The “evidence” is irrelevant if it actually has no effect on the outcome already decided, right? If Ella believed that planting evidence, or standing on one foot, or whistling Dixie, or crossing her eyes, somehow created “evidence” that would convince you that Jacob created the mess in the room, she’s free to do so. Maybe she’ll even convince you. But that will not even slightly change the fact that Ella made the mess (if in fact she did).
If you want to make this more analogous to the paradox, there has to be some mechanism (like opening a box) that will demonstrate to you the true facts. At that point, you’d concede, I’d hope, that planting evidence, standing one one foot, whistling Dixie, and crossing her eyes aside, Ella actually made the friggin’ mess, and all the “evidence” was a red herring that had no effect on who is to blame.
To make it even more analogous, it would be something like this: In most instances where you ultimately discover that Ella made the mess, you have found that you’re wearing a red cardigan. No reason why, but that was the case. In instances when you discovered Jacob was the culprit you happened to be wearing a Motley Crue concert tee-shirt. Another series of strange coincidences (or was it?). Now, if you suddenly ran to your room and put on a red cardigan, that will not somehow provide “evidence” that Ella is the likelier culprit. At that point, she either did it, or she didn’t.
Same with the paradox’s boxes. The money’s in there, or it isn’t. Nothing you do at that point provides “evidence” that makes it likelier that the money will be in there. Anything you do at that point will be…wait for it…completely disconnected to whatever is in the box. On account of it ALREADY BEING IN THERE (or not) and all. Indistinguishable’s reference to “evidence” is, as I pointed out, silly.
Not that it apparently is of great concern to anyone, but as concerns the probabilistic reasoning I am employing, I want to note that a better parenthetical would have been “(both when people take both boxes and when people just take the one [both cases having been observed plentifully])”.
Okay, Indistinguishable, you present the notion that the facts in the past are dependent on “our” knowledge of them. Exhibit A:
Emphasis mine. What do the facts care about us? What we know or don’t know about them is irrelevant to whether they have a truth value.
You, sir, are some variant of a solipsist. As I’m not one (yet), this notion of yours that the actual facts are dependent on your or anyone’s perspective on them is a very strange one to me. I mean, suppose there are three people looking at that bit of yours. It’s locked at, say, a 1, since the people observing know it’s a 1. Then one of those persons looks away or leaves or dies or something. Two people are still aware of the bit’s value, so it has to stay a 1. Then another one of those remaining people goes away too; however there’s still one person left watching, so the bit is still locked at being a 1. Then somebody takes that last person down and passes them around, and there’s no more people observing on the bit. Oh no! Nobody’s looking at the bit! It might suddenly spontaneously turn itself into a 0! Or sneak off and kill some poor cat in a box! Or maybe go take a holiday vacation for two in the Bahamas with your wife!!! You’d better run back and look at it -ah, good, it’s still a 1. Better stay there watching it, because if everybody forgets what its value was, it’s going to start going crazy, turning into a probability and maybe going out and mugging people, or doing god knows what. (Or, wait, if god knows, then the bit’s not forgotten, and so can’t run off and do evil. Phew! At least theists don’t have to worry about this bizarre theory of yours.)
Now, I hear that small children actually do worry about this. If an object disappears from sight, such as being closed in a box, then they worry that it might have vanished, and therefore like to check. A parent leaves the room, and they might start to worry that they’ve stopped existing. I think the concept that things don’t actually disappear when you look away is one known as ‘object persistence’, and most people learn it at some stage of their childhood.
Now, I’ll be the first to admit that i occasionally have a relapse. Sometimes I’ll suddenly get the willies and go check on something to make sure it’s still where I left it. But in general, I believe that unless there’s some way that something could move or change itself (like, being a living child or infant), that I can reasonably expect it not to wink out of existence when I turn my back on it.
Facts in the past are like that as well. This might be something of a blow, but those facts don’t care whether you or anyone else knows them or not. Come on, how could they? They’re just statements with truth values; they’re not even actualized in reality, much less alive or something. And if facts don’t care about whether people know them or not, then they wouldn’t wait for everyone to forget about them to discard their fixed truth values and go spinning wildly off into the unknown. Instead, things would just change at random all the time. Either that, or facts would stay fixed all the time. Which one of those two cases do you think occurs?
All the evidence I have indicates that past facts do not change. Now, this evidence I have certainly does not extend to past facts that nobody has information about, true. However, to claim that weird things only happen in the areas we can’t understand is a god in the gaps argument. Unless you have proof, I’m inclined to disbelieve it. (And, of course, proof for that is impossible.)
