ever time travelled. — Luke

OK, Bob makes the machine and uses it to go from 2024 to a new timeline starting at 1990. Any point on the original timeline before Bob vanishes from it is the time before the first travel event. There is no time on the new timeline before the first travel since it starts there, kind of per last-Tuesdayism.
I lost track of the question about this 'time before'. Are we talking about say 2023 on the original timeline or am I still getting it wrong?

Why can there be no odometer on the time machine counting jumps?

There can't be one on the machine that jumps in the loop. Bob's machine can have an odometer, no problem.
The contradiction: Suppose, just before the jump, the odometer reads x. It arrives at its destination (8 seconds in the past in my popper example) and immediately increments the thing to x+1. This contradicts it leaving 8 seconds later with a reading of x.

Sorry to be unclear again. What I meant was: how could I be in the past as a time traveller prior to the first use of the time machine.

Just repeating the same question doesn't make it clear. Are we talking about Bob and the copy-timeline scenario? If so, you need to specify which timeline you're referencing when talking about one thing being prior to another.
In general, if one considers that Bob builds the machine and first uses it in 2024 and uses it to go to 'the past' (no timeline specified), then since 'the past' is typically considered to be prior to 'the present', Bob is in the past as a time traveler (in 1990 or whatever) prior to 2024 since 1990 is often considered to be prior to 2024. So that's how he's in the past prior to first using the thing. It's the whole point of the machine to be able to do this.

What are these different causal rules?

That's what you are apparently trying to figure out. I don't know either, so I'm also exploring. What I don't do is presume the usual rules, such as that a place that almost looks like the state of things in 1990 is prior to the state of things in 2024. I also don't presume that the cause of a thing is necessarily prior to the thing. That's a pretty obvious one to throw out.

There are still causes and effects, it seems.

Agree. We're trying to keep that. The loop is causally closed, so I don't see it as a contradiction. The cause of the 8-second guy is his own travel event 8 seconds later.

The older self can teach the younger self about time travel technology and the younger self can then use that knowledge in order to time travel from the future to the past. Or, the younger self can steal a time machine from the museum and then later use that time machine in order to donate the time machine back to the museum. The only different causal rule appears to be that there can be no first time travel event or that we are not allowed to talk about the first time travel event, for some unspecified reason.

There is no first time for the loop, or if there is, it's the only time. There is after all but the one jump, per the external timeline, presuming its a simple loop. Only the machine's timeline has multiple jumps, plus its contents if those contents go from arrival all the way back into the machine at departure.

So some causal rules are okay, but not others?

The ones not OK lead to contradictions. The looping machine having its own 'first time' leads to a contradiction. It would effectively be an odometer going from 0 to 1, and we showed how that is a contradiction.

We may never ask/explain how a time machine came into existence in the universe

It came into existence by traveling from 'the future'. You can ask and that's the answer. That universe allows that sort of causality.

but it's okay to ask/explain how a time machine came into existence in a museum?

By being donated of course.

I think the rule you find hard to discard is that all seemingly artificial things must somehow be invented and assembled at some point, and the examples we show are consistent without all those steps. Sure, the machine is built in the teaching loop, but the technology knowledge (the inventing) is the loop, information that is never gleaned, but is merely passed on.

It's logically self-consistent as long as we never consider a loop as having a first time travel event or what preceded it, it seems.

You're trying to find a logical inconsistency, and I don't see one. Before the loop, the machine simply doesn't exist, nor does it after. The 8-second machine exists but for 8 seconds. Not time to study and figure out how its done, something the museum guys might decide to attempt.

But how is it logically self-consistent that there was never a first time travel event?

The same way that the lack of the most eastern point isn't a logical inconsistency? It's only inconsistent if you presume there must be a first time (on the machine's timeline), so that's apparently a wrong thing to presume. There's a first time on the world's timeline. Isn't that enough? This presumes that the external world is itself not a loop. There are hypotheses that suggest otherwise, a sort of cyclic model of the universe.

Does time or causality work differently in these scenarios such that it would be impossible to trace back to the first time travel event?

The infinite-age universe hypothesis similarly suggests the impossibility of tracing back to a first event. A loop without a beginning is not in contradiction with anything.

Not if it is part of a loop. The whole 'must be a first time for everything' is only a rule in a universe like ours, intuitive to us, but not true in the sort of scenario we're discussing.
— noAxioms

How is it "not true"? It doesn't seem to me that it's not true; it seems that you just want me to ignore it.

It's true in our universe because I cannot think of a scenario where at some earlier time there is not a mug, and at a later time there is a mug, and that there it a beginning to the mug's timeline. The timeline of the mug and that of the rest of the universe is completely parallel, so there must be a 'first moment' for it. In this alternate universe, the mug timeline might not be parallel. It still has a first (and only) time in the universe timeline, but not on its own timeline, which isn't parallel to the one 'outside'.

We're dispensing with evolution, too?

If a machine that loops and is never created can exist in some consistent way, then so can a creature than has no evolutionary ancestory. It just appears from some retrocausal event, and its existence somehow eventually plays a role in that eventual retrocausal event.
So it's like humans have no evolutionary ancestors (despite the biology folks suggesting otherwise). But some time far in the future, when humanity is near its end but they develop time travel, the put a couple back to year -4000 and name them Adam & Eve.
Geez, the religious folks would jump on this if it wasn't supposed to be blamed on God instead of retrocausality.

I suppose I could alter P2 to say that time machines involved in human time travel need to be created at some point.

OK. The 8 second machine is created in front of me at some point, and un-created 8 seconds later when it vanishes. Works for me.

Or, better still, P2 could say that there must be a first human time travel event associated with the human use of a time machine or time travel device/technology (assuming that any such events occur).

Dangerous to use the word 'first' when the temporal ordering of things is not objective. I think that's where a lot of the trouble comes from.

You are effectively telling me to ignore how the time machine came into existence originally.

Nope. It came into existence when it first appears, not 'uncaused'. It doesn't exist at any time before that, so that is it coming into existence. It gets donated to the museum some time later and yet later is stolen and vanishes from existence forever after as it causes the earlier event.

now you're invoking fanciful beings that can time travel without any time machines and other magical shenanigans in order to try and save the "self-consistent" logic of causal loops.

I just made them up as another example which isn't directly self contradictory.

I should probably withdraw my language that these things are not created. They are, but the causes of their creation are events that are future events as measured by the world-outside timeline, a timeline which we are presuming to be reasonably linear and therefore orderable.

Am I supposed to be the guy in the box/time machine, because this doesn't sound like a causal loop

No, you are the spectator who has somebody use your popper and then take it from you. The person in the box is, well ... something else. It is along those lines that you should tear this apart. A human makes a great odometer, and you can't have an odometer, so the guy is perhaps not human?

it's just a guy using a time machine to go back in time every 8 seconds to do the same thing repeatedly. It's unlike the other causal loop scenarios because it's not clear that I ever become the guy in the box/time machine. Or was that part left unsaid?

Yes to the first. No, it's never you. You're left behind being befuddled, remember? You never see him again. It very much is a loop, and a very tight one.

I thought by "un-created" you meant that the time machine was not created or did not exist. Did you mean "uncaused"?

Un-create means to cease existing. From the perspective of the linear timeline, Any traveler uncreates his machine and himself. It's just gone leaving not even disassembled parts. Of course on the machine's timeline, it just has an external environment change and isn't an act of creation or uncreation at all.

Somebody could catch a video of uncreation on their phone. You're were taking a video of a train crossing because you heard a train in the distance. The video records a blue car suddenly appearing (an act of spontaneous creation) at the crossing which then crosses and continues on. The lights go on and the gates come down, just as an identical blue car pulls up and vanishes (uncreates) at the exact spot where the prior car appeared. You have a video of time travel in action, and it even worked in the universe of Bob where a new timeline is created each time. The video is subsequently sold to the time machine sales people who use it to pitch their product.

Right, that's why I've been arguing that time travel only makes sense on multiple timelines

For the most part I agree. But single-timeline travel isn't necessarily contradictory so long as one does not make choices known to be different than those made before. It does require a sort of lack of free will as it is often defined.

I was just trying to restrict it only to causal loops that do involve time travel, in case you were about to bring up any causal loops that don't.

Any loop in time is contrary to the sort of linear ordering of all events that we find intuitive. No, it doesn't have to be labeled 'time travel'. A cyclic universe is a nice loop that isn't considered time travel because there is no linear timeline laid alongside the loop.

so it does not follow that every timeline would quickly have a population of zero. — Luke

Agree that if the people (especially those on the original timeline) fear the readily available devices, they wouldn't get used and the population remains.

The people on the alternate timelines would have solid evidence of it working. A car gets caught by a train crossing, so it goes back 1 minute in time and sneaks across the tracks before the gates come down. The traveler can see her own car doing it as she approaches, and the bystanders can witness it as well. The person doesn't always reappear. Sometimes the traveler is gone forever. Kind of pot luck, a game of Russian roulette.

What I meant was: how could I already be in the past before I have ever time travelled?

If you time travel to the past, by definition you end up somewhere 'before' the event where you initiated the travel. I kind of lost track of the context. Are we talking about the loop here?

I could already be in the past (on a single timeline) if I had time travelled before but, given causality, there must have been a first time that I ever used the time machine to time travel.

I don't see how that follows with the loop scenario. There would be no 'first time' to a loop. As I said, there can be no odometer on the machine counting jumps. That would be a contradiction.
If it's just teaching the younger-self how to do it, then every jump is the only jump, so I guess that would count as the first (and only) time, or at least the one jump that defines the simple loop.

How could I already be in the past prior to that?

I don't understand this. If the jump is from 2024 to 1990, then 1990 is 'the past' destination, and you are not in a past that is prior to that except perhaps as a young person, the one that you teach.

That sort of logic only holds water because there are no time machines possible.
— noAxioms

Why are no time machines possible? That's not something I've said.

We're presuming they're possible, hence the logic you give being fallacious. Things that are impossible in this universe are not impossible in this alternate universe where time travel makes for different causal rules. A loop is valid under the new rules. It doesn't violate anything except the rules of this universe.

We can just discard causality and assume that time machines don't need to have had a first ever use, and we can conveniently disregard whatever history led up to that first ever use?

