Actually, the bottom line is that if we ever invent a time machine, it is our moral duty to find the creator of Hawaiian Pizza and murder him in cold blood...
I liked this post, but I didn't find any serious objections to nitpick with, which is kind of a bummer. My assumption in time-traveling thought experiments is generally something like this idea in Quantum Mechanics (is it the Multiverse thesis?) that every indeterminate state whose wave function collapses actually generates two universes, one in which we got state A and one in which we got state B. I don't think time travel is possible, but if it were, I'd suppose it will create two timelines: the one you departed from (which, if you were to return, would be completely unchanged in present and past) and a new one, in which the changes you effected have become the unavoidable past that actually happened.
I also agree that our attachment to those 'impossible to avoid' scenarios doesn't seem very rational, but is probably deeply imbued with cultural , religious and intellectual traditions (Greek Tragedy! An omnipotent God and his Divine Providence). In the killing-Hitler scenario, I imagine some parts of the past would change very little (there'd still be a 1st World War, a Russian Revolution and the Great Depression), while others would change a range of probability in their changes (given a defeated, mostly undemocratic, vindictive and militaristic Germany, it is likely that something like an authoritarian regime would have toppled the Weimar Republic and engaged in at least *some* warring). Historians love to debate the importance of individual agency in history, with periodic pendulum swings from Great Men to Geological-like Great Processes. Some specific individuals do have a disproportionate amount of influence on events, but they are exceptional, and I'd guess even them have an upper bound.
What does this view imply about the grandfather paradox?
My objection to killing Hitler is that the future you create will be one whose past is no longer compatible with you going back in time to kill Hitler; after all, there's no more Hitler to kill!
It's unclear how you resolve this: does the "new" timeline overwrite the "old"? What notion of time do the terms "new" and "old" refer to?
What will your personal memories consist in? Will you remember that you killed Hitler? Will you remember the old world in which WWII happened? Or will your memories change so that you always remember a Hitler-less world? When will that change happen? Will you suddenly come to beside a dead baby in 1889 with no memory of getting into a time machine?
It seems like this point of view leads to too many inconsistencies with no obvious resolution.
Hey! You ask "What does this view imply about the grandfather paradox?" The answer is: Absolutely nothing! I'm not making any claims about whether any kind of time travel is coherent or possible. I want to sidestep all that. I'm just using fictional time travel as a way to illustrate what the thesis of causal determinism says and what it doesn't say!
Hm, okay. Well, I've added a couple of short sentences now make clear that I'm not making any claims about the coherence of possibility of time travel. Not sure if that'll make any difference for readers, but hopefully it does!
I totally agree with this view. And unlike personal identity I even agree with you that this one isn’t depressing at all! In fact randomness would be somewhat more depressing because it would mean we aren’t control. Determinism means we always do what we decide to do. It’s just that who we would be and thus what we would do was predetermined. But that’s good for us because it would have been bad if we had not been predetermined to exist.
I do think quantum mechanics might make determinism slightly false. I don’t think it matters much on normal timeframes, but it could butterfly effect over long time frames.
Hey thanks for the comment! And yeah, quantum mechanics might throw a wrench in the theory, but just barely.
You probably know this better than I do, but the jury's out sbout how exactly to interpret quantum mechanics and some interpretations are deterministic. Nonetheless, some aren't, including, from what I understand, the most common interpretation.
For this reason, some philosophers will talk about near-determinism: determinism at the macro-level but not the micro-level. I omitted all this for simplicity's sake, but if you just substitute "near-determinism" for "determinism" the whole way through the piece, everything works out. Maybe I'd better add a footnote.
I wouldn’t say I’m an expert. My understanding would be that we know for sure there are things we can’t predict, so it seems like randomness. And then Bell’s Theorem, which is now conclusive basically, rules out local hidden variables that could explain this randomness. There could still be non-local hidden variables. I don’t know much about that but it seems very strange, though perhaps no more strange than quantum mechanics as a whole or the very idea of true randomness.
I’ve had a bit of a discussion about all this with a couple of people who know a lot more about physics than I do (Misha and Glen) on Twitter. From what I gather based on that, what I’ve said about deterministic interpretations of quantum mechanics is correct, but again, definitely out of my depth here. Anyway, if it interests you, the conversation is under my post about this piece!
Yeah I think everything is consistent. I’m 100% sure general relativity does not conflict with determinism at all so Misha is correct and Glen is off base there.
If someone goes back in time to kill him, then someone came from the future to kill him in his own time. Why would he have done that, since it would have been impossible for him to do what he did because he would have been dead?
