Why is the A-Theory of Time Attractive?

post by Tyrrell_McAllister · 2014-10-31T23:11:24.608Z · LW · GW · Legacy · 89 comments

I've always been puzzled by why so many people have such strong intuitions about whether the A-theory or the B-theory1 of time is true.  [ETA: I've written "A-theory" and "B-theory" as code for "presentism" and "eternalism", but see the first footnote.]  It seems like nothing psychologically important turns on this question.  And yet, people often have a very strong intuition supporting one theory over the other.  Moreover, this intuition seems to be remarkably primitive.  That is, whichever theory you prefer, you probably felt an immediate affinity for that conception of time as soon as you started thinking about time at all.  The intuition that time is A-theoretic or B-theoretic seems pre-philosophical, whichever intuition you have.  This intuition will then shape your subsequent theoretical speculations about time, rather than vice-verse.

Consider, by way of contrast, intuitions about God.  People often have a strong pre-theoretical intuition about whether God exists.  But it is easy to imagine how someone could form a strong emotional attachment to the existence of God early in life.  Can emotional significance explain why people have deeply felt intuitions about time?  It seems like the nature of time should be emotionally neutral2.

Now, strong intuitions about emotionally neutral topics aren't so uncommon.  For example, we have strong intuitions about how addition behaves for large integers.  But usually, it seems, such intuitions are nearly unanimous and can be attributed to our common biological or cultural heritage.  Strong disagreeing intuitions about neutral topics seem rarer.

Speaking for myself, the B-theory has always seemed just obviously true.  I can't really make coherent sense out of the A-theory.  If I had never encountered the A-theory, the idea that time might work like that would not have occurred to me.  Nonetheless, at the risk of being rude, I am going to speculate about how A-theorists got that way.  (B-theorists, of course, just follow the evidence ;).)

I wonder if the real psycho-philosophical root of the A-theory is the following. If you feel strongly committed to the A-theory, maybe you are being pushed into that position by two conflicting intuitions about your own personal identity.

Intuition 1: On the one hand, you have a notion of personal identity according to which you are just whatever is accessible to your self-awareness right now, plus maybe whatever metaphysical "supporting machinery" allows you to have this kind of self-awareness.

Intuition 2: On the other hand, you feel that you must identify yourself, in some sense, with you-tomorrow.  Otherwise, you can give no "rational" account of the particular way in which you care about and feel responsible for this particular tomorrow-person, as opposed to Brittany-Spears-tomorrow, say.

But now you have a problem.  It seems that if you take this second intuition seriously, then the first intuition implies that the experiences of you-tomorrow should be accessible to you-now.  Obviously, this is not the case.  You-tomorrow will have some particular contents of self-awareness, but those contents aren't accessible to you-now.  Indeed, entirely different contents completely fill your awareness now — contents which will not be accessible in this direct and immediate way to you-tomorrow.

So, to hold onto both intuitions, you must somehow block the inference made in the previous paragraph.  One way to do this is to go through the following sequence:

  1. Take the first intuition on board without reservation.
  2. Take the second intuition on board in a modified way: "identify" you-now with you-tomorrow, but don't stop there.  If you left things at this point, the relationship of "identity" would entail a conduit through which all of your tomorrow-awareness should explode into your now, overlaying or crowding out your now-awareness.  You must somehow forestall this inference, so...
  3. Deny that you-tomorrow exists!  At least, deny that it exists in the full sense of the word.  Thus, metaphorically, you put up a "veil of nonexistence" between you-tomorrow and you-now.  This veil of nonexistence explains the absence of the tomorrow-awareness from your present awareness. The tomorrow-awareness is absent because it simply doesn't exist!  (—yet!)  Thus, in step (2), you may safely identify you-now with you-tomorrow.  You can go ahead and open that conduit to the future, without any fear of what would pour through into the now, because there simply is nothing on the other side.

One potential problem with this psychological explanation is that it doesn't explain the significance of "becoming".  Some A-theorists report that a particular basic experience of "becoming" is the immediate reason for their attachment to the A-theory.  But the story above doesn't really have anything to do with "becoming", at least not obviously.  (This is because I can't make heads or tails of "becoming".)

Second, intuitions about time, even in their primitive pre-reflective state, are intuitions about everything in time.  Yet the story above is exclusively about oneself in time.  It seems that it would require something more to pass from intuitions about oneself in time to intuitions about how the entire universe is in time.