You need to discard the causality rules of this universe, yes. The rules are different in the universe we're discussing. With the loop scenario, there is no 'first ever' to it. You can't count them. The loop is just there, and is self-consistent.

but if a creature evolved a way to do it, then there must have been a first time that they ever time travelled.

Not if it is part of a loop. The whole 'must be a first time for everything' is only a rule in a universe like ours, intuitive to us, but not true in the sort of scenario we're discussing. Yet again, a simple counterexample falsifies your assertion. So maybe this time traveling creature never evolved, but just is. Again, there are movies depicting pretty much this.

P2 is unacceptable. It's like trying to prove God by asserting that the universe needs creating at some point (which is itself a self-refuting argument). An un-created time machine does not violate any rules in a universe where time travel is possible.
— noAxioms

If you accept that one cannot travel without a time machine (P1) - at least, for the sake of argument - then it follows that a time machine (or the means for time travel) must be created or have evolved or somehow brought into existence in some manner.

I will not. We're discussing the possibility of closed loops, and loops falsify P2.

it's merely assuming the universe must have been brought into existence (which is quite self-evident).

I don't find that evident at all. It violates Einstein's theories for starters, which suggests that time is part of the universe, and not something in which the universe is contained and in need of being created.

Hopefuly we can agree to the standard scientific view that the universe's existence began with the big bang,

Pop science view maybe. OK, if one confines one's definition of 'the universe' to just what evolved from the big bang, then a good deal of them would suggest a larger structure from which that bang was initiated. But there is no before/after without the sort of time that boiled out of the bang, so calling it 'before' is misleading.
If you consider the universe to be the entire quantum structure, which includes all the stuff that 'springs' from it, our bang being one of them, then that structure is not something 'created' or 'caused'. It cannot be, both terms implying a larger container for something we're defining to be the largest container.

but even if we might assume that the universe has always existed

Sorry to be so buggy, but I don't buy that either. The phrase once again implies a universe contained by time, and not the other way around. Yes, there are those that suggest something like that, in denial of Einstein's postulates.

To say that time machines have always existed is more like saying that waffle irons have always existed.

Your wording suggests that the machine exists at all times, which isn't the case. It exists in the loop in the museum case. It doesn't exist at other times.
In the teaching case, it is built by young Bob in say 2022 with knowledge from his older self. Bob can then use to travel all over the place, here and there, to say the restaurant at the end of the universe. Eventually he goes to 1990 and teaches his younger self the secrets. Then he's off again to see even more wonders. Point is, there no point in time where that machine cannot be unless it has a limited range or something.

An un-created time machine does not violate any rules in a universe where time travel is possible.
— noAxioms

Sure, but it would imply no time travel.

Same counterexample falsifies this.

Imagine you're holding one of those party poppers that you pull and it explodes a bit of confetti around. You're about to do it and a box appears in front of you from which a some guy jumps out and explodes his own party popper as he says "three!". Then he grabs yours (unpopped), and apologizes, says the box is a time machine that goes back 8 seconds, the says "One, two, ..." and the box disappears, leaving you simply befuddled. That's what an 8 second loop looks like.
It's not too hard to take that one apart, but not by the logic you've been attempting.

Why can't there be a first time in a loop? Loops are immune to causality?

The kind of causality rules you're thinking of don't exist in a universe with time travel. A first time for a loop would contradict its existence, which is travel from the other end of the loop and not somewhere else.

P4 [the initial time travel event cannot cause itself (e.g. by a prior time travel event using the newly- invented time machine).] is not irrelevant. I'm saying that if a loop involves time travel (as the examples in the SEP article do), then we can consider the first ever time travel event in that loop and what preceded it. Unless you are arguing that there is no causality in a loop or that time travel loops and time machines in loops have always existed? Why should causal loops be immune from causality; from having been caused? It seems like a bit of magic.

Not following. There a possibility of a loop that doesn't involve time travel? Example please.
You're then referencing the 'first ever' go-around for a loop that cannot have such a thing. So that's what you mean by 'initial travel event'. There is no such thing for a loop, so I must withdrawm my 'is OK' assessment of it.

Unless more than one person used the same time machine to time travel together. — Luke

Right. Neglected that bit.

I don't understand why the planet would quickly have a population of zero in all timelines though.

well, if everybody had one and knew it worked, I suppose they'd all use it and exit any particular timeline. It's sort of like heaven: The sales pitch is great, but if it's such a better place, why does nobody voluntarily hit the button and go there? It's because from the perspective of the original timeline, it just looks like you vanish, never to be seen again. There is zero evidence that it is safe, let alone works.

Bob would continue to exist on any timeline he travelled to (at least, until he dies).

And Bob is missing from every timeline except one. Of course on the other timelines, there may be many people that attest to having traveled, and the evidence is there that it works. Those timelines would empty out faster than the original, if only from people going back to times when there were still people to meet.
Nobody on these worlds knows who the actual time traveler is (the one that created this world), not even Bob.

But how could I already be there before I time travel?

With a time machine of course. That sort of logic only holds water because there are no time machines possible.

If I go back 250 million years to see the early evolution of mammals, I'm sorry, but humans will never evolve from that timeline.

How?

Evolution is a chaotic function. The popular term for it is 'butterfly effect'. The killer asteroid is not chaotic, so you don't alter that, but evolution is a random process, and you've totally altered that. People are not an inevitable result of the state of 250 million years ago. It's an inexpressibly low chance even without the traveler mucking things up. OK, that last statement presumes a lack of hard determinism. Our discussion has a lot of quantum interpretation implications as well as implications for interpretation of time. The SEP article didn't mention the former.

Maybe it all boils down to this. I'm arguing that causal loops require a start; that there must be an initial time travel event which causes the loop in the first place

No. The whole point of them is that they are uncaused. They'd not be a loop if they were caused. That it doesn't fit in with your notion of singular causality is irrelevant since all those rules must be discarded with reverse causality.

My premises would be that:
P1 - one cannot time travel without a time machine
P2 - time machines need creating at some point
P3 - there must be an initial time travel event following the creation of the time machine, when the time machine is first used to time travel
P4 - the initial time travel event cannot cause itself (e.g. by a prior time travel event using the newly- invented time machine).

I added bold labels. Let me know if I did it wrong.
P1: I said I would accept this for this purpose, but there is no such requirement. If time travel was possible, somebody might be able to do it just by willing it. If a machine can do it, why can't a creature evolve a way to do it. The premise is something like saying you cannot get to grandma's house without a car. Well, that's false since evolution has given us a means of machineless locomotion.
P2 is unacceptable. It's like trying to prove God by asserting that the universe needs creating at some point (which is itself a self-refuting argument). An un-created time machine does not violate any rules in a universe where time travel is possible.

The argument falls apart there: an unacceptable premise, which again, has a counterexample which falsifies it.

P3 seems false. I might make a time machine but never use it. We presume you mean the machine in the loop, so yes, it just happens to get used (the 'first time' say) in the story you are creating. I put 'first time' in scare quotes because there can't be a first time in a loop: there cannot be an odometer on the machine that records how many jumps it has taken.

P4 is OK, but seemingly irrelevant since your story involves only a single time travel event, no loop at all.
I listed them all as postulates (and no conclusions) since none are worded as conclusions. I don't see any conclusion of the impossibility of a closed loop. The whole things doesn't seem to discuss loops at all. It discusses only a created time machine, not a looping one.

By your rules, a person can only be in the presence of but one actual time traveler, even if other people on the timeline also remember time traveling.
— noAxioms

I'm considering Bob to be the first ever time traveller.. — Luke

You're not reading my comment. I said that by your rules, a person can be in the presence of at most one actual time traveler. We could have a factory that made them like bags of cheetos, and everybody used them to get to appointments and catch the traffic light that just went yellow. If they were used like that, the planet would quickly have a population of zero in not just the original, but all the timelines. Despite that prediction, no person would ever be in the presence of more than one actual time traveler, which is the one and only person that created the specific timeline the person finds himself in (if he's still in it and hasn't left already).
Actually, nobody would use the machines, due to the overwhelming evidence of it being nothing more than a self-annihilation machine. So good thing Bob is the only person that has one, and only Bob fails to exist in pretty much any of the timelines.

if they were to travel to the past then that would be altering something about the past.

But the way you describe it, it isn't really the past, just a different timeline which maybe looks like 'the' past, but is actually just another line, 'a' past at best, one of many. There is only one 'the' past, and you didn't go there.

Well, I wouldn't call [a read-only verision] "travelling to the past", That is just somehow viewing the past at the present time of the viewers.

Except he can't leave (turn off the TV so to speak). OK, I agree that it stretches the definition too much. But if he's there at all, history is gone. If I go back 250 million years to see the early evolution of mammals, I'm sorry, but humans will never evolve from that timeline. Your very presence destroys that, although it doesn't prevent the asteroid that wipes out whatever is there instead of the dinosaurs.

the time machine must have been built by someone else in order for it to have been stolen and then donated to the museum. — Luke

Does not follow. That sort of reasoning is only valid if time travel is not possible. The whole point is that it was never built. — noAxioms

You will need to explain why my objection does not follow.

Try to state the logic of your statement formally. What are the premises? How does your conclusion (that the closed-loop machine must have been built) follow? One of your premises is perhaps that all things need creating at some point, but that premise begs a universe with no closed time curves.

If you don't have a time machine then you can't time travel

I accept that premise, at least for purposes of this issue.

so you can't then obtain that time machine (or its technology) via time travel.

Why not? It works, does it not? This is worded as a conclusion, not an additional premise. I don't accept it since 1) it doesn't follow from the premise, and 2) it is easily falsified by counterexample.

Again, try to word it more formally, and the errors will stand out better. I think the example of learning the technology from the future works better than actually being handed the machine.

Which hole are you referring to? Entropy?

The one we discussed: the machine needing to exist for infinite time without showing any wear. Hence better to be handed the plans than to be handed the machine. The movie predestination works that way. It depicts a closed loop, without the infinite-age issue.

The loop could have started only 10 years ago.

It wouldn't be a loop if it had. Loops don't have a start.

he should not be surprised by his sudden appearance at an earlier time, unlike everyone else on the new timeline (who we would assume have never encountered a time traveller before). — Luke

By your rules, a person can only be in the presence of but one actual time traveler, even if other people on the timeline also remember time traveling.
Any witnesses to Bob's first moment's on that timeline will be surprised by the apparent sudden appearance of Bob, despite the fact that he has been there all along and by definition doesn't appear somewhere where he wasn't just before. There simply isn't a 'just before' on that timeline.