You cannot change the past.
Again if he was killed at that time, there would not have been the future we know.
I feel like the article could be easily clarified by making it unambiguous what you mean by “change the future” in the intro, and by dropping the focus on causal determinism as a theory about events being fixed in the present in the second section.
With regards to the former, for example, when I read your assertion that causal determinism (CD) is compatible with the notion that we can change the future, given time travel, I interpreted that to mean that on CD we can change the future given everything that happened in the past (including your time travel shenanigans) and present. That, in essence, if I sort of knew what I did in the past before stepping into the time travel machine, I could go back in time and take a different approach to get a different future outcome. Which of course is both false and not your intended meaning.
After all, when we speak of change, we must mean a modification to some baseline state, but what is that baseline state if not a future predicted/determined by the events I knew of prior to my time travel shenanigans?
And regarding your definition of CD being about the present, first that’s not quite how philosophers and others usually define the term. Typically, causal determinism is interpreted as thesis about every event (past, present, and future) being determined.
Second, it confusingly and unintentionally implies that you are somehow contrasting causal determinism and fatalism by suggesting the former is only focused on the present (and not the past or future) while the latter is expanded to encompass the future (this was doubly puzzling for me, given my previous misunderstanding of your intended meaning of “change the future”). But that is both incorrect, and neither Dennett’s nor your intended meanings. So (to me at least) this created unintentional confusion.
Hey, thanks for the comment, but I'm a little confused about a few things you say. Instead of addressing all of them in one shot, I'll just address the main one for now.
You consider the possibility that "if I sort of knew what I did in the past before stepping into the time travel machine, I could go back in time and take a different approach to get a different future outcome."
You then say that that is "both false and not [my] intended meaning." But unless I've misunderstood what you're saying, that is indeed my intended meaning and I think it's true! Could you clarify?
Oops, I think I skimmed the baby Hitler section too fast! In any case, I interpreted the baby Hitler example to be a case of time travel where your previous time travel actions did not occur in the past (so more like alternate reality travel). Let me know if that’s not what you meant, and whether you think that even with every event in the past, including what exactly you did in the past as a time traveler, being fixed- you can still change the future. I indeed assumed (erroneously I take it?) in my first comment that the answer would be no. Feel free to let me know if I made any other misunderstandings! Apologies.
To be clear, my position is that
A) Real causal determinism is the position that all events (past, present and future) are determined, given a set of fixed initial conditions in the distant past and the laws of nature- these events cannot be changed. We are destined to do something, given all the temporal events in the past and present.
B) Nonetheless what we do makes a ‘difference’, in the trivial sense that if our actions had been different, by definition the events in the past, present and future would be different. So the future would indeed be different, but it’s trivial in the sense that a change in the past is required.
D) This is also Dennett’s position (see in particular his comments on pg.565 of the article you linked, on us being destined to do something on determinism. He admits this is true, but argues it is irrelevant for free will).
I assumed in my first comment that you agreed to A-D, but I’m not sure if that’s the case anymore. Sorry for the confusion.
Hey, no problem! Yes, I agree to A-D. I’d hesitate to use the word ‘destined’ myself, if only because it sounds so fatalistic: Some force (which??) has decreed that this must happen and will pull some strings behind the scenes to make sure it does. But of course, this is merely a terminological point. I know that’s not what you mean.
Now as for your first question: What I say is that I want to avoid committing myself to any particular model of time travel. Now of course, if what you have is a closed causal loop of the kind depicted in 12 Monkeys or Prisoner of Azkaban, then you can’t change anything, since the timeline included the time travel the whole time.
But that kind of model is incompatible with my claim that, by traveling back to the past, you change something about the past state of things and so should be able also to change the future. Granted, I don’t make this clear enough. Maybe I should rework that bit in the interest of clarity. At the very least, maybe it’d be a good idea to add a footnote.
But there are all kinds of models (none of which make any sense, of course), where there’s a single timeline and yet one travels back to the past, thus introducing oneself into the single timelines. There are also ones that create new timelines. Either of these models work with what I’m claiming (though in the latter case, what you’d be changing is the new timeline, not the original).
Hope this clears things up! The main lesson in all this is: Yeah, time travel’s a huge headache!
Actually, the bottom line is that if we ever invent a time machine, it is our moral duty to find the creator of Hawaiian Pizza and murder him in cold blood...