(The question of whether future facts are similarly fixed, or as you put it, whether “what will happen, will happen”, is simply another way of stating the question “Is the universe deterministic or not”? We don’t have direct memories of future object or fact persistence in our memories (or at least I don’t), and last I heard, the juries are still out on what the secondary evidence indicates about the predictability of our universe. We may very well be deterministic. If we are, then probabilities are as flawed a predictor for the future as they are for the past. However until I learn otherwise, I figure I might as well use probabilities to describe the future, and facts to describe the past. I believe that’s what the people who specialize in probabilities and truth theory do, anyway.)
This is all beside the point relating to Newcomb’s paradox, though. Not only are the contents of the box explicitly an unchanging fact, somebody knows about that bit so it couldn’t change anyway: the predictor who set it.
[/aside]
Okay, let’s review the position you appear to be holding regarding Newcomb’s paradox. As far as I can tell, it’s that you know that the contents of the box are fixed, but until you know what’s actually in there, it’ll make you feel more confident in your odds if you make the bet that discards the $1000. You base this decision on the notion that the predictor is a good bet, and you’ll feel more confident if you make the choice that ‘affects your evidence for what is in the box’ in a more positive way. So the effects of either choice are:
Choosing one box: Your confidence that you’ll get the million increases dramatically, your actual odds of getting the million are perfectly unaltered, and the actual amount of money you get is less by $1000.
Choosing two boxes: Your confidence that you’ll get the million plummets, your actual odds of getting the million are perfectly unaltered, and the actual amount of money you get is greater by $1000.
I believe that this statement of the options correctly incorporates your perspective, while remaining as true as possible to the actual scenario. So, it looks to me like you’re choosing the short-term pleasure of basking in the belief that you’re a sure winner over the result that actually gets you more money. This sort of thinking is what keeps the lotteries in business. (And the lotteries do stay in business, so it’s no surprise that there isn’t a popular concensus about Newcomb’s Paradox.)
The problem is designed to present just this choice: Good feelings, or actual results. It also presents the risks of thinking probabalistically about past events. Suppose there’s a one in a million billion trillion chance of some specific event or series of events ever happening. Then suppose you come across evidence that that specific improbable series of events actually occurred! If you apply probability to the situation as a determiner of whether to accept the facts, you’ll reject them, and poof: you’re a creationist.
Statistics are an extremely useful prediction tool. However, when applied to past events, they are a little out of their element. True, they still work, after a fashion; they let you make an ‘educated guess’. But the reality of the matter is that what happened, happened, and that’s that. If your evidence helped you guess right about what happened and profit from it, then cool, but you don’t accomplish this by attempting to doctor the evidence to support your desired outcome. Which is what you’re trying to do when you say that choosing one box is good “evidence” that the predictor filled the box. Sorry, no. The box is full or not. Fiddling with the evidence doesn’t do anything but instill in you false confidence, and keep in mind that you’re paying for that confidence. You’re not paying to increase your actual odds of getting the million; those are either 100% or 0% depending on the predictor’s past actions, and they’re not going to change based on your current actions.
We’ll note that the glass box scenario presents a slightly different scenario: there is still no difference in actual odds and you still lose money if you take the one box, but the “Good Feelings” element is removed. Since “Good Feelings” are why you’re throwing away the money in the unaltered scenario, it makes perfect sense that removing that incentive will cause you to switch the the only strategy that has any incentive now: the choice that gets you more actual money.
The accuracy of the predictor is, as Stratocaster pointed out, nothing more than bait on the hook. The higher the accuracy you imagine him to have, the more alluring the bait, but the fixed contents of the boxes are the barb that pierces the worm. There is only one correct choice here, and throwing away a sure thing to try and twiddle the odds on an already determined outcome isn’t it.
Note: the business I said about taking the two boxes being the correct choice in a deterministic world, that was phrased poorly. In that scenario, the correct choice is still to take both boxes, however the choice to take both boxes when the predictor put the million in both boxes is unavailable. It simply does not exist; a person with the million in front of them can’t choose two boxes; regardless of the fact that taking only one is the poorer choice. That is why, as a non-interactive observer watching my automoton body carry out its scripted choices, I would by mildly pleased to see it choose the one box and take the million. Since it was impossible for the body to make the correct choice and take the full $1001000, it might as well get something for the effort of trundling along its predetermined path.
However, if you have any free will at all, you should take both boxes. Anything else is throwing away good money after a ship that’s already sailed.