Also, instead of considering the new timeline as a copy, you could consider it as a re-writing of history, but one which does not eliminate the original timeline.

That would be a different convention. The new timeline is a rewrote-history according to traveled-Bob, and the old timeline becomes the copy from which he originated.

if they were to travel to the past then that would be altering something about the past

There are stories/scenarios in which nothing is altered. It's more like watching the past on TV since nothing there can detect you.

Incidentally, based on my very amateur understanding, I had thought that once the Einsteinian "time traveller" had returned to Earth, the same amount of time must have elapsed on Earth as it has for the traveller, given the time dilation effects of turning their ship around in order to return. When I read about the twin paradox long ago, I figured that although one twin can be in the future of the other, there is no way to transmit information to the Earthbound twin which could give them advanced knowledge about the future and that they must both return to the same proper time when they meet again. However, I admit that I don't fully understand these things and I'm probably way off. Besides, those sorts of time travel scenarios involving that type of "time travel" are not what I had in mind here.

The twin goes out and comes back, and the two twins are no longer the same age. Not sure what you've been reading, but the proper time going out and back is less than the proper time of a direct path between the two events where the depart and meet up again. None of this requires anything communicating or having knowledge of what the other is doing.
It works better with pregnant women, who make great clocks. Betty and Veronica both get knocked up the same day and Betty takes off for the stars. Betty comes back in 9 months her time and has her baby. She meets who she presumes to be Veronica also giving birth in the same hospital, but it turns out the woman is actually Veronica's daughter giving birth to Veronica's grandchild. 20 some years have passed on Earth since Betty left and the other woman is merely the spitting image of her mother.

Anyway, SEP considers that to be time travel.

The scenario wants us to imagine that this is a logically-sealed causal loop. However, the time machine must have been built by someone else in order for it to have been stolen and then donated to the museum.

Does not follow. That sort of reasoning is only valid if time travel is not possible. The whole point is that it was never built.

It wasn't the time traveller that built it, so it cannot be the donation by the time traveller that causes the existence of the time machine.

The existence is caused by its own time travel to the past. Such is the nature of closed loops. Still, in my prior post I pointed out a hole in that story.

...imagine a time traveller who explains the theory behind time travel to her younger self: theory that she herself knows only because it was explained to her in her youth by her time travelling older self.

That version works better since it mostly solves the problem I identified.

I don't buy the fact that the time traveller could not have done something else.

The closed-loop scenarios illustrate free will (or more precisely, the lack of it) better than any discussion about reality where there's no pragmatism to it.

There are several fiction stories with closed loops. Predestination is one of them.

[The machine] stays perfectly new at all times, which isn't plausible for something that is thousands of years old.
— noAxioms

I don't follow why it must be thousands of years old.

It takes say 10 years from donation to museum to getting stolen. It ages 10 years during that time after which it goes back 10 years and does it again, and again... Infinite age since it's stuck in a loop. Somebody has to do one excellent refurbishment effort somewhere during each 10 years.

I imagine the causal loop in these scenarios to be a much shorter period than thousands of years.

It's a loop. It has no finite length, just a period, just like there's no end to walking east.

It's not a copy of Old Bob, since he time travels from the original timeline to the new timeline. — Luke

That part is just you saying it. It could be just as easily said that everybody (including old Bob) in the copy timeline is a copy. The machine could split Bob just like it splits everything else. The story doesn't go like that, but the story could go like that. It would still be time travel of a sort, especially from the PoV of the Bob on the created timeline.

What's the point (or possibility) of time travelling to the past if it is to leave the past completely unchanged?

I take it you're not a historian. Those guys would love a machine that lets them go back, even in a way that cannot alter anything, just watch.

This is the only way to avoid contradictions, paradox and violations of causality.

I think we're doing considerable damage to causality if any of this were plausible. OK, the Einstein time travel doesn't violate causality, but I personally don't think that one counts even if it meets the SEP definition.

Right, but I'm attempting to point out why I think single-timeline examples of time travel are senseless, and why I believe that a second timeline is necessary to avoid contradiction or paradox.

The SEP article gives several examples of a single timeline without paradox, Some of the best are the loop ones, including a case where you don't even need to invent/build the machine. You just give it to your younger self when you're done with it.
There is a sort of paradox with that scenario which is how the machine experiences no entropy: It stays perfectly new at all times, which isn't plausible for something that is thousands of years old.

Most designations of identity have pragmatic reasoning and are thus not arbitrary. This doesn't, so the question needs asking, and the answer needs justification.
— noAxioms

I am arguing that Old Bob cannot have been in the past originally, because Young Bob had not yet grown up to build a time machine or to time travel. — Luke

You don't seem to understand my point, which is that there is not obvious convention as to if the old-Bob in the copy timeline is the same old-Bob from the original timeline. The usual conventions for saying this person is the same person that looked like him yesterday. "I bought a can of beans yesterday": True? By convention, yes, the person who bought the can of beans is the same person that submitted this post. We know that because we know the convention. There is no convention for crossing timelines. To me it looks like old-Bob commits suicide, but builds a copy of himself (and the machine) in a timeline with a copy of everything else. The convention could just as easily say that.

This could be achieved by something like 'waiting', which the SEP article categorises as not time travel. — Luke

The SEP does consider travel to a parallel timeline (Meiland, 1974 or Deutsch and Lockwood, 1994) to be time travel despite the lack of rigorous simultaneity convention between separate timelines. Your idea seems to illustrate the same issue. What was this alternate timeline doing before Bob traveled to it? Was it different in any way? Did it have a 'present' 2024 that was altered by Bob's appearance in what was considered to be 1990 at the time?
I don't think your timelines are parallel like the ones discussed in SEP. The question seems more approriate for the above mentioned authors.

However, you have suggested that the inhabitants of the second timeline are merely "copies". — Luke

You seem to regard them as copies yourself, as evidenced by several comments (my bold):

There is no point or possibility of travelling to a destination if you are already there.

So by this wording, the young Bob that gets killed is not Bob. He is not already there, but is rather killing a copy, somebody else, having left the young Bob that is actually himself back in the original timeline unkilled.

The younger self does not time travel; the older self does.

If the two of them were the same person, this would be a direct contradiction. But you seem to regard them as not the same person. So if (actual) Bob goes to some parallel world in 1990, and waits several years for the perfect opportunity to take out the young-Bob copy1 that is there. The moment comes, and he fires his gun only to find it wasn't loaded. Opportunity lost, and there won't be another one. But he has a time machine, so he goes back a day and loads the gun that yesterday-Bob (also a copy) can use to complete his task (of killing young-Bob copy2, leaving young-Bob copy1 un-shot back in the first alternate timeline).

My point of all that is that your comment is true only if you assume yesterday-Bob is a copy. Yesterday-Bob (the one with the empty gun) has time traveled (he's the original Bob). Yesterday-Bob copy has not, him being a different person with a false memory of having time traveled. If he was the same person as Bob, then he very much as time traveled, explaining his presence in 1995 and his memory of 2024.

I'm saying time travel is senseless on a single timeline — Luke

All the examples of 'is time travel' at the top of the SEP article are single-timeline examples. I'm not saying that traveling 'sideways' to a different line is or is not time travel, but I'm saying that those examples cannot all be senseless. Yes, they all have potential paradoxical consequences, all discussed in the article.

My answer was that it is Old Bob from the original timeline who time travels and kills his younger self (on the new timeline). — Luke

My reason for asking was to figure out justification of that assertion. I'm not saying it's wrong, just an arbitrary designation. Most designations of identity have pragmatic reasoning and are thus not arbitrary. This doesn't, so the question needs asking, and the answer needs justification.
You wanted to explore the implications. I'm trying to do that.

Fiction is just that: Non-evidence, so it doesn't in any way constitute an argument one way or another.
Apparently Kirk was also split by the transporter, but not identical. So the story changes as the plot requires.

But it should, in principle, be possible to make a complete copy (à la Thomas Riker), who feels, thinks.. exactly the same as Will Riker. — Walter

This seems to be an assertion against body-soul dualism, not for it. The trek writers have always sort of somewhat presumed monism, but the subject has come up before with Dr. McCoy disliking the transporter since he considered it a copy/suicide machine. He said it always made a copy and destroyed the original. He simply chose a different convention.

Presuming that one isn't lacking something supernaturally critical (such as memories stored in the mind instead of physically), neither would know which is the original, and the designation seems to fall to convention. I might say that the transporter successfully beamed him up and mistakenly left a copy behind. The one on the planet is the copy. That's just a different convention. Identity is just an abstract convention, and the human identity convention stops working in sufficiently alien situations like say mitosis. Say I have 3 fuse strings tied at a common point. One end is lit, creating a flame. The flame gets to the junction and goes both ways down the other two fuses. Which is the original flame? Our convention doesn't answer that, and neither does physics.

You use the word 'soul', which has connotations of an entity with identity that gets held responsible in the afterlife for its choices made in life. If it doesn't hold memory, then it cannot know it is being judged, let alone why. If it does, then it will be quite obvious which Riker was the copy since the copy would have no memories. Maybe, rather than be a zombie (which you seem to assert otherwise), it would be assigned a new 'soul', which seems to absolve it of all the sins of the original, such as that time he took a leak on the captain's chair, earning him the informal title of 'number one'.


All this said, MWI interpretation says you are copied all the time, with no distinct 'original'. The convention is that each copy (and everybody that interacts with it) assumes it is the original. Is dualism then totally incompatible with MWI? I think not. If it were somehow proven, the dualists would find a way to bend their story, but the 'soul' thing would really take a hit. Which one is responsible for some random bad choice made at some point?

I wouldn't call that time travel in the relevant sense. The SEP article attempts to draw the relevant distinction: https://plato.stanford.edu/entries/time-travel/#WhaTimTra — Luke

I read almost all of the SEP article. Thanks for the link. Didn't know it had a page on the subject.
Apparently what I am doing right now does count as time travel, so long as I move. Motion-related time dilation apparently counts according to the article. I would not have said that.