 

1 [ETA: In this post, I use the words "A-theory" and "B-theory" as a sloppy shorthand for "presentism" and "eternalism", respectively.  The point is that these are theories of ontology ("Does the future exist?"), and not just theories about how we should talk about time.  This shouldn't seem like merely a semantic or vacuous dispute unless, as in certain caricatures of logical positivism, you think that the question of whether X exists is always just the question of whether X can be directly experienced.]

2 Some people do seem to be attached to the A-theory because they think that the B-theory takes away their free will by implying that what they will choose is already the case right now.  This might explain the emotional significance of the A-theory of time for some people.  But many A-theorists are happy to grant, say, that God already knows what they will do.  I'm trying to understand those A-theorists who aren't bothered by the implications of the B-theory for free will.

89 comments

Comments sorted by top scores.

comment by calef · 2014-11-01T00:15:13.957Z · LW(p) · GW(p)

I feel like this is just a really obnoxious argument about definitions.

I especially feel like this is a really obnoxious argument about definitions when the wiki article quotes things like:

"Take the supposed illusion of change. This must mean that something, X, appears to change when in fact it does not change at all. That may be true about X; but how could the illusion occur unless there were change somewhere? If there is no change in X, there must be a change in the deluded mind that contemplates X. The illusion of change is actually a changing illusion. Thus the illusion of change implies the reality of some change. Change, therefore, is invincible in its stubbornness; for no one can deny the appearance of change."

So, to taboo a bunch of words, and to try and state my take on the actual issue as I understand it (including some snark):

B theory: Let there be this thing called spacetime which encodes all moments of time (past,present, future) and space (i.e., the universe). The phenomenal experience of existence is akin to tracking a very particular slice of spacetime move along at the speed that time inches forward, as observed by me.

A theory: My mind is the fundamental metaphysical object, and moments of "time" can only be oriented with respect to my immediate phenomenal experience of reality. Trying to say something about a grand catalog of time (including the future) robs me of this phenomenal experience because I know what I'm feeling, and I'm feeling the phenomenal experience of existing right now, dammit! Point to that on your fancy spacetime chart!

Read this way, I suppose the most succinct objection of the A-theorist is: "If all of spacetime exists, all reference frames are equivalent, etc. etc., why am I, in this moment, existing right now?" To which, I imagine, a B-theorist would respond by saying, "Because you're right here," and would then point to their location on the spacetime chart.

But this isn't actually an argument about what time is like. It's an argument about how whether or not we should privilege the phenomenal experience of existing--of experiencing the now. That is, does me experiencing life right at this moment mean that this moment is special?

I suppose I can see why people that aren't computationalists would be bothered by the B theory, because it does rob you of that special-ness.

Replies from: Florian_Dietz, TheAncientGeek, Tyrrell_McAllister
comment by Florian_Dietz · 2014-11-01T18:27:42.435Z · LW(p) · GW(p)

After reading your comment, I agree that this is probably just a semantic question with no real meaning. This is interesting, because I completely failed to realize this myself and instead constructed an elaborate rationalization for why the distinction exists.

While reading the wikipedia page, I found myself interpreting meaning into these two viewpoints that were probably never intended to be there. I am mentioning this both because I find it interesting that I reinterpreted both theories to be consistent with my own believes without realizing it, and because I would like to see what others have to say about those reinterpretations. I should point out that I am currently really tired and only skimmed the article, so that probably wouldn't have happened under ordinary circumstances, but I still think that this is interesting because it shows the inferential gap at work:

I am a computationalist, and as such the distinction between the two theories was pretty meaningless to me at first. However, I reinterpreted the two theories in ways that were almost certainly never intended, so that they did make sense to me as a reasonable distinction:

  • the A theory corresponds to living in a universe where the laws of physics progress like in a simple physical simulation, with a global variable to measure time and rules for how to incrementally get from one state to the next. I assume for the purpose of this theory that quantum-mechanical and relativistic effects that view time non-linearly can be abstracted in some way so that a single, universal time value suffices regardless. I interpreted it like this because I thought the crux of the theory was having a central anchor point for past and future.

  • the B theory corresponds to living in a highly abstracted simulation where many things are only computed when they become relevant for whatever the focus of the simulation is on. For instance, say the focus is on accurately modelling sapient life, then the exact atomic composition of a random rock is largely irrelevant and is not computed at first. However, when the rock is analyzed by a scientist, this information does become relevant. The simulation now checks what level of detail is required (i.e. how precise the measuring is) and backpropagates causal chains on how the rock came to be, in order to update the information about the rock's structure. In this way, unnecessary computations are avoided. I interpreted it like this because I thought the crux of the theory was the causal structure between events.