I don’t see where Indistinguishable said here anything about the facts about the past depending on us. Rather, he described a hypothetical situation in which the facts are independent of us, in the strong sense that our beliefs (or lacks thereof) regarding these facts are irrelevant to the truth of these facts. In other words, in Indistinguishable’s hypothetical, the moon landing occured no matter what we believe or fail to believe about it. This makes the facts independent of us.
Except for the last sentence, I can’t see how the above can be said by someone who is in an substantial disagreement with the position Indistinguishable and I have been arguing for.
If the consistency of facts were independent of us, they’d very likely change while we were looking. They sure don’t seem to. If you assume that the fixed nature of past facts are dependent on our knowledge of them, then this is not a problem. However, if you don’t assume that you are the center of the universe, then there is compelling evidence that past facts are indeed fixed, whether or not we’re looking.
As I pretty much already said in the paragraphs following the one you quoted.
Well, I did phrase it poorly.
Sometimes the only option is only the best option by virtue of the better option not actually being an option.
I think the reasoning behind that is as follows? That if they don’t change when we’re looking, it seems our looking has something to do with their remaining the same? Hence the fact that they don’t change when we’re looking indicates that their consistency does depend on us after all?
That’s what I think you’re saying, though I hesitate to affirm this with any confidence sense if you said what I think you said, you’re a solipsist of some sort, and this is exactly what you said you find strange (and wrong) in Indistinguishable’s post.
I also hesitate to affirm this interpretation of your statement because the inference contained in the statement as interpreted doesn’t follow. If they don’t change when we’re looking, this says nothing about whether our looking has something to do with their not changing–because for all we know, they don’t change when we’re not looking, either. (Indeed, I don’t think any participant in this discussion thingks they do!)
But if my interpretation of the reasoning behind your statement is incorrect, then I am at a loss to say what else you might have meant.
Where “this,” I think, refers to the fact that the truth of facts doesn’t seem to change when we’re looking. But if that’s what “this” refers to, then this statement confuses me as well. If I assume that the fixed truth value of past facts depends on my knowledge of these facts, then things could change while I am looking at them, i.e., opportunities for such a change would arise whenever my state of knowledge changes.
True enough, but I do not see how this has to do with what Indistinguishable said, nor do I see exactly how it is relevant to your previous comments in this post, as I’ve explained above. Indistinguishable set up a hypothetical situation in which something happened in the past, but the evidence for it now is not available to us. Indistinguishable’s hypothetical presume that the past fact is fixed, even though not only are we not looking, we can’t look because the evidence has been made inaccessible to us. As far as I can tell, there is nothing in those presumptions with which you ought to disagree by your own account.
Ah, I’ve confused you, probably becuase I was assuming that you shared the odd perspective that Indistinguishable seems to be presenting, and was responding to that position more than your actual words. My bad; I apologise.
The sentence should have been more properly typed "If the consistency of facts were independent of us, and past facts were in fact not fixed with perfect consistency, they’d very likely change while we were looking. They sure don’t seem to. " This is a brief proof by contradiction, intended to disprove the middle clause there, which I mistakenly assumed was obviously the subject under discussion, and so for brevity omitted entirely. :smack:
I’m thinking that on this particular subject, we’re in perfect agreement (even though I managed to hide that fact for a while). Past facts do not change when we’re not looking.
The thing is, I don’t think that’s the hypothetical that Indistinguishable is presenting. I think he’s saying “If you don’t have any information about a particular past fact, how do you know that it has a fixed truth value? Perhaps it doesn’t.” If that’s not what he’s saying, then I don’t know what this side discussion is even about, because he’d have confused me as efficiently as I confused you.
(And yeah; absent considering the contents of the box “in flux” due to being “unknown” (give or take the knowlede of the predictor), I don’t see why a proof by cases couldn’t be applied here. Unless, of course, you’re in it for the fleeting good feelings, rather than the actual money.)
It seems I have a fair amount of reading to do to catch up before I can give a proper response to everything; however, begbert2, I think you misread “Suppose there was absolutely nothing about the current state of the universe which could tell us if there was a moon landing in the past or not.”. The “us” in that was just conventional linguistic filler; that particular sentence wasn’t supposed to be making any real reference to human beings and their mental states, or any such thing. It was just meant to say “Suppose there was absolutely nothing in the current state of the universe which gave any indication as to whether there was a moon landing in the past or not.”