The definition they referenced (but were not totally satisfied with) came from Lewis (1976) who says that time travel is when two different kinds of time (coordinate time and proper time say) don't match. The clock on the wall says one thing, but my watch says another. But physics allows that sort of stuff all the time. One is always proper time, but the other is something else. The SEP article seems not to know physics enough to actually use the correct term of 'proper time', the length (interval) of any worldline, which is what a clock measures if it follows that worldline.

The article doesn't ever reference 'branching', but it talks about traveling to other parallel timelines, like those lines already existed, and apparently already have the time machine appearing from nothing in them, much to the surprise of any witnesses.

The article does seem to accept travel to a 'nonexistent' time in the past under presentism. All of section 4 discusses the various interpretations of time and their implications.

My point was that it is senseless for Bob to travel to the past if he is already there — Luke

Him already being there was the point: To alter what he (younger self) would have otherwise done. I see no reason why the younger self cannot have already time travelled before. Another mistake could be made, 'necessitating' a second correction. I put it in quotes because the mistake cannot be corrected on the more original (more real?) timeline.

I don't think so. I'm assuming that Bob returns to the same past that he lived through when he was younger — Luke

Poor assumption. If I'm to 'kill grandfather', I'd have to go back at least a century. Maybe I want to witness the asteroid taking out the dinosaurs. You can't put in a rule that says you can only travel a short ways to some past with you in it somewhere.

You also contradict yourself. You say on one hand that it is senseless to go to a time when you exist, and on the other hand you're presuming Bob does this 'senseless' thing.

I don't; that's the scenario of the Grandfather paradox. — Luke

Any travel to the distant past will destroy the history you know. Everyone talks about critical events that make a change, but just appearing and stepping on a bug is enough. That said, killing grandpa isn't necessarily paradoxical. Maybe you're not actually related to him, but rather the mailman. I know my grandfather was a cheater. Why can't grandma be?

Call it a copy if you will but this is the only way that time travel is possible. — Luke

Well, besides the fact that it isn't possible in the first place, there are valid scenarios discussed in SEP that allow travel to the original timeline. CTCs are one example.

I assume that it is the Bob (or Luke) from the original timeline — Luke

I don't make that assumption. I try to work it out.

As I replied earlier, I wouldn't call this recreation of another time in the present time to be time travel. — Luke

I'm not recreating a time. I'm just moving a Studebaker forward in time by a century. That's pretty much exactly what you're describing, except in the forward direction. So tell me why that's not what I did. How do you back the assertion that the car didn't travel through time, but Bob (also going forward say) did. Was it the lack of a fancy time machine looking device with blinking lights and stuff? There's plenty of fictions where the machine isn't necessary.

They've done quantum teleportation, which counts as time travel according to SEP. They put something in a booth and it teleported it to another a couple hundred km away. Arguably not time travel, but my question is: Is the thing at the far booth (and no longer at the near one) the same object, or a perfect copy? They were asked this question, and replied: "What possible difference does it make?".

The time traveller was either never at the destination time and cannot return there without contradiction (having two conflicting histories on a single timeline), or else they were always there and therefore cannot "return" there. — Luke

How is the 2nd clause different from the first? Both just seem to say that you can't travel to your own timeline, which is partly silly because I am doing it now. IOW, does forward time travel necessitate a branch in timelines?

One can pretty easily do it just by having a space ship that travels faster than light, and fiction is full of that as well.

Bob must travel to, and insert himself into, a past time at which he didn't always already exist as a time traveller.

Why this restriction? I go back to 1955 (standard destination). Hang around until 1970, and go back to 1960 this time, where "I" already am as a time traveler. What's wrong with that? Can he also make a 3rd branch off the original timeline? Can I, having just made the machine, branch a new line off some other timeline where I never existed in the first place, say some version of 1980 where my parents didn't survive WWII?

Meanwhile, why do you want to kill anybody? The young-Luke you find back there is not you since 'you' is presumably on the original timeline. You've no reason to kill this other person or for that matter, anybody. If you kill yourself, have you killed Luke, or did a copy kill himself?
I think you need to consider the question I asked about the Studebaker in my first post in this thread (about post 57). Is that time travel? If not, why not? What is your machine doing that my example with the Studebaker did not?

I often bring up the famous rhetorical question that Albert Einstein asked his friend on an afternoon walk (I think it was Abraham Pais): 'Does the moon continue to exist when we're not looking at it?'

I think the answer is obviously 'yes' but the question I would like to ask is, why did he feel compelled to ask it in the first place? Why did it bother him? — Wayfarer

The quip was said in the early days of quantum theory when what is now known as the Copenhagen interpretation was 1) pretty much all they had, and 2) was strictly an epistemological interpretation, concerning what was known about a system and not what was. Ontologically, only the Wigner interpretation (leading to solipsism) suggests that human observation has anything to do with what is.

So Einstein perhaps was vocalizing some of the apparent implications of quantum experiments, that one cannot know the state of a system (moon in this case) between measurements. And indeed this is true. The moon is not in a specific state relative to anybody on Earth since it is over one second away and any measurement of it is quite old. That statement is wrong if one presumes counterfactuals.

They give philosophical questions very short shrift. — Wayfarer

That they do. Wrong forum to ask that sort of stuff. But most of the forums that do allow it don't have the sort of expertise found there. I mean, I'm a mod on one of them, and apparently 'top dog' on things like relativity and maybe QM, which is pathetic since I would utterly fail a college level exam on either subject. I learned enough to glean informed implications of both theories on philosophical topics, but not enough to actually do the higher mathematics.

I dabbled in mathematical Platonic realism (especially as it applies to Tegmark's MUH) and found it lacking in explanatory capability. The kind of ontology that makes sense to me simply does not apply to most mathematical structures functionally different than our own. I'm of course not 'most other contexts'.

So many of the actual philosophy forums suffer from a lack of posting standards, resulting in a negligible signal to noise ratio.

A: Yes, but only up to the point where the rate is so high that the interaction between different electrons can no longer be neglected. — Wayfarer

OK, should have thought of that. I was kind of thinking photons, which don't interact with their neighbors nearly as significantly as something like a charged particle. So I pictured a laser weapon aimed at the slits...

It's the nature of that existence which is the philosophical conundrum. It's not as if it's precise position and momentum is unknown, but that it's indeterminable. It will be found whenever it is observed, but the sense in which it exists when not being observed is what is at issue.

The moon was measured. It's still there despite it not being measured at the moment (like it's possible to ever not measure the moon from anywhere as close as Earth). The proton is like that, but with not quite as many 9's to express the probability of it still 'existing'.
The conundrum of which you speak seems to be that the proton in fact hasn't an objective location/momentum at all between measurements. The moon, being classical, isn't like that. But quantum theory doesn't say that the particle doesn't 'exist'. A few interpretations say it does in fact have these properties at all times, but they're just interpretations. The others might still say it 'exists', in the manner of say energy, charge, baryon & lepton number conservation. It can't just not-exist. It just lacks objective properties that put it in a specific state.

Note that if I say something different from the physics-forum guys, they trump me. There are some really solid experts over there, and I don't often respond to questions for fear of putting my foot in my mouth.

Another note: I've not been reading this thread, so not sure how the non-classical nature of QM has to do with 'best argument for (or against) physicalism'.

The Schrödinger equation's solution is called a wave function. If one simplifies the equation considerably it has the form dQ/dt=kQ, which has solutions involving e^it=cost+isint, giving it repetitive or wave-like characteristics. — jgill

Thanks for the clarification, which was mostly about the terminology. Yes, it definitely has wave-like characteristics.

But what is the probability wave, other than a distribution of probabilities? The answer to the question ‘where is the particle’ just IS the equation, right up until the time it is registered or measured. So the answer to the question ‘does the particle exist’ is not yes or no. The answer is given by the equation. So you can’t unequivocally say ‘it exists’ - you can only calculate the possibility that it might. (This torpedoes Democritus ‘atoms and the void’ by the way.)

So - does that mean ‘yes it is?’ - let’s ask noAxioms. — Wayfarer

The probability of measuring some part of a system can be computed from the wave function. I've not heard the result of that computation being referred to as a 'wave', but I'm sure it is somewhere.
Does the particle exist? That's a counterfactual, so there is only a yes/no answer given an interpretation that posits counterfactuals. Quantum theory would simply give a probability of measuring it here or there, or not at all. You can confidently say about some proton that it 'exists' mostly because outside of the sun, protons are pretty stable * and don't just cease existing, so it exists but you don't know exactly where it will be next measured.

I was asked if 'yes it is' is correct, in reference to:

But the wave form of the particle is not the probability wave of the particle is it? — jgill

No it is not. The wave function of the particle describes its quantum state. The probability of where it might be computed from that wave function, but the wave function itself is not a 'probability wave'.

it is well-known that if only one particle at a time is fired in the double-slit experiment, a wave interference pattern still occurs.

Right. This shows that the interference pattern (from a continuous beam say) is not due to the photons interacting with each other.

But the intriguing thing is that even if you increase the rate, you still get the same pattern (up to a point).

Up to a point? What happens if you go beyond that point, other than the slits melting or something? Got a citation?


* ¹⁵O (with a half-life of a couple minutes) decaying into ¹⁵N is an example of an everyday non-violent end of a proton that might be observed in a lab here on Earth. A PET scanner apparently uses exactly this reaction to study oxygen / blood flow.

You said you're creating a new world
— noAxioms
I have never said this. — Luke

Not exactly in those words:

there must be two (or more) parallel timelines in order for time travel to make sense. The timelines branch off into two or more timelines following the first time travel event. — Luke

That says a parallel timeline [world] is needed, created since it doesn't otherwise exist. The 2nd sentence implies the 1990 new timeline branches off the 2024 'travel' event, which means no actual travel, just a universe creation event at 2024.
You say this, but perhaps worded it poorly and meant actual travel to an actually existing 1990, so it isn't something (a whole universe) that you need to manufacture.

my ontology of time involves a blend of presentism and eternalism (in short, that without presentism there is no 'progression of events', — Luke

OK, you acknowledge that the concept of a timeline implies the lack of presentism. There is no need for a 'progression of events'. Time travel under eternalism simply involves a worldline that is discontinuous, or doesn't follow a timelike path. So we ditch the presentism altogether, and that gives us a 1990 destination which we select as our target.