I liked this post, but I didn't find any serious objections to nitpick with, which is kind of a bummer. My assumption in time-traveling thought experiments is generally something like this idea in Quantum Mechanics (is it the Multiverse thesis?) that every indeterminate state whose wave function collapses actually generates two universes, one in which we got state A and one in which we got state B. I don't think time travel is possible, but if it were, I'd suppose it will create two timelines: the one you departed from (which, if you were to return, would be completely unchanged in present and past) and a new one, in which the changes you effected have become the unavoidable past that actually happened.
I also agree that our attachment to those 'impossible to avoid' scenarios doesn't seem very rational, but is probably deeply imbued with cultural , religious and intellectual traditions (Greek Tragedy! An omnipotent God and his Divine Providence). In the killing-Hitler scenario, I imagine some parts of the past would change very little (there'd still be a 1st World War, a Russian Revolution and the Great Depression), while others would change a range of probability in their changes (given a defeated, mostly undemocratic, vindictive and militaristic Germany, it is likely that something like an authoritarian regime would have toppled the Weimar Republic and engaged in at least *some* warring). Historians love to debate the importance of individual agency in history, with periodic pendulum swings from Great Men to Geological-like Great Processes. Some specific individuals do have a disproportionate amount of influence on events, but they are exceptional, and I'd guess even them have an upper bound.
Thanks for reading! Yeah, I'd agree except that I actually really like Hawaiian pizza. I can't help my perversion. It's a disease!
What does this view imply about the grandfather paradox?
My objection to killing Hitler is that the future you create will be one whose past is no longer compatible with you going back in time to kill Hitler; after all, there's no more Hitler to kill!
It's unclear how you resolve this: does the "new" timeline overwrite the "old"? What notion of time do the terms "new" and "old" refer to?
What will your personal memories consist in? Will you remember that you killed Hitler? Will you remember the old world in which WWII happened? Or will your memories change so that you always remember a Hitler-less world? When will that change happen? Will you suddenly come to beside a dead baby in 1889 with no memory of getting into a time machine?
It seems like this point of view leads to too many inconsistencies with no obvious resolution.
Hey! You ask "What does this view imply about the grandfather paradox?" The answer is: Absolutely nothing! I'm not making any claims about whether any kind of time travel is coherent or possible. I want to sidestep all that. I'm just using fictional time travel as a way to illustrate what the thesis of causal determinism says and what it doesn't say!
Fair enough; I just think the time travel framing confuses things too much.
Hm, okay. Well, I've added a couple of short sentences now make clear that I'm not making any claims about the coherence of possibility of time travel. Not sure if that'll make any difference for readers, but hopefully it does!
Oh sorry, I don't think that was necessary! I just found the time travel thing confusing
I totally agree with this view. And unlike personal identity I even agree with you that this one isn’t depressing at all! In fact randomness would be somewhat more depressing because it would mean we aren’t control. Determinism means we always do what we decide to do. It’s just that who we would be and thus what we would do was predetermined. But that’s good for us because it would have been bad if we had not been predetermined to exist.
I do think quantum mechanics might make determinism slightly false. I don’t think it matters much on normal timeframes, but it could butterfly effect over long time frames.
Hey thanks for the comment! And yeah, quantum mechanics might throw a wrench in the theory, but just barely.
You probably know this better than I do, but the jury's out sbout how exactly to interpret quantum mechanics and some interpretations are deterministic. Nonetheless, some aren't, including, from what I understand, the most common interpretation.
For this reason, some philosophers will talk about near-determinism: determinism at the macro-level but not the micro-level. I omitted all this for simplicity's sake, but if you just substitute "near-determinism" for "determinism" the whole way through the piece, everything works out. Maybe I'd better add a footnote.
Update: I added a footnote!
I wouldn’t say I’m an expert. My understanding would be that we know for sure there are things we can’t predict, so it seems like randomness. And then Bell’s Theorem, which is now conclusive basically, rules out local hidden variables that could explain this randomness. There could still be non-local hidden variables. I don’t know much about that but it seems very strange, though perhaps no more strange than quantum mechanics as a whole or the very idea of true randomness.
I’ve had a bit of a discussion about all this with a couple of people who know a lot more about physics than I do (Misha and Glen) on Twitter. From what I gather based on that, what I’ve said about deterministic interpretations of quantum mechanics is correct, but again, definitely out of my depth here. Anyway, if it interests you, the conversation is under my post about this piece!
Yeah I think everything is consistent. I’m 100% sure general relativity does not conflict with determinism at all so Misha is correct and Glen is off base there.
Good to know!
If someone goes back in time to kill him, then someone came from the future to kill him in his own time. Why would he have done that, since it would have been impossible for him to do what he did because he would have been dead?