The idea was that I wanted to see why you hold that propositions like “There was a moon landing in the past”, even in such a hypothetical situation as that there would remain no telling trace of such a thing, are somehow more concrete or fixed or determined or something than propositions like “There will be a Mars landing in the future”. I want to understand the grounds for this fundamental temporal asymmetry in your perspective. I could understand taking both propositions to be “fixed”, I could understand taking both to be similarly not “fixed”, I could even understand claiming that “fixed” is not a terribly useful or meaningful concept. But what makes me curious is why you would take the one to be “fixed” and the other not, even when the only difference between the two is basically that one makes reference to lower-numbered years than the present and the other to higher-numbered years.
Or, to get at part of it another way: you keep saying that if one considers the truth value of (any? all?) statements about the future to be “fixed”, then one must be accepting a sort of determinism (presumably, that the state of the universe at all future times is entirely calculable from the state of the universe at the present time). Fine. But do you also maintain the reverse, that since you do believe that statements about the past are “fixed” in just this manner, you must be accepting a sort of reverse determinism (that the state of the universe at all past times is entirely calculable from the state of the universe at the present time)?
If yes, then… really? You take it as a necessary principle that no two different past states could both be capable of leading to the same current state? That everything about the past is calculable from facts about the present, that no information can be truly deleted or erased?
And, if no, then: why do you think that a particular perspective on propositions about the future compels determinism, while not thinking that the same perspective on propositions about the past compels reverse determinism?
Why should past facts care whether the universe remembers then?
Whether you pin the fixedness of the universe on your own knowledge or “our” knowledge or all the knowledge embodied in the current state of the universe, the situation is the same: there’s a range of past times, place, facts, and events that are “remembered”, and outside of that range are things that have been “forgotten”. You are drawing a magical distinction between remembered and forgotten facts that there is no practical basis for believing in.
And no, I don’t believe that no two past states could result in the same present state. What I’m saying is, whether a fact is “forgetten”, by you or I or the state of the universe itself, is irrelevent. It has no bearing on the fact’s actual truth value. Sure, it has a bearing on whether we can determine that truth variable, but the truth doesn’t depend on what we know anyway, so that doesn’t matter.
Take, for example, a chessboard. This will represent the state of the universe. Every move made on the chessboard is an advancing state of the universe. Now, suppose that the players start out by each moving their knights out, first king’s side, then queens side. Now, there’s nothing in the state of the board that indicates that the king-side kights were moved first. The same board-state could have also been acheved by moving the queen-side knights out first, or some mixture of the two movement orders between the two players.
So, there is no way of telling from the state of the board that the king-side knights were moved first. Does that mean that it is no longer a fact that the king-side knights were moved first? Of course not. Forgotten events, even ones that the entire universe has ‘forgotten’, still happened. The truth of the past doesn’t care if it can be determined or not. How could it care?
My argument for the perseistence of the past has nothing to do with requiring all facts to be knowable; the argument is actually the same as the purely solipsistic argument that I presented before: There is no evidence that the past fluctuates and objects are impermanent within the range of the past that we know about, and there’s no reason whatsoever to assume that events we don’t know about aren’t similarly consistent. This applies to all events we don’t know about, whether the universe knows them or not.
Now, let’s talk about determinism. Determinism is where the future is utterly inflexible; the events at any given time are fixed. There are two ways that I can think of that this could occur: the entire timeline could be fixed ‘in advance’, such that our notion that there is a “present” time is merely an illusion; or the mechanics of the universe could be such that the course of future events are caused by past and present events, such that no portion of the universe is not left to chance. (Typically the term ‘determinism’ refers exclusively to the latter scenario, but I will discuss both for completeness.)
If the universe is literally prescripted, it may be analogized to a story in a book. In this case, like in various books, predictions could be made that “just happen” to come true, without any in-story reason for them or for the foreknowledge at all. This is also the secnario that would occur if the future were fixed for the same reason and in the same manner the past is; “what happened, happened”, and “what will be, will be” would both simply mean “what is, is”.
The problem with this theory is that the evidence doesn’t support it. In a static timeline, there is no “present”; why do we seem to percieve one? Our belief (well, my belief) in the fixedness in the past is based purely on memories of mine and others that I know about; there are no such memories of the future, so there is therefore no reason to conclude that the future is similarly fixed. And in general, there are not accurate predictors all over the place, taking advantage of the fixed timeline to make accurate predictions.
In short, this scenario is little more than a thought exersize, like the one where you imagine all of time, memory, past, and self to be an illusion, and imagine yourself a static timeless entity, frozen in the act of navel-gazing. There’s no reason to give credence to it, and a couple reasons not to.