Now we do some retro-causality magic and branch a new timeline off of that point, which doesn't alter the original line at all except for Bob's abrupt disappearance from 2024, a violation of a bunch of conservation laws that we are ignoring. How is this Bob in the new timeline the same Bob as the old timeline? I mean, with the usual parallel timelines (MWI here), the Luke in some other world is not you, but somebody else.

There's no contradictions with it because killing the copy young-Bob isn't killing old-Bob's actual ancestor.
— noAxioms

You can call it a "copy" if you like. There are two parallel timelines, after all: one timeline in which Young Bob grows up to build a time machine and travel to 2024 and another in which Young Bob gets killed by Old Bob. — Luke

Sounds like a copy to me. Old Bob is a continuation of the not-murdered original young Bob, not the Bob that gets murdered.

However, what supposedly happens to Old Bob in your single timeline scenario after he murders Young Bob?

I don't have a single-timeline scenario. Heck, I don't have a scenario at all. Just trying to figure yours out. I've changed my guess significantly based on what you've said and based on some past comments that I read again. Is it better now?
What was our point here? If we can do this impossible thing, no contradiction need exist (except for the magic in doing it). I've long since expressed that the branching solution resolves the grandfather paradox.

This would be the only realistic result — AmadeusD

Sort of like Marty (or his picture of his older siblings) beginning to fade as he slowly destroys any possibility of his parents hooking up. Hollywood loves this idea despite the paradox it creates.

Could you explain why it must be a "new thing, a copy" of 1990 recreated in 2024 and why Old Bob cannot actually travel back to 1990? — Luke

No I can't. You won't let me discuss interpretations at all. You said you're creating a new world, not altering the original, in effort to avoid the paradox. That means an act of creation of a new world.
You said that 2024 is the antecedent state, so that means the alternate (copy) 1990 state was created at that time. It's all I have to work with. I see why the video says it needs a lot of energy.

The original 1990 doesn't exist anymore. You can't travel to somewhere that doesn't exist. You have to create it, a copy of it. I'm running with that assumption when trying to understand what you're suggesting.

There's no contradictions with it because killing the copy young-Bob isn't killing old-Bob's actual ancestor.

What precedes old Bob's appearance in 1990 is the use of the time machine in 2024 — Luke

It sounds like your machine doesn't travel at all then. It manufactures a new world in 2024 that looks like how things were in 1990. It's a new thing, a copy. The time is still 2024, but the calendar hung on the wall is set to 1990. Rather than going through the bother of putting a copy of old-Luke (and the machine) in this newly created world, it would save effort by just creating the world like it was but without young-Luke.

The original 2024 timeline marches on, without you and the machine if the universe-creation process involves the destruction of the machine and its occupant, and still with you if it doesn't involve that and only places a copy of you and it in the new world created.

Anyway, if you hand-wave away all the physical reasons why this cannot be done, I have no problem envisioning time-travel scenarios that are free of paradoxes.

Not sure if there was anyone witnessing the Big Bang — Corvus

We did. It's not like it happened a finite distance away and the view of the bang has already passed us by. Of course the really early events are obscured by the opaque conditions back then. The window through which we look took a third of a million years or so to turn transparent. By that measure, nothing could 'see' the big bang since it was all obscured behind a blanket until then.

Sure, I am not saying it is not allowed to have conjectures and hypothesis on time travel. My point was the claim that "If X, Y, Z, then time travel is possible." remains as a hypothesis until X, Y, Z had been proved as truths which complies to the objective facts in the actual world. — Corvus

Nothing ever gets proved. I can go to grandma's house if I have a car, and the weather is acceptable, and if I draw breath. But technically I cannot prove any of those.
Point is, requiring 'proof' is going to far. Evidence of X,Y Z is probably enough for plausible time travel. Right now, that evidence is very negative.

I am not sure what the physical clock measures. — Corvus

It measures proper time, which is very defined in both interpretations of time. It doesn't measure the advancement of the present, or the rate of the flow of time. That sort of time is more abstract, and there is no empirical way to detect it, let alone measure it. So maybe we're talking past each other when I reference the sort of time that clocks measure, vs you referencing the latter.

But one might still demand to prove the existence of the parallel time lines, before progressing further. — Corvus

Only if it is claimed that they necessarily must be. We're assuming them here to see if it makes time travel possible. It doesn't, but it does remove some of the issues and paradoxes.

From my own perspective, time doesn't exist. It is a mental concept. — Corvus

A physical clock measures something. Hard to deny the existence of something that can be measured.
You seem to get around this by defining time differently than, well than how physics defines it, which boils down to 'what a clock measures'. I agree that the coordinates we assign to time is pure abstraction.

I just transcribed most of the very short video. — Luke

Saved me from typing it. Most of the thanks was for that.

— Luke

As you may recall from previous discussions on time, my ontology of time involves a blend of presentism and eternalism (in short, that without presentism there is no 'progression of events', and without eternalism there is no timeline(s) of events). If eternalism solves a problem for time travel, that's great. — Luke

How about a growing block model then? The past exists. You can go to it, but since it is 'the past', you cannot change it. So a new branch is created (MWI style, but with physics violations), very much like your Bob story. I think that would satisfy both of us. The video presumes (I think) one would have to recreate the entire past state of the universe, hence the excessive energy required.

Oh, then we are in agreement and I've wasted my keystrokes. I thought the grandfather paradox indicated that time travel itself is paradoxical? — Luke

Wasn't wasted. Your Bob example showed how that paradox can be easily avoided.
Another way is to scratch the parallel world and let Bob simply destroy his younger self, and the time machine appears in 1990 uncaused. It's going to do that anyway (in violation of physics), but we're supposed to be ignoring known physics for this exercise.

OK, I said it wasn't paradoxical, but it's still a violation of the physics that we're ignoring. If sending information outside of the cause's future light cone constitutes a paradox, then its still a paradox.

And apparently Bob fails in his effort to destroy the bad thing resulting from his technology.
— noAxioms
Not with the spawning of a new, second timeline (once old Bob time travels back from 2024). — Luke

The old timeline still has the bad technology. It just doesn't have Bob anymore. If it's just Bob that's the problem, he could fix that quick without bothering to build the machine.

The antecedent state would be old Bob's time machine transporting him from 2024 to 1990 — Luke

No, the antecedent state would be 1990 minus 1 second. That cannot produce an old-Bob.
Either that or he didn't actually go to 1990, but simply rearranged the entire state of the 2024 universe to correspond to what it looked like in 1990, which seemingly is what the video envisioned.

Physics doesn't allow a vehicle to just materialize from nothing. But I'm told to ignore this inconvenient problem. Hollywood depicts it frequently, and they can't be wrong, right?

The problems associated with time travel cited in the video are as follows — Luke

Great summary, thanks. All packed into less than a minute to boot.

1. Time is not a physical object that can be moved or manipulated. It's simply a measurement of the progression of events.

The video author seems also to presume presentism, implying that time itself would have to be re-wound (and the entire universe with it) in order to 'go back', rather than time being left alone and just the traveler going somewhere.
In non-presentist terms, it would require a discontinuous, or non-timelike worldline. Well, the worldline is just an abstraction, so it being discontinuous is not in itself a problem Several people have proposed valid methods to do (forward) time travel utilizing discontinuous worldlines.
Anyway, I have a problem with number 1.

2. The laws of physics, including the laws of thermodynamics, make it impossible to go back in time.

Not at all. But it presumes a self-contradictory version of dual-presentism, that the universe causality is made to go backwards (less entropy) but real time continues to go forwards.
Fact is, my abrupt appearance in 1955 would not violate entropy laws at all, nor would it violate thermodynamic law. It makes a hash of causality, but that's not brought up in this point.

3. The idea of travelling back in time would violate the laws of causality, meaning that an effect cannot occur before its cause.

This one has teeth, but is worded wrong. Causality doesn't say an effect cannot occur before its cause, it says that the effect (information travel) cannot occur outside the future light cone of the cause. The future light cone is physical and objective (not frame dependent). The plane of simultaneity (referenced by the word 'before') is frame dependent and an abstraction, at least it is under Einstein's theory. It isn't under presentism of course, so that assumption yet again.
Anyway, yes, backwards (not forwards) time travel would violate causality laws, if they're valid laws.

4. Time travel raises numerous paradoxes, such as the grandfather paradox, in which travelling back in time and changing a past event would alter the present and create a contradiction.

Closed time loops are allowed under relativity, but like several other things, that doesn't mean there are any at a classical scale. Time travel isn't itself paradoxical.
Also, what did grandfather ever do to deserve this abuse? If you want to illustrate the paradox, go back 5 seconds and kill yourself, or otherwise disable the machine, which would probably happen anyway with a 2nd machine materializing right in the same place.

5. Even if time travel were possible, it would require immense energy and advanced technology beyond our current capabilities.[/quote]This is nonsense. 'If impossible thing, then [arbitrary unfalsifiable conclusion]'. The energy requirements are meaningless unless a method to do it is proposed.

Point 2 says that time travel is impossible due to the laws of physics

Mostly point 3 that actually says that, seemingly the only point that isn't straight up unbacked conjecture.

Late in 2024, Bob enters his time machine for the first time and sets course for the year 1990.

A nit: He has to set his course for an event, which has 4 coordinates, not just one. Pretty much all the fiction (except xkcd) seems to forget that. Everything moves, but it is always assumed that the machine will reappear at the same map-location as it left despite the motion of stars, planets, etc. OK, Dr Who doesn't work that way. It's a car, and it travels in space as much as time.

However, I will argue, there must be two (or more) parallel timelines in order for time travel to make sense. The timelines branch off into two or more timelines following the first time travel event. Let's call them timeline A and timeline B. Timeline B differs from timeline A only by the addition of the time traveller (and all that causally follows).

OK, the 'spawn a new timeline' explanation. Yes, that avoids the grandfather thing, but doesn't resolve the physics violation of the machine in the first place, in particular, what caused the 1990 state with two Bob's in it.

On the original timeline (A):
1980(A) - Bob(A) is born
1990(A) - Bob(A) has the inspirational idea for time travel technology
2024(A) - Bob(A) builds his time machine and travels to 1990
2025(A) onwards - the world continues on its course of the original timeline (A)

And apparently Bob fails in his effort to destroy the bad thing resulting from his technology.

On the second timeline (B):
1990(B) - Bob(B) arrives in his time machine.
...
However far-fetched this may seem, it does not violate causality and leads to no apparent contradictions.