In essence, the A theory would correspond to a mindless, brute-force computation, while the B theory implies a deliberate, efficient computation that follows some explicit goal. This is nowhere near what the A and B theory actually seem to say now that I have read the article in more detail. In fact, the philosophical/moral implications are almost reversed under some viewpoints. I find it very interesting that this is the first thing that came to mind when I read it.

Replies from: Tyrrell_McAllister
comment by Tyrrell_McAllister · 2014-11-01T21:49:16.880Z · LW(p) · GW(p)

I probably should have written "presentism" and "eternalism" instead of "A-theory" and "B-theory". Does the dispute between presentism and eternalism also seem to you to have no real meaning?

Replies from: torekp, Florian_Dietz
comment by torekp · 2014-11-03T01:31:04.096Z · LW(p) · GW(p)

It's worse than your typical verbal dispute IMO, because in this case the two verbal conventions could live happily side by side, without over-complicating our communications. All we need to do is be careful with tensed verbs. I haven't argued for this. But try it out for yourself, and see if it works. Here is a blog post I read that I think supports my view, even though the author winds up in a different post thinking there is a genuine puzzle.

Edit: Luke_A_Somers seems to have beaten me to it. Further edit: A-theory, on some ways of fleshing it out at least, may be richer than presentism.

Replies from: Tyrrell_McAllister
comment by Tyrrell_McAllister · 2014-11-03T02:14:06.382Z · LW(p) · GW(p)

The dispute between the A-theory and the B-theory is not a dispute about whether, say, "A-series" talk is valid. Everyone agrees that A-series talk (past, present, future) and B-series talk (before, during, after) are both valid.

The dispute is about which kind of talk is more "fundamental". In particular, if A-series talk is fundamental, then, it seems, there must be an objective fact about which time is "present", and this fact is independent of the time at which the question is asked. If the A-theory is true, then asking which time is "present" is like asking "Who is torekp" rather than "Who am I", because the answer to the first question doesn't depend on who is asking.

To the make the analogy tighter, asking which time is "present" is like asking "Who is torekp?" in a world where the name "torekp" rotates through the population in a systematic way. Yes, the answer is different at different times, but the answer changes without regard to who is doing the asking. Similarly (on the A-theory), the answer to which time is present changes (in some elusive sense), but the change happens without regard to when the question is asked.

Replies from: torekp
comment by torekp · 2014-11-04T01:01:32.522Z · LW(p) · GW(p)

That's a great explanation (the non-relative, non-indexical now, and "who is" analogy - not the "fundamental" talk, which just makes me cringe). But that's A-theory, not presentism, which is being explained, right? This paper claims there's a distinction, at least in that one can be a presentist without endorsing A-theory.

Replies from: Tyrrell_McAllister
comment by Tyrrell_McAllister · 2014-11-04T22:12:26.597Z · LW(p) · GW(p)

But that's A-theory, not presentism, which is being explained, right? This paper claims there's a distinction

Yes. One can certainly be an A-theorist without being a presentist. Some people really have subscribed to so-called "moving spotlight" theories. (Hermann Weyl was an example.)

I'm less convinced that anyone was ever a presentist but not an A-theorist. The paper you cite doesn't convince me for at least the following reasons.

First, the paper doesn't even argue that any non-A-theorist presentists have ever actually existed. Rather, the paper attempts to show that such a theory is, as it were, technically possible.

Second, I don't buy that the paper succeeds even at this. The author constructs the theory in Section 4. But the constructions essentially depends on a loophole: A-theories must posit A-properties, he says, but existence is not a property. Then, in Section 5.3, he deals with what seems to me to be the obvious reply. He allows that maybe A-theories only require A-facts, and not necessarily A-properties. If existence is a fact, then his construction fails. His reply is that "it is still possible to be a presentist without being an A-theorist: we need simply deny the existence of facts. ... If there are no facts at all then there are no existence facts. ... This is not an unreasonable view. There are metaphysical systems that do not posit facts—versions of substance theory, bundle theory, and so on."

I find this unconvincing. I don't know enough about these other theories to know how they get by without facts. But I suspect that they introduce some kind of things, call them faks, that do the work of facts. I suspect that the A-theory could just as well be held to require only that there are A-faks.