The second and most common use of the term determinism refers to the case where the current state of the universe (the positions and velocities and so forth of everything in it) unvaryingly causes a single, specific ‘next state’. The notion being that, if no uncertainty is introduced into the system, successive state will follow successive state, with the next state always being completely ‘determined’ by the prior state, for all successive states forever.
The opposite of determinism is where to differing ‘next’ states could occur based off of the same prior state. (This happens in chess games, for example; in most cases there are multiple different possible states the board can end up in the next turn from of a given board state.)
Now, there is some evidence that something like determinism occurs at the macro level; when you toss a ball in the air, it generally follows a predictable path, especially in “optimum conditions”, and this path seems to be determined by the prior velocities and postitions causing the successive velocities and positions, for the full course of the flight. However, I believe that there is still uncertainty about precisely how determined some things actually are, and whether some things might introduce random, or at least not perfectly determined elements, thus causing the future to have multiple possible paths.
If determinism does not hold, then events in the future may vary in an unpredictable and unconstrained manner, from the perspective of the current state. This makes prediction increasingly inaccurate the further out the prediction is, and as more non-determined elements come in alter the course of events. People like to think that their minds, though not entirely random, incorporate a sufficent number of nonpredictable events that they cannot be predicted with certainty at all. I believe we call that notion “Free will”.
Now, note that determinism is based on causation. As we have rehashed repeatedly, causation is not reversible, so under this the more common model of determinism, “reverse determinism” is an incoherent concept. I can’t even think what you’d mean by it.
I wouldn’t necessarily say they should. I was trying to see what your view was.
But why should future facts “care” whether the universe predicts them either, then?
Ok. Now why doesn’t this apply to facts about the future as well? Sure, I don’t know, and nothing in the universe can tell, what the outcome of tomorrow’s coin flip will be. Same as I don’t know, and nothing in the universe can tell, what the value of yesterday’s erased bit was. Why does the statement “Yesterday’s erased bit was a 0” have a truth value today but the statement “Tomorrow’s coin flip will be tails” does not have a truth value today?
Fine by me. It is a fact that the king-side knights were moved first. And at the beginning of the game, it was also a fact that the king-side knights would be moved first. If the next move is of a pawn, then it is a fact right now that “The next move will be of a pawn.” Right after the king-side knights were moved, it was a fact that “The queen-side knights will move next”. Sure, nobody knew/knows these facts, these facts couldn’t in any way be determined from the current state; there was/is no rule demanding that these facts be true; things could have been different. All the same, those were the facts. At the beginning of the game, the statement “Many pieces are capable of moving next, but as it happens, the king-side knights will move next” was a fact. No?
I don’t believe I’m saying the past “fluctuates”, whatever that means. I’m just saying the past is like the future; if you maintain that statements about the past have definite truth values, despite our inability to know them (and this is perfectly fine by me), then you should grant this to statements about the future as well.
Ok then. I wouldn’t call the latter a situation where the future is “utterly inflexible, period”; the latter is a situation where the future is entirely constrained by the past. There’s more than one possibility for the future; it’s just that different futures require different pasts. Given any one past, there’s only one matching future, but this doesn’t mean the future is utterly inflexible, any more than reverse determinism (“no two different pasts can match the same future”) means the past is utterly inflexible.
Except, of course, for referring to the past and to the future. But sure.
The same reason we perceive a “here”. There’s no global, fixed “here” or “now”; these are terms that drawn on deixis. There’s “now” at this point in the story, “now” at that point in the story. That’s not so odd.
If “fixed” means “The state of the universe at time t is the same as the state of the universe at time t”, then this is tautologous; you don’t need any evidence for it, it’s necessarily true.
If “fixed” means “The state of the universe at time t is pretty much the same as the state of the universe at time t + epsilon” (e.g., what’s in my closet today will be pretty much the same as what was in it yesterday), then I think we have good reason to conclude that this holds for the future just as much as it holds for the past.
If “fixed” means something else, please elucidate.
There are tons of accurate predictors all over the place. The only reason I’m able to have this conversation with you is because I make accurate predictions: that you will reply in English rather than French or gibberish, etc.
Sure, these aren’t infallible predictions, but they’re damn reliable. (If you think their fallibility is somehow important, why? Memories aren’t infallible either.).
I think it’s a mistake to conflate “free will” with unpredictability. I mean, you can use the term however you want; however, for the purposes to which it is usually put to work, I think there is a much more useful sense it can be given, which correlates well enough with the ordinary use of the word that it has legitimate claim to be associated with it, which has nothing to do with unpredictability.