You cannot change the past.
Again if he was killed at that time, there would not have been the future we know.
5 cases of time travel and the explanation : https://arthur3144.substack.com/p/time-travel
I feel like the article could be easily clarified by making it unambiguous what you mean by “change the future” in the intro, and by dropping the focus on causal determinism as a theory about events being fixed in the present in the second section.
With regards to the former, for example, when I read your assertion that causal determinism (CD) is compatible with the notion that we can change the future, given time travel, I interpreted that to mean that on CD we can change the future given everything that happened in the past (including your time travel shenanigans) and present. That, in essence, if I sort of knew what I did in the past before stepping into the time travel machine, I could go back in time and take a different approach to get a different future outcome. Which of course is both false and not your intended meaning.
After all, when we speak of change, we must mean a modification to some baseline state, but what is that baseline state if not a future predicted/determined by the events I knew of prior to my time travel shenanigans?
And regarding your definition of CD being about the present, first that’s not quite how philosophers and others usually define the term. Typically, causal determinism is interpreted as thesis about every event (past, present, and future) being determined.
Second, it confusingly and unintentionally implies that you are somehow contrasting causal determinism and fatalism by suggesting the former is only focused on the present (and not the past or future) while the latter is expanded to encompass the future (this was doubly puzzling for me, given my previous misunderstanding of your intended meaning of “change the future”). But that is both incorrect, and neither Dennett’s nor your intended meanings. So (to me at least) this created unintentional confusion.
Just some food for thought. Thanks.
Hey, thanks for the comment, but I'm a little confused about a few things you say. Instead of addressing all of them in one shot, I'll just address the main one for now.
You consider the possibility that "if I sort of knew what I did in the past before stepping into the time travel machine, I could go back in time and take a different approach to get a different future outcome."
You then say that that is "both false and not [my] intended meaning." But unless I've misunderstood what you're saying, that is indeed my intended meaning and I think it's true! Could you clarify?
Oops, I think I skimmed the baby Hitler section too fast! In any case, I interpreted the baby Hitler example to be a case of time travel where your previous time travel actions did not occur in the past (so more like alternate reality travel). Let me know if that’s not what you meant, and whether you think that even with every event in the past, including what exactly you did in the past as a time traveler, being fixed- you can still change the future. I indeed assumed (erroneously I take it?) in my first comment that the answer would be no. Feel free to let me know if I made any other misunderstandings! Apologies.
To be clear, my position is that
A) Real causal determinism is the position that all events (past, present and future) are determined, given a set of fixed initial conditions in the distant past and the laws of nature- these events cannot be changed. We are destined to do something, given all the temporal events in the past and present.
B) Nonetheless what we do makes a ‘difference’, in the trivial sense that if our actions had been different, by definition the events in the past, present and future would be different. So the future would indeed be different, but it’s trivial in the sense that a change in the past is required.
C) Fatalism goes above and beyond this because it suggests that our actions are pre-ordained by mystical forces/outside agency (e.g. God), see the SEP intro: https://plato.stanford.edu/entries/determinism-causal/#Int
D) This is also Dennett’s position (see in particular his comments on pg.565 of the article you linked, on us being destined to do something on determinism. He admits this is true, but argues it is irrelevant for free will).
I assumed in my first comment that you agreed to A-D, but I’m not sure if that’s the case anymore. Sorry for the confusion.
Hey, no problem! Yes, I agree to A-D. I’d hesitate to use the word ‘destined’ myself, if only because it sounds so fatalistic: Some force (which??) has decreed that this must happen and will pull some strings behind the scenes to make sure it does. But of course, this is merely a terminological point. I know that’s not what you mean.
Now as for your first question: What I say is that I want to avoid committing myself to any particular model of time travel. Now of course, if what you have is a closed causal loop of the kind depicted in 12 Monkeys or Prisoner of Azkaban, then you can’t change anything, since the timeline included the time travel the whole time.
But that kind of model is incompatible with my claim that, by traveling back to the past, you change something about the past state of things and so should be able also to change the future. Granted, I don’t make this clear enough. Maybe I should rework that bit in the interest of clarity. At the very least, maybe it’d be a good idea to add a footnote.
But there are all kinds of models (none of which make any sense, of course), where there’s a single timeline and yet one travels back to the past, thus introducing oneself into the single timelines. There are also ones that create new timelines. Either of these models work with what I’m claiming (though in the latter case, what you’d be changing is the new timeline, not the original).
Hope this clears things up! The main lesson in all this is: Yeah, time travel’s a huge headache!