Um, that's a blatant violation. 'Old Bob' in 1990 is not the result of an antecedent state. If 2024 is the antecedent state, then the rest of this new timeline is not the result of that other antecedent state.

They stay exactly 1 meter away from each other for eternity. Is there time? — Philosophim

Well, you said 'for eternity', which implies time. But I agree that there is no meaningful time without change of some sort. You speak of an observer, but observing at all cannot take place.

To further your example, the particles need not remain a fixed distance apart. In a 2-particle universe, the distance between them is meaningless, as is rotation, so there is no need to state that they remain a meter apart, or remain in a given orientation.

Lets say the particles are a little misshapen — Philosophim

Now you go too far. A misshapen particle is not just a particle, but a collection of them. A misshapen particle has extension, and if it has extension, the distance between the particles becomes a meaningful multiple of that extension.


As for the OP

It is still the case that, if the universe had a begining (Cosmic Inflation, the Big Bang) then there was a time T0 where things existed and did not exist in any prior state. — Count Timothy von Icarus

It is not necessarily the case since eternalism suggests no such thing, as your seem to realize:

As I understand it, this problem is one of the things that makes eternalism popular. It would seem to solve the problem by saying that only what exists, including all moments, exists without begining or end. I find eternalism as a whole problematic for reasons that would make this post too long, so this never appealed to me. — Count Timothy von Icarus

OK, so it's a problem for your chosen interpretation of time. Einstein's relativity theory assumes eternalism by assuming the existence of spacetime. There are alternative theories that are more along the lines you suggest, but those theories took almost another century to be fleshed out, being in denial of the big bang, black holes, and other things that fall out of relativity theory.

So yes, even the eternalists have a problem if they assert that the universe is real. How does one explain the reality of whatever one declares to be real? It is sort of the 'why is there something and not nothing?' question. You have to answer that as well, even given your chosen interpretation of time.

But the point where the moving wheel touches the road is not a fixed point on the rim — Agree-to-Disagree

Ah, gotcha, and that made me re-read the way the OP was worded, and I think your take is more correct than the way I saw the question being asked.

The answer varies with starting assumptions. — frank

Trying to figure out which starting assumption (unstated) would reach an one (but not all) the other answers. For instance, I could assume that circle B is rotating, or that circle A is slipping, but either assumption leads to any of the answers being possible.
I'm also from a physics background where rotation is absolute, not something that is frame dependent.

The video does explain that. — Wayfarer

I didn't watch the video, so I'm just commenting on what is shown in the still shot. Surely the makers of the test selected one of the five answers as being 'correct', and surely somebody must have guess that selection, either in ignorance or in realizing that the correct answer isn't an option. The title suggests that this answer is selected by nobody, which is implausible.

If you insist that you can travel into the past or future in your imagination — Corvus

I never said anything about imagining. The comment to which you are replying was a reference to your pressumption of presentism. LuckyR seems to presume it as well:

Time travel exists, but only to the future, never the past (since, as stated) there is no "past" to travel to. — LuckyR

I personally don't insist that there is no "past' to travel to. I give equal ontology to all of spacetime, not just one 3D slice of it. Reverse time travel (as typically envisioned) is not possible because it would constitute transfer of information outside of somebody's future light cone, something which relativity forbids, and something which has never been demonstrated .

Strictly speaking there is no tomorrow in reality.
...
There is only "Now" for the whole universe and its members.

Your opinion, not mine. "Tomorrow" is a relative reference, sort of like (one km to the east). There is no objective location that is 'one km to the east', but relative to any given reference location in a place where 'east' is meaningful, there is.

What you call tomorrow is in your imagination as a concept or idea.

I am not speaking as an idealist when I made the comment. To me, 'tomorrow' is just as real as 'one km east of here'. All of Einsteins theories presume the same, but it is admittedly a presumption. There are alternatives to his theories that don't make this presumption, but they came almost a century later.

It's fine to presume such things, but a topic about time travel seems to require that the 'destination' exists in order for you do deny your ability to get there. You can't argue that it can't be done because only the present exists, because there's no way to prove that opinion.

But the point where the moving wheel touches the road is not following the path of a cycloid. It is a point moving in a straight line at the same speed as the car is moving. — Agree-to-Disagree

All this is wrong. A point on the rim of a rigid not-slipping wheel IS folling the path of a cycloid (not well depicted in the drawing which shows the path coming in from an angle instead of vertically), and is very much is stationary relative to the road, not the car. The axle is moving at the speed of the car, and no point on the wheel is ever stationary relative to the axle while the car is moving.

Am I reading your comment wrong? It seems you're just asserting things that are obviously wrong.

A thought problem along similar lines (or curves): "The SAT Question that Everyone Got Wrong" — wonderer1

The reason everyone gets it wrong is because the correct answer (4) isn't one of the options. It isn't because nobody can figure it out correctly. I've had that problem on a different test (not multiple choice) and got it right, as did a fair percentage of others.

It would be recreating the past in the future. — LuckyR

Recreating a piece of some past state. Indeed, this isn't time travel being described.
I can build a new 1928 Studebaker, even giving it the same serial number as one made in that year. Has that car time traveled or is it just a new thing? I satisfied the conditions of the OP by doing so. Is it even a Studebaker if I built it instead of the defunct company?

There's several physics violations made by the OP, mostly that the state of some system can be fully measured (violating Heisenberg uncertainty), counterfactuals, and violations of entropy, the latter of which can be fixed by recreating a finite state, just a system in a box, not everything.

The question for the OP scenario then becomes, how is the thing created in the box you? Suppose you recreate a 'you' while you're still around, outside the box. Clearly the created thing in the box is not you. Nothing has 'traveled' into the future.

That said, yes it has. If you look at the state of Earth in 1990, you'll find me there. Have I time traveled to 1990? It seems I have. The statement presumes there is a 1990 in existence to examine, which pretty much means presentism is abandoned, which says only today exists.

You cannot travel into a place where the destination doesn't exist. We are all nailed into the present until deaths under the universal law. — Corvus

This is what I mean. Corvus seems to assume presentism with this statement. The whole notion of time travel seems to assume otherwise, that there are 'other times' available as valid destinations.

Impossibility of time travel seems to be one of the universally necessary truth. — Corvus

Funny that you will nevertheless travel to tomorrow. I plan to see you there.
People talk about time dilation being time travel. It isn't any different than doing the same thing sitting still. You get to 'the future' either way, assuming you live long enough to get to the target destination.

On a different note, closed time loops are valid solutions to Einstein's field equations. So are tachyons, and nobody has any reason to suspect either actually exists anywhere, but it isn't mathematically impossible. They don't for instance violate things like the grandfather paradox, and don't require branching of timelines.

No it doesn't. See my other argument in that post: — Michael

OK, we're at an impasse. I did see the argument, and it begs, and you don't see that. We can both just repeat our stances forever.

1. A foo is a four-sided triangle
2. Foos are a metaphysical impossibility
3. Therefore triangles, if they exist, do not have four sides — Michael

This is yet again a begging argument. The whole purpose of the argument is to somehow determine how many sides a triangle has, which means we need to start from an agnostic position of not knowing. You don't do that. Step 2 says that Foos are impossible, which you cannot demonstrate unless you beg that triangles have something other than 4 sides.

The analogy is also poor since the Foo and the not-4-sided triangle are not physically identical.

The other issue isn't a begging one, but seems to hinge on another sort of fallacy.

We can conceive of something that is physically identical to us not having consciousness, therefore it is metaphysically possible for something physically identical to us to not have consciousness, therefore consciousness isn't physical. — Michael

So I can conceive of a universe that is physically identical to ours, except momentum being conserved isn't the result of supernatural intervention. A rock, in the absence of an external force acting on it, could continue at its velocity indefinitely without help from the supernatural entity carrying it, or however that works. Therefore it is metaphysically possible for something physically identical to us to not require the magic, therefore momentum isn't physical.

That I do believe leverages the same logic, but I don't know the name of the fallacy. Never mind the fact that I also leveraged the begging of the magic. This example was meant to point out the fault in concluding that momentum (or anything of your choice like say mass or a clock running) isn't physical.

Step 1 doesn't define consciousness. — Michael

It doesn't define what it is, but it blatantly defines it to be something not physical. You've not refuted this in any way.

Here's a different argument:

1. A p-zombie is physically identical to us but has no consciousness

No different than before. This is the same statement, stating right up front that consciousness is non-physical.

2. P-zombies are a metaphysical impossibility
3. Therefore consciousness, if it exists, is physical

3 doesn't follow from the prior statement. 1 asserts that consciousness exists, so 3 cannot say 'if it exists'. 3 should read 'consciousness exists, and is not physical'. It follows directly from 1 and line 2 is superfluous.

The argument is valid.

It isn't. 3 directly contradicts 1, regardless of the actual nature of consciousness.

Line 1 is just a definition. — Michael

A begging definition then.
1. A p-zombie is physically identical to us but has no consciousness[/quote]

Physically identical implies that the difference is non-physical. 'We' have something non-physical that the physically identical zombie doesn't. That's very much begging the conclusion. I mean the conclusion is drawn by step 3 without any additional unreasonable premises.Step 1 defines consciousness to be supernatural, and step 5 asserts that we have it, per that definition. How is this possibly not begging?

So I attempted to interpret this in a non-begging way, allowing consciousness to be a physical process. In that case the p-zombie would not by physically identical, but rather some physical difference rendering him unconscious,and an unconscious person does not plausibly behave like a conscious one. It would be like asserting that you with all your senses cut off (not to mention voluntary motor control), would still be able to function without anybody noticing the difference.
Under physicalism, both the conscious and unconscious people would behaves as their physical circumstances dictate. Somebody behaving as his physical circumstances dictate does not imply that he is not conscious, at least not until the argument is accepted, but it is fallacious.

An unconscious being is not a metaphysical impossibility, but I do notice that nothing in the 6 points mentions the fact that the zombie behaves like the conscious one, with no way to detect the difference. It doesn't take a medical professional to detect the difference between a conscious and unconscious person. They're in different physical states, so they might be physically identical except for whatever states render him unconscious, such as say sleep or anesthesia.

My apologies for replying to days-old posts, but I've been otherwise occupied.