To use the same old example, I predict that the majority of replies to this thread will be in English. It’s not infallible, but I think we can agree that it’s pretty damn reliable a prediction. Does this ability to make accurate predictions mean that you aren’t exercising free will when you choose to reply to this thread in English? It’s a pretty safe bet, given how much I enjoy chips and salsa, that I’ll buy some next time I’m at the grocery store. Will I not be exercising my free will in doing so, just because it’s predictable?
To me, saying that an action was taken of one’s free will is something like saying that it was taken voluntarily; but there’s nothing demanding that your volition be unpredictable, or free of all influence from the past. To think otherwise seems to me an all too common bit of silliness.
Well, I told you what I’d mean by it; in the relevant sense, determinism means that given all the information about the state of the universe at some particular time, it is possible to calculate all the information about the state of the universe at future times. Reverse determinism, correspondingly, means that given all the information about the state of the universe at some particular time, it is possible to calculate all the information about the state of the universe at previous times. What’s incoherent about that concept?
I don’t think they do. Remember? I’m the guy who’s into “facts=past, future=probabilities”. “Future facts” reads as near-gibberish to me. (Just like “reverse determination”.)
Because I have no reason to believe, and don’t believe, that the the future is currently fixed.
In an nonpredetermined world, statements about the future don’t technically have a truth value yet; they have a probability. For certain facts that probability may approach or equal 100%, but there’s no reason to think that it has to.
Why on earth should I grant that? There’s no evidence for it. I am not going to come into this assuming timeless (book-analogy) predeterminism.
We only have one present; we only have one past. The universe might have lost or overwritten enough information in itself that an infinite variety of theoretical past-states could be theoretically matched up with the current present without contradiction, but that doesn’t change the fact that only one of these pasts happened. The only question in my mind is whether the future is fixed or not, either through timeless booklike determinism (which sounds to be what you’re talking about) or determinism through an endless stream of inevitable conclusions (which doesn’t).
But, we don’t seem to experience all the nows at all the points in the story. We seem to be experiencing one single now, that advances through the story at a constant rate. This is not really consistent with a timeline that exists all at once, where “past” and “future” are terms without referent, since there is no particular “now” in the constantly-all-existing-at-once-timeline.
“Fixed” means “The state of the universe at time t is either already defined to a specific, unvarying state, or it’s not defined yet but still can be predicted with 100% probability based on extrapolation from a prior defined state.”
Your first definition only applies in the ‘booklike’ scenario, where all future times have already been defined. Your second definition, with its “pretty much the same”, both does not meet the standard of “fixed”, and does not hold to the future as much as it does for the past except in the deterministic scenarios, where there’s actually no uncertatinty at all either way.
By “accurate” I mean “perfectly accurate” or “infallible”; all of your ‘tons’ of predictors are not accurate. (And as for your prediction: @#$%!!)
“Damn reliable” predictors are possible in a nondeterministic world, especially when you allow predictors that are limited in what they predict. We’re talking about determinism vs. nondeterminism here, right? So only (perfectly) accurate predictors matter at all.
How predictable is it? If you have no choice, then your will is not “free”.
How silly is saying that “an action is not taken of one’s free will if it was entirely involuntarily; where your volition is entirely predictable, and/or entirely dominated by the influence from the past.”?
That’s not “reverse determinism”; that’s detective work. And it does not follow from anything that it is possible to calculate all the information about a past state of the universe from the present state of the universe. I don’t think that’s at all true. (Simply drop the earth into the sun and I think that lots of data will be lost beyond the reach of reverse calculation.) And I think this holds true even in a fully deterministic universe. (This does not mean that the past is not fixed; it just means that information can be destroyed. Which shouldn’t suprise anybody.)
Causation, even fully deterministic causation, is not reversible. Sorry, but not every idea or notion in this universe can be reversed and still be true. It doesn’t work with A ^ B -> A v B, and it doesn’t work with time, causation, or causal determinism. Sorry, that’s just how it is.
(And there are certainly contexts in which we at least speak of past probabilities. When playing poker, despite other people having already been dealt particular hands, I still reason to myself “Well, there’s x% probability that Jim has an ace, y% probability I’ll get one on the river, etc.”. Seems to me, the concept of probability can be applied in lots of situations of information analysis; it has nothing particularly to do with the future.)
Yes, well, my question was, why do you believe that the past is “currently fixed” and not the future?
Ok, why does “The bit that was erased yesterday was a 0” have a truth value at this very moment rather than a probability, while “The coin flip tomorrow will be tails” does not have a truth value at this very moment, merely a probability?