Replies from: torekp
comment by torekp · 2014-11-05T22:29:41.936Z · LW(p) · GW(p)

OK, thanks

comment by Florian_Dietz · 2014-11-02T08:37:53.539Z · LW(p) · GW(p)

The meanings are much clearer now.

However, I still think that it is an argument about semantics and calef's argument still holds.

Replies from: Tyrrell_McAllister, TheAncientGeek
comment by Tyrrell_McAllister · 2014-11-02T18:21:41.052Z · LW(p) · GW(p)

Does the argument over interpretations of QM also seem like just semantics to you?

For example, when Eliezer advocates for MWI over Copenhagen, is he mistaken in thinking that he is engaged in a substantive argument rather than a merely semantic one?

Replies from: Florian_Dietz
comment by Florian_Dietz · 2014-11-03T08:56:00.406Z · LW(p) · GW(p)

No, the distinction between MWI and Copenhagen would have actual physical consequences. For instance, if you die in the Copenhagen interpretation, you die in real life. If you die in MWI, there is still a copy of you elsewhere that didn't die. MWI allows for quantum immortality.

The distinction between presentism and eternalism, as far as I can tell, does not imply any difference in the way the world works.

Replies from: Tyrrell_McAllister
comment by Tyrrell_McAllister · 2014-11-03T17:57:07.129Z · LW(p) · GW(p)

No, the distinction between MWI and Copenhagen would have actual physical consequences. For instance, if you die in the Copenhagen interpretation, you die in real life. If you die in MWI, there is still a copy of you elsewhere that didn't die. MWI allows for quantum immortality.

Analogously, under the A-theory, dying-you does not exist anywhere in spacetime. The only "you" that exists is the present living you.

Under the B-theory, dying-you does exist right now (assuming that you'll eventually die). It just doesn't exist (I hope) at this point in spacetime, where "this point" is the point at which you are reading this sentence. When you die in the A-theory, there is not a copy of you elsewhen that isn't dying. The B-theory, in contrast, allows for a kind of Spinoza-style timeless immortality. It will always be the case that you are living at this moment.

(As usual in this thread, I'm treating "A-theory" and "presentism" as being broadly synonymous.)

If you think that other points of spacetime exist, then you're essentially a B-theorist. If you want to be an A-theorist nonetheless, you'll have to add some kind of additional structure to your world model, just as single-world QM needs to add a "world eater" to many-worlds QM.

Replies from: Florian_Dietz
comment by Florian_Dietz · 2014-11-03T19:17:03.148Z · LW(p) · GW(p)

I'm not entirely sure what you mean by 'Spinoza-style', but I get the gist of it and find this analogy interesting. Could you explain what you mean by Spinoza-style? My knowledge of ancient philosophers is a little rusty.

Replies from: Tyrrell_McAllister
comment by Tyrrell_McAllister · 2014-11-03T19:53:49.258Z · LW(p) · GW(p)

Sorry just to throw a link at you, but here is a link :)

http://kvond.wordpress.com/2008/07/03/spinoza-on-the-immortality-of-the-soul/

That post discusses one interpretation of Spinoza's notion of immortality. The basic idea is that the entire universe exists in a timeless sense "from the standpoint of eternity", and the entire universe is the way it is necessarily. Hence, every part of the universe, including ourselves, exists eternally in the universe. Because the universe is necessarily the way it is, no part of it can ever not exist.

comment by TheAncientGeek · 2014-11-03T00:00:28.455Z · LW(p) · GW(p)

However, I still think that it is an argument about semantics

You mean the original distinction, or your computationalist reconstruction? (Which is not at all accurate in my view)

Replies from: Florian_Dietz
comment by Florian_Dietz · 2014-11-03T08:49:43.974Z · LW(p) · GW(p)

The original distinction. My reconstruction is what I came up with in an attempt to interpret meaning into it.

I agree that my reconstruction is not at all accurate. It's just something that occurred to me while reading it and I found it fascinating enough to write about it. In fact, I even said that in my original post.

comment by TheAncientGeek · 2014-11-02T23:50:21.127Z · LW(p) · GW(p)

The main objection to B/eternalism is that it doesn't explain change, as either reality .or illusion....as the WP article says.