The argument is that:
1. A p-zombie is physically identical to us but has no consciousness
2. P-zombies are not a metaphysical impossibility
3. Therefore consciousness, if it exists, is non-physical
4. Therefore either physicalism is false or nothing is conscious
5. We are conscious
6. Therefore physicalism is false — Michael

This argument seems to depend on consciousness having zero benefit or purpose. It would never have been selected for since it brings zero benefit. The argument makes somewhat more sense if one is in denial of evolution of course.
The p-zombie can function identically without the consciousness (as I do). It reports the same experience (except I don't in this conversation because I've been stripped of the vocabulary necessary to do so). So line 3 should be "Therefore consciousness, if it exists, is undetectable". It seems to be more of an argument for epiphenomenalism.
Lines 1 and 5 beg the conclusion, making the argument fallacious. I claim I am not 'conscious'. I would be lying if I said I was.

If physicalism was true, the non-conscious being would make the exact same argument as above, per line 1, demonstrating that the argument carries no weight.

We can conceive of something that is physically identical to us not having consciousness — Michael

This also begs the conclusion.

External stimuli such as light and sound stimulate its sense receptors, these signals are sent to the brain which then responds by sending signals to the muscles causing it to move in the manner appropriate to navigate the stimulus. — Michael

How do you know that this isn't a description about how you work? I mean if it was, then by definition you would not know it, so I guess I am asking how you would report that you know that this isn't a description of how you work.

P-zombies have no free will. Everything they do is a physical effect of prior physical causes. — Michael

How do you know that you have this sort of free will? Given many definitions of free will (that your choices are not the result of physical causes), I agree with your argument above. But then this zombie has no idea why anyone would benefit from that sort of free will. It sounds like a curse.

I think they're impossible too. — flannel jesus
Impossible because conscious experience is physical ... — Michael

By your definition it cannot be. You've made that very clear.

I wouldn't say impossible, but it's ludicrous to think there would be a couple of p-zombies carrying on, what to us would appear to be a deeply personal heartfelt conversation, while in fact their conversation is simply meaningless noises they are making for no reason. — wonderer1

Only because the language forbids using half those words for what the zombie is doing. It very much claims 'heartfelt', 'meaningful', etc, but they're apparently all lies. The zombies doesn't know that they're lies.

I am simply explaining that “I believe that I am a p-zombie” is false if he is a p-zombie and irrational if he’s not. — Michael

On that note, I present this:


One of these is a duck. It swims, quacks, and avoids predators. The other is a p-zuck, which does none of that by definition. It simply behaves as its physical circumstances dictate. No more.

It is impossible by any test to tell which is which.
The duck quacks, and reacts to a quack noise by another. The physical circumstances of the p-zuck dictate that it produces a series of air vibrations that the duck interprets as a quack, but the 'quack' statement as understood by the duck is false because that which produced the vibrations does not quack.
The duck also is not sure if it is itself a p-zuck since it cannot tell the difference and as far as the duck can tell by any test imaginable, it also behaves as its physical circumstances dictate. So it's potential conclusion either way is not irrational. It truly doesn't know. Neither does the p-zuck, but what it is doing is not 'concluding' or 'believing' and it cannot communicate its conundrum because it has no available language by which it can express what its physical circumstances are dictating.

I would say, if your claim were true, it would be a revolutionary finding — hypericin

It wouldn't be a finding at all. If it was true, nobody (not even I) would know for sure. Of course, I'm sure that 'knowing' things (and being 'sure') are all forbidden. But I do have whatever it takes to pass an interview for a technical job, even if it isn't knowledge. I have claims of it on my resume, all false apparently.

Apologies to RogueAI for somewhat hijacking this topic. It's about p-zombies still, but not much about having children. I have no way of knowing if my kids are conscious or not. Not sure if a new topic would cover any ground not already covered.

You could just say “I am a p-zombie”. — Michael

Most of us don't know it. It isn't easy to tell until you get a glimpse of the bit being missed, sort of like having your vision restored after cataracts have reduced you to near grayscale levels.

You’ll need to explain it in these terms. — Michael

Hence my attempt with the car, which very much is aware of its surroundings, but 'aware' is perhaps one of those forbidden words. It all smacks of racism. They basically degraded black slaves by refusing to use human terms for anything related to them, using cattle terms instead,.It made it easy to justify how they were treated. "Cows don't feel pain. Neither do p-zombies. It's not immoral to set em on fire."
I'm sorry, just because we're not conscious in the Chalmer way doesn't mean we don't hold beliefs. I refuse to withhold perfectly understandable terms when no alternatives are offered.

I don't agree with this. "so he says he feels pain, not knowing that it isn't real pain". That's an epistemic issue, not a truth issue. For any x, if x is not feeling pain/hurting, and x says it is feeling pain/hurting, x is wrong. X is saying something false. — RogueAI

I didn't say the statement "so he says he feels pain, not knowing that it isn't real pain" was true.
I said the statement, "pain hurts" is true regardless of what utters it.

But this isn't true for you. — RogueAI

That's just using a language bias to attempt a demonstration of a difference where there isn't one. A Roomba cannot be conscious because you define 'conscious' to only apply to humans. That doesn't demonstrate that a pimped-out Roomba isn't doing the exact same thing, it only means that the Roomba needs to pick a different word for the exact same thing, and then tell the human that he isn't that sort of conscious because he's an inferior human.

That's the sort of argument I see coming from everybody. The word is not legally applied to you, therefore you you're not doing what the chosen race is doing.

But belief is a conscious mental activity. P-belief/p-consider is incoherent. It's missing a necessary condition for anything that remotely resembles believing and considering. — RogueAI

What is the computer doing then when it processes data from a camera pointed at a table. The computer 'concludes' (probably a forbidden word) that there is a table in front of the camera in question, and outputs a statement "there seems to be a table in front of the camera". You say it's not a mental activity. I agree with that. That usage of "mental activity" only applies to an immaterial mind such as Chalmers envisions. So OK, you can't express that the computer believes there's a table there, or that it concludes that. How do you phrase what the computer does when it does the exact same thing as the human, which is deduce (presumably another forbidden word) the nature of the object in the field of view.
If you can't provide acceptable alternative terms, then I'm sorry, the computer believes there's a table there. Deal with it.

The answer is that it would behave as its physical circumstances dictate. — Banno

So I claim I'm doing. But how would you express what the p-zombie does when it correctly identifies the table in front of it that it cannot 'see'?

We don't behave this way — RogueAI

I do, and you don't. So why are we indistinguishable (except for me deciding to stop imitating the language you use for that which I cannot ever know)?

I don't have a computer algorithm in my head. My head does not implement a Von-Neumann architecture, even though I'm capable of simulating one, and a Von-Neumann machine is capable of simulating me. None of that is true of you. The lights are out for me, but I don't know it since I've never experienced the light and had that 'ooooohh' moment.

By definition they behave as we do. This includes belief. — hypericin

Well, per Michael, this includes false claims of belief. I'm doing something that I think is belief, but Michael says it's by definition false.

When humans say "pain hurts" it's true. When a p-zombie says it, it's not. — RogueAI

No, it's true by definition, regardless of what says it. The p-zombie might not feel actual pain, but says he does anyway since he very much detects the undesirable sense of injury, and he has no actual reference to what true pain feels like, so he says he feels pain, not knowing that it isn't real pain, just an interpretation of sensory data.

And, I'm not so sure a p-zombie would say that. Why would they lie?[/quote]Lies are intentional. The p-zombies doesn't know. I only suspect because there's something missing, something many others find obvious, but not all.

Wouldn't they just be confused about what "hurting" is?

You're definitely confusing me when y'all say there's a whole vocabulary that I'm not allowed to use, and without giving me replacement words. So I use the words.

I assume you are telling the truth when you say "pain hurts"

It is truth, but 'hurts' to me is just detection of signals of injury. It's not like I lie and don't actually get this sensory input. But the extra bit, that which I would be totally unaware except for people talking about how obvious and inexplicable it is, only the talk of that makes me aware of something more that should be there.

Perhaps they have desires and urges we're not aware of. — RogueAI

I would hope so. They invented sex after all.

Does pain hurt? Does it feel bad? — RogueAI

Yes, but any p-zombie or human would say that. It's not a question that distinguishes the two cases. I've been taught that 'hurt' and 'feel bad' are appropriate ways to express the state of my information processor when it detects signals originating from nociceptors. Most self-driving cars don't have these, so in that sense, the car is a poor example.

Not sure what having children has to do with being conscious. Things have been having children since eukaryotes, which arguably are not particularly conscious.

1. “I consider myself to be a p-zombie” is false because you are a p-zombie and so don’t believe anything. — Michael

OK, 'belief' is one of those things reserved. It is not appropriate to say that a self-driving car believes that steering onto the soft shoulder at speed would be a poor choice. Different language must be used.
All that said, the statement "I consider myself to be..." does not use the word 'believe'. If you equate them, then you really need to pony up a generic word that applies to cars and such which consider some judgement to be 'true'.
But a p-zombie making the same statement would be one of truth, even if the phrasing is designated to not be allowed by you.

The statement “I consider myself to be a p-zombie” is only true if you are not a p-zombie and so no rational person can believe themselves to be a p-zombie.

OK, so it's false, only because the actual p-zombie is not allowed to use the phrasing. The p-zombie differs on that ruling.

It is a thought experiment, it is an open question whether it is believable or not. — hypericin

Belief in it is critical to the argument. The p-zombie apparently isn't allowed to 'believe', so there's seemingly no position from which an actual p-zombie can argue his case.

Chalmers is full of descriptions of all this miraculous stuff that isn't physically possible. It's so obvious to the people actually conscious (his definition), leaving people like me wondering what the miracle is. There's nothing inexplicable. No hard problem, so the conclusion is that Chalmers experiences something I don't, something that cannot be conveyed to me, which is sort of like trying to convey the experience of seeing red to Mary. I can pick red out of a box of crayons by sight, as can a robot. Mary was never given a chance to, so I'm not sure if the analogy is apt.

I cannot think of anything I can do that a machine cannot. Chalmers has access to something that obviously is out of reach of the machine. The conclusion is that I'm missing what he experiences.

Are you conscious?. — hypericin

I react to data from my senses. So yes, I am conscious in the same way that a self-driving car is conscious of the traffic around it.
No, not conscious in the Chalmers sense, something more, something that machines cannot have. I've looked for the 'more' part, the part that is inexplicable, and find only the automaton.