It’s obvious that “The bit that was erased yesterday was a 0” has a truth value yesterday, just as “The coin flip tomorrow will be tails” has a truth value tomorrow. But why does the former get a truth value today rather than a probability, while the latter gets a probability today rather than a truth value?
Ok. Tell me again why you granted it to statements about the past, then, so that I may better grasp the missing aspect with the future.
And, perhaps you can at least accept this: Whatever the evidence that led you to grant it to statements about the past, it should at least be conceptually possible that such evidence could have been granted for statements about the future as well, right?
Fine by me. But, going along with that, I would then also say “Though multiple futures could be theoretically matched up with the current present without contradiction, that doesn’t change the fact that only one of those futures will happen.”
What do you mean we don’t experience all the nows at all the points in the story? We do; we just don’t experience them all at one time, that being what it means for them to be different "now"s. Consider a movie reel. This is your timeless book determinism, but it’s no contradiction that the characters in the movie are “experiencing one single now that advances through the story at a constant rate.” It’s just that the “one single now” which is experienced on cel #21341234 is different from the one which is experienced on cel #21341235.
(I don’t know why I bothered bringing up movie reels; this point about “timeless booklike determinism” could be made just as well with a, uh, book. But movie reels seem, in my mind, to bring the proper intuitions a little more strongly).
What a long definition. What is the major use of this definition? It does have a lot of stuff about the past built in. When you say “already defined”, do you mean “is defined currently” or do you mean “is defined currently or was defined at a previous instance in time”?
What about the corresponding notion of, let’s called it blixed, meaning “The state of the universe at time t is either currently defined [or will be defined at a later instance in time] to a specific, unvarying state, or it’s not but still can be calculated with 100% probability based on extrapolation from a later defined state”? Can my “blixed” be put to as many uses as your “fixed”? Why is your “fixed” a useful notion for analyzing Newcomb’s paradox/the smoking gene paradox/any of that? Would my “blixed” be as useful, too?
I’m not sure what you mean by “Your first definition only applies in…”. My first definition is a… definition. I guess you are saying that (currently) the statement “The state of the universe in the year 3000 is the same as the state of the universe in the year 3000” fails to be a true statement. I doubt you are holding that it’s false; is your position perhaps that it is meaningless?
Again, what? “does not hold to the future”? I don’t know what this means. It’s a prospective definition; what does it mean for a definition to hold to the future?
Perhaps you are saying that my claim that “the state at time T and the state at time T+epsilon are pretty similar” holds true for past T but not future T. But… really? You don’t think that what’s in my closet today is pretty much the same as what will be in my closet tomorrow?
Well, ok… why would the absence of infallible predictors be evidence against a static timeline?
I agree that damn reliable predictors are possible in a nondeterministic world, absolutely. We’re not talking about determinism vs. nondeterminism here except insofar as you think it’s necessary for us to be talking about them. Primarily, we’re (or at least I am) talking about “Do statements about the past/about the future currently have truth values or probabilities or what?” and “Can one have free will, even if one’s actions are reliably predictable?” and several other things, subsidiary to a discussion of Newcomb’s paradox(/the smoking gene paradox/etc).
It’s very predictable, but not infallibly so, let’s say. I mean, you know as well as I do just how predictable it is that the replies to this thread will be in English. It’s at exactly that level of predictability. Is that level of predictability compatible with free will/choice/whatever?
I don’t demand that the predictor in Newcomb’s paradox be infallible; I’ve been perfectly happy, as I’ve said, to imagine that he has made incorrect predictions on occasion, but has displayed an overall track record of dominating success. My reasoning has never depended on the correlation between his predictions and my actions being “infallible”, just on it being very good.
You’ve put two notions in there (“involuntary” and “predictable/influenced by the past”), and I’m not sure how you mean to combine them, by a disjunction or what? But, when it comes to “entirely predictable/entirely influenced by the past”, if by “entirely” you mean “infallibly”, then your statement is not entirely silly, though I would still happen to disagree, but that doesn’t matter. What matters is that, if by “entirely predictable”, you mean to encompass also “not infallibly, but to a very large degree of accuracy”, then, yes, this is a silly thing to say. I can predict lots of things to a very large degree of accuracy which I think it is silly to deny are instances of free choice.
I didn’t say it followed from everything, except if you consistently applied those principles you demand of the future to the past as well. It is your claim that “if one believes that statements about the future currently have truth values, then one must be a determinist”. I don’t believe that claim of yours, but at any rate, what I’m saying is, if you were consistent about it, it seems to me you should also accept “if one believes that statements about the past currently have truth values, then one must be a reverse determinist”.