B theory: Let there be this thing called spacetime which encodes all moments of time (past,present, future) and space (i.e., the universe). The phenomenal experience of existence is akin to tracking a very particular slice of spacetime move along at the speed that time inches forward, as observed by me.A theory:

The problem is how the tracking works. If you were allowed some sort of moving cursor, that would easy. But you're not.

comment by Tyrrell_McAllister · 2014-11-01T18:15:18.477Z · LW(p) · GW(p)

In this post, I use the words "A-theory" and "B-theory" as a sloppy shorthand for "presentism" and "eternalism", respectively. The point is that these are theories of ontology ("Does the future exist?"), and not just theories about how we should talk about time. This shouldn't seem like merely a semantic or vacuous dispute unless, as in certain caricatures of logical positivism, you think that the question of whether X exists is always just the question of whether X can be directly experienced.

(I've added this as a footnote to the post.)

comment by Luke_A_Somers · 2014-11-01T00:48:30.624Z · LW(p) · GW(p)

It seems to me that they're both perfectly valid and indeed equivalent. Like a change of basis, though, you will seem to have rearranged everything.

ETA: Actually, this IS a change of basis.

Replies from: pragmatist, Tyrrell_McAllister
comment by pragmatist · 2014-11-01T12:03:51.125Z · LW(p) · GW(p)

How is this just a change of basis? To me it seems more like B-theorists have an appropriately coordinate-independent conception of the underlying reality whereas A-theorists are privileging one particular coordinate representation.

Replies from: Luke_A_Somers
comment by Luke_A_Somers · 2014-11-01T15:59:10.425Z · LW(p) · GW(p)

In history-of-the-universe space, time-translations are just changes of basis. The difference between A and B is what time you assign to be '0'.

When you're thinking about yourself, it's appropriate to privilege facts pertaining to yourself. Like, if I'm on a roller-coaster, I will do most of my thinking about accelerations in my personal reference frame. This is a stupid reference frame to use for anything else, even for thinking about the person sitting next to me.

I guess the issue is, it's easier to think of a block universe in B-theory-mode?

Replies from: pragmatist, JeremyHahn
comment by pragmatist · 2014-11-01T16:20:48.192Z · LW(p) · GW(p)

The A-theory and B-theory perspectives are not related by a time translation.

B-theory says that the only objective temporal facts about events in space-time are relative facts. Given two points X and Y, you can ask questions like "Is X in the future relative to Y?" and "What is the space-time interval separating X and Y?" But B-theorists say that it makes no sense to ask questions like "Is point X in the future?" without at least implicitly relativizing the question to some other space-time point or region.

The A-theorist says that there is an absolute answer to the latter sort of question, and it does not depend on any implicit relativization. It is an objective fact about point X whether it is in the future or not. Not whether it is in the future of point Y, not whether it is in the future of the space-time region in which the question is being asked, but simply and non-relatively whether it is in the future. B-theory recognizes no such absolute property of future-ness or past-ness.

I don't see how this difference can be interpreted as a mere change of basis. I think you are attributing to A-theorists an implicit relativization to the space-time region in which the discussion is taking place; you're assuming that they have simply chosen a reference frame with a zero in that region, and that all their claims about past and future are actually relative to this frame. But A-theorists are explicit that this is not what they mean. They believe that an event is either in the future or not, and that this is a fact that is independent of reference frame, just like whether X is in the causal past of Y is independent of reference frame.

Like I said in another comment, A-theorists propose radically different ontologies for space and time, and it seems you are under-estimating how radically different these ontologies actually are.

Replies from: TheAncientGeek, William_Quixote
comment by TheAncientGeek · 2014-11-03T00:10:19.386Z · LW(p) · GW(p)

That is trueish, but the point of introducing a distinction between relative facts (before and after) and absolute facts (past and future) is to get a handle in change/becoming,

Note also that presentism is not exactly equivalent to A series.

Replies from: pragmatist
comment by pragmatist · 2014-11-03T12:19:19.244Z · LW(p) · GW(p)

That is trueish, but the point of introducing a distinction between relative facts (before and after) and absolute facts (past and future) is to get a handle in change/becoming,

This is the stated motivation, although I must confess I have no idea how the A-theory is supposed to be even a partial explanation of becoming.

Note also that presentism is not exactly equivalent to A series.

This is true, but I don't think I conflated the two in my post. I didn't say anything about the existence/non-existence of past and future entities or space-time locations. I was talking about the A-theory, not about presentism, although the two are regularly treated as a package deal in contemporary metaphysics.

I actually think the presentism/eternalism distinction is more likely susceptible to shminux's charge of vacuity than the A-theory/B-theory distinction.