I find it very hard to believe. — hypericin

Then the P-zombie argument falls flat because it is unbelievable that something could behave identically externally without that extra thing on the inside. The argument hinges on not being able to tell. So you must believe.

Do p-zombies ever look like they're in pain, or sad, or happy? — RogueAI

They'd act very different if they didn't look like that. As said in terminator, I sense injuries. The data could be called "pain.". I react in a way that attempts to minimize that pain, sometimes quite irrationally. The face expressions? Those seem to come from subconscious places to which I have no direct access. A sleeping (unconscious) person will still wince in pain given certain stimuli.

P-zombies don't have urges or emotions, so why would they reproduce? — RogueAI

They're supposed to be indistinguishable, so yes, they would display urges, emotions, and children. I consider myself to be a p-zombie. OK, my kids are a bit off the curve, but I display emotions and talk about qualia, mostly because I learned the language from you non-zombies, not because there's that inexplicable extra bit that seemingly defies 'physical explanation'.

Parfit, Personal Identity wonderfully covers the topic from a metaphysical point of view, but it seems you're looking more for sources discussing the biological origins of one's sense of identity.

Please show how "all points on a two dimensional surface that are equidistant from the center"
and these exact same points form four straight sides of equal length in the same two dementional plane. — PL Olcott

First you've mentioned 'straight sides'. Also first you've mentioned a 2D plane. This comes back to my point, which you predictably have totally missed besides it being emphasized multiple times: State your premises, because the assessment of 'impossible' or not depends on them.

I can still do it despite these new definitions.

On the surface of the Earth, — magritte

Tip O' the hat to magritte, who actually paid attention to my point, and similarly did not see a restriction to Euclidean space, which is interesting because our universe is not Euclidean, so it's not an obvious assumption to make. One can draw a square circle on the 2D closed plane of the surface of (an idealized) Earth. It would have four equal length straight sides.

changing the subject to the not logically impossible — PL Olcott

I think only a mod can change the subject of a topic, so it is safe.
So yes, just be more careful when selecting your logically impossible thing, because so far none of them has qualified. Then, having found some suitable impossible requirement, in order to make your point in the subject line, you need to come up with a scenario that lists it as a requirement. That was also missing in the OP. So let's say for argument sake that a square circle (suitably defined) is impossible. What argument would plausibly list that as a requirement?

It is logically impossible to draw a square circle because it must be perfectly round AND not round at all with four equal length sides. — PL Olcott

Not logically impossible. I suppose it depends on your definitions, and 'is perfectly round' is a poor definition of a circle. More like 'all points on a two dimensional surface that are equidistant from the center. Given that and 'four equal sides' (and maybe four equal angles to eliminate a rhombus) as a square, it is not logically impossible.
Relativity theory ran into that. It proposed something logically impossible, except to somebody who is willing to drop some obvious but unstated premises.

It is clear that the impossibility of creating a CAD system that can correctly draw square circles places no limits on what computers can do.

The inability to do an impossible thing isn't a limit? What is 'impossible' if not a limit?

It is less clear that requiring a program H to report on the behavior of another program D that does the opposite of whatever H says is a logical impossibility when we see that program H1 can correctly say what D will do.

Godel showed that the program H cannot solve the task to which it is put. You seem to know this since you reference the halting problem. I was just discussing this very thing in another thread.

So the when the {halting problem} requires a program H to always say whatever program D will do includes programs that do the opposite of whatever H says this is requiring the logically impossible, thus the same as requiring a CAD system to correctly draw square circles.

Again, I don't find it logically impossible. All you need is to prevent access of H to the inputs of D. This again illustrates the fact that you've presuming conditions which have not explicitly been stated, same as the square circle.

I am talking about creating a perfectly round thing that
cannot be round because it has four equal length sides,
thus a logically impossible square circle. — PL Olcott

Agree, squaring the circle is an exercise in what can and cannot be constructed with a compass and straight edge. But the thing you list as impossible can be done, even without drawing a rhombus, which fits your definition of a square.

Just cut the circle (without moving anything) into four 90 degree arcs. Make them a different color if that helps to visualize it. There you go. Perfectly round figure that is also a figure consisting of four equal length sides. That was really trivial. Point is (the important part): Don't be so hasty to declare something impossible.

First, the robots that beat humans at rock, paper, scissors 100% of the time essentially cheat (that is they don't have any "thinking" algorithm) — LuckyR

I disagree about it not thinking, but yes, it leverages its far faster reaction time to see what the human hand is going for and bases its choice on that. For that to work, it must think, and very quickly. It would be instructive to put the robot in front of a mirror.
It takes far less thinking to just pick a seemingly random choice, and quite a bit more to learn its opponent's habits and begin to anticipate subsequent moves.

As to robot vs robot, since there is no "data" to enter into the algorithm to truly predict what the other robot will choose, the programmer has used some other way to come up with a choice.

Well, the data is that it knows exactly how the other robot is programmed, and thus it knows that its choice is determined. But the programmer knows the task to be mathematically impossible, leaving him with nothing to attempt the task.

If one knows what the algorithm is, one can reproduce it.

But the robot does know its opponent's algorithm, and is nevertheless incapable of predicting its outcome. That's the point I'm trying to make: illustration of the difference between determinism and predictability. You seem to see it.

As to pondering leading to different choices with the same input, I agree with you that humans commonly use the same analysis based on memories, emotions, objective variables such as price etc, however the priotization of the numerous variables leading to different choices in essentially identical situations is a common human experience.

That it is, but the antecedent states are never the same. Something critical is different when a different choice is made. Humans cannot be aware of that because they're never get to do anything twice. An antecedent state is never repeated. So the common human experience is of not always choosing the same thing, which, coupled with a naive assessment of identical antecedent state, results in what others have called 'the illusion of freedom'.in choice. Perhaps this is what you're getting at. The 'other factors' is simply antecedent states that are not identical. A huge factor is simply memory of the last time this choice came up, and wanting to not do the same thing all the time, hence picking a different ice cream today because you remember vanilla from last time.

If what you imagine is going on in the Black Box of human decision making was actually true, when faced with a decision between two choices of equal merit, humans would be unable to make a choice, yet we do every day.

Humans don't freeze in that scenario. It's called a metastable state, and we have a very fast mechanism to break such states, as is necessary for survival.
It comes up frequently in electronics, say a race between the Jeopardy buttons to indicate the one contenstant that pushed his button first. What if it's a tie? The circuit cannot freeze nor can it pick both. A metastability breaker is put in to resolve the condition. Mathematically, the task is impossible, but the resolution hammers the probability of continued metastability down to super low probabilities.

Maybe you can guess what sort of field I work in based on some of that.

Well the programmer of the robot can predict with 100% accuracy what his robot is going to choose. — LuckyR

Can he now? Maybe if the program is really trivial, making no attempt to work out what the other robot would do. Otherwise, you need to back that statement up.

He can of course just do a test run and see what it outputs. Given identical antecedent states and a deterministic algorithm (both were specified), then it has to do that each time. But that isn't the programmer 'predicting' anything. It's past tense at that point, not a prediction of the future, but it is a prediction in terms of future runs, that it must output the same thing as all prior runs.

The machine of course cannot win at rock paper scissors. It by definition has to fail despite the fact that we programmed it to work out what the other robot will pick (paper say) and choose a winning output (scissors) instead. Despite it knowing the antecedent state of its totally deterministic opponent, it is incapable of making that prediction. Hence the story illustrating the difference between determinism and predictability.


For the record, they have robots that they trot out on talk shows and such, that play rock paper scissors against humans. It has a hand as its output. Each iteration effectively starts at the same 'antecedent brain state' and are fully deterministic. They win every time, but they never pit one against a similar robot, at least not on camera. It is entertaining as a human to try to beat it. Are humans so lacking in free will that a simple robot can predict you every time? No, it doesn't work that way. It's more of a illustration as to why machines are better at some tasks than are humans.

Unless his program notes (accurately) that the three options are equally probable and therefore it chooses randomly.

These are deterministic algorithms. A computer has no instruction for randomness, so no, there is no equal probability. Neither do you have such a randomness amplifier, even though creatures would have evolved one had there been any survival benefit to it.

As to the factor, lay persons call it thinking" or "choosing". I'm not much into labels.

I think the word 'pondering' went by. Anything (human, machine, whatever) can do that much even under hard determinism, so thinking" or "choosing" only lets you make a better choice. It isn't what is going to make it possible for an antecedent state leading to many possible resultant states. You seem to conflate a good choice with a free one. Your choice is not free by the definition you give.

Again, there are very few options available. One is that multiple resultant states are chosen, and then a truly random selection is made between them. So held the feeble old lady across the street, or take advantage of the feebleness and snatch her purse. You, being a good person, want to help her, but you also want the choice to be free, so you 'roll the dice' instead, and it picks 'mug her' this time (but not every time). So glad you have that free will.

The other option is possession. Your body/brain wants to mug her, but a supernatural demon has possessed you and makes your body (in violation of natural law) help her across the street instead. The demon is now responsible for the actions of your body, and not your own criminal brain, which is safely shoved into some kind of epiphenomenal state, lacking the free will to act out its evil intent. For this to work, somewhere natural law has to be violated, severing the intent of the brain, and instead reacting to something with no natural cause. It should be pretty easy to spot such an organ.

I’d love to hear the details of this trivial experiment. — LuckyR

You have a machine that is entirely deterministic and classical: It executes instructions, and its purpose is to win rock-paper-scissors. Make it as simple or complex, slow or fast as you want. Its opponent is an identical robot, and both of them know that. The way the program works is to predict what the other robot will do (something that is completely determined ahead of time) and play the output that beats that.
That's a demonstration of deterministic vs predictable.

If a Determinist can’t use his Determinism to predict outcomes, what is the practical value of this Determinism?

Who ever claimed that the view had a practical value? I suppose its value lies in the fact that true randomness doesn't come into play, a priority that ranks high in some people's opinion, notably Einstein who was quite vocal about his distaste of randomness (and non-locality).

You are still evading my question, trying to steer things to another track.

How can you believe in free will when you cannot identify the 'other factors' that allow more than one potential outcome to a given antecedent state? Or is this one of those faith things where you believe in something that you think sounds desirable when in fact you have no evidence for it? (for both having it, and for it being a good thing to have)