If you don’t think consistency demands that, then tell me what the argument is that concludes determinism from statements about the future having truth values. We’ll see what goes wrong when one tries to turn it into an argument that concludes reverse determinism from statements about the past having truth values.
Sure, but plain-vanilla logic is symmetric with respect to time. If an argument form was logically valid, it would remain logically valid with all the temporal references in it flipped. Whatsoever is logically demanded of the past is logically demanded of the future and vice versa, just the same as one couldn’t hope to construct a logical argument which established something about movement to the east without there also being a corresponding argument establishing something about movement to the west.
Now, this doesn’t mean there’s no way out. Just as one could hope to prove something about the east without having to accept the same of the west by taking an asymmetric premise (for example, by taking the premise “The sun rises in the east”) [or definition (for example, by taking the definition “The east is the direction in which the sun rises”)], so one could do with time. But such premises would be only contingent facts, not logically necessary ones, and also such premises and/or definitions need to be explicitly brought out and justified. (And, of course, correspondingly flipped arguments would still go through with the correspondingly flipped premises; it’s just that we might have reason to reject the flipped premises).
But you seem unwilling to engage in this identification and justification, or accept that there is even any need or use for it; rather than say “Hey, it’s interesting that there’s this temporal asymmetry in my views. I wonder what the premise is from which this springs, and why it should only be accepted in one direction”, you seem to be thoroughly resigned to “That’s the way it is, I’ve already told you, and there’s no point investigating any further. Why are you wasting your time?”.
I’m not scrolling through all of those at 03:20 am.
The brain needs to be a determined “program” for the paradox to be a paradox at all, and the god needs to have access to both the code for the brain and all the variables it might come across.
If the god, given the code and variables, can see that this is a brain that would find the logic of taking two boxes (choice two) appealing, he won’t put the million anywhere.
If he sees that you believe in choice one, he will put the million there.
This follows from the proposition of the paradox that the god is omniscient and can predict the future perfectly, and choice one is the only valid choice.
And I forgot to add… since most people will understand (either before or after choosing) that after taking b2 and finding the million there, they could always open b1 and take the $1000, most people who decided to only take b2, would find it empty, except those who for some reason would be naturally inclined to leave b1 even if they found the million in b2 and knew that there still had to be $1000 in b1.
So! The righteous would get the million, the rest would get $1000. This isn’t a paradox, it’s a fable.
Again, the paradox doesn’t require that the predictor be infallible, that brains be completely deterministic, or any of that. It’s good enough that, say, over the hubazillions of trials in every which configuration, 90% of the time that you took both boxes, there was no million bucks, while 90% of the time that you took just the one box, it had the million. For large enough hubazillion, it seems you should be led to conclude “Hm, maybe this guy is so good at predicting me when I take both boxes, the chances are rather high that I won’t get the million, while when I just take the one, the chances are rather high that I will get the million. (rather high = something around 90% probability). Therefore, I say it’s best to take just the one box”.
Note that nothing in there depends on infallibility, on the brain being deterministic, on anything so strong. All it depends on is that this predictor can predict you rather reliably, a much weaker requirement.
Indistinguishable(I’ll asume you replied to me, as you had the last post before me): In which case, 90% of the Righteous would get the million, 90% of the rest would get the $1000.
What if you removed the proposition that god would remove the money if you tried to cheat or randomize your choice? Then, if you opened b2, and found nothing, you knew you were destined to open b1 at some point. If you then sealed it so you couldn’t open it, would you become immortal? the first thought that comes to mind is that the god would put the million there, so you could not seal b1 and become immortal, but it is highly unlikely that none of the brains capable of devising such an immortal-scheme would be capable of opening b1 after opening b2 and finding the million there, so what’s the solution here? Either, you open b2, find nothing and know that sealing b1 will make you immortal (as you can’t die before you open it, and you can never open it), og you open b2, find a million, and know that you can open b1, since not only righteous people would think of sealing b1.
ie, either you’re immortal, or you find $10001000
I believe the solution to my own paradox, and an alternative solution to the first, is quite simply that an omniscient god wouldn’t be able to create a situation creating an unsolvable paradox that required him solving it, and thus it’s just absurd, and no more of a paradox than what would happen if you went back in time (provided, of course, that you really can’t go back in time).
I think I’ll just wait and let Indistinguishable tell us which one of us has got him right.