Replies from: TheAncientGeek
comment by TheAncientGeek · 2014-11-03T13:07:27.840Z · LW(p) · GW(p)

This is the stated motivation, although I must confess I have no idea how the A-theory is supposed to be even a partial explanation of becoming.

I don't think it's supposed to be an explanation of becoming, I think it's supposed to be a model of time that takes becoming into account. Explanations have to ground out somewhere.

I actually think the presentism/eternalism distinction is more likely susceptible to shminux's charge of vacuity than the A-theory/B-theory distinction.

The opinion has born put forward a number of times, but I am still waiting for someone to substantiate it by putting forward an explanation of how change is equivalent to stasis.

I can see how "time is passing through me" is equivalent to "I am passing through time" ....but those are two dynamic theories.

comment by William_Quixote · 2014-11-06T12:33:43.233Z · LW(p) · GW(p)

Do these differnt ontologys have consequences for how we understand physics? For example, suppose society generated a quantum random bit and if it was 1 we allocated all future social resources to pushing Mercury into the sun. If under B theory the future allready exists, then there is a fact of the matter as to whether Mercury is still there. Which then implies that there is a determic result that will happen we we generate the quantum random bit used to make this decision.

I'm no expert on quantum mechanics, but whether or not the result of quantum randomness is in fact determined based on a variable allready in the universe but hidden from us seems like something physics may say something about

comment by JeremyHahn · 2014-11-02T16:51:48.155Z · LW(p) · GW(p)

I think that the distinction may be clarified by the mathematical notion of an affine line. I sense that you do not know much modern mathematics, but let me try to clarify the difference between affine and linear space.

The A-theorists are thinking in terms of a linear space, that is an oriented vector space. To them time is splayed out on a real number line, which has an origin (the present) and an orientation (a preferred future direction).

The B-theorists are thinking in terms of an affine line. An affine line is somewhat like the A-theoriests real line, but it doesn't have an origin. Instead, given two points a & b on the affine line, one can take their difference a-b and obtain a point on the real line. The only defined operation is the taking of differences, and the notion of affine line relies on a previously defined notion of real line.

Replies from: Tyrrell_McAllister, Luke_A_Somers
comment by Tyrrell_McAllister · 2014-11-03T00:17:42.719Z · LW(p) · GW(p)

The A-theorists are thinking in terms of a linear space, that is an oriented vector space. To them time is splayed out on a real number line, which has an origin (the present) and an orientation (a preferred future direction).

I think that this analogy is accurate and reveals that A-theorists are attributing additional structure to time, and therefore that they take a hit from Occam's razor.

However, to be fair, I think that an A-theorist would dispute your analogy. They would deny that time "is" splayed out on a number line, because there is no standpoint from which all of time is anything. Parts of time were one way, and other parts of time will be other ways, but the only part of time that is anything is the present moment.

(I'm again using A-theorist as code from presentist.)

By the way, off-topic, but:

The only defined operation is the taking of differences, and the notion of affine line relies on a previously defined notion of real line.

This is true if affine space is defined as a torsor for the reals as an additive group, but you can also axiomatize the affine line without reference to the reals. It's not clear to me whether this means that you can construct the affine line in some reasonable sense without reference to the reals. Do you know?

Replies from: JeremyHahn
comment by JeremyHahn · 2014-11-03T04:19:33.457Z · LW(p) · GW(p)

I have always heard the affine line defined as an R-torsor, and never seen an alternative characterization. I don't know the alternative axiomatization you are referring to. I would be interested to hear it and see if it does not secretly rely on a very similar and simpler axiomatization of (R,+) itself.

I do know how to characterize the affine line as a topological space without reference to the real numbers.

Torsors seem interesting from the point of view of Occam's razor because they have less structure but take more words to define.

Replies from: Tyrrell_McAllister
comment by Tyrrell_McAllister · 2014-11-03T13:20:21.734Z · LW(p) · GW(p)

I do know how to characterize the affine line as a topological space without reference to the real numbers.

This is what I was referring to. The axioms of ordered geometry, especially Dedekind's axiom, give you the topology of the affine line without a distinguished 0, without distinguishing a direction as "positive", and without the additive structure.

However, in all the ways I know of to construct a structure satisfying these axioms, you first have to construct the rationals as an ordered field, and the result of course is just the reals, so I don't know of a constructive way to get at the affine line without constructing the reals with all of their additional field structure.

Replies from: JeremyHahn
comment by JeremyHahn · 2014-11-04T00:36:21.740Z · LW(p) · GW(p)

You might be able to do it with some abstract nonsense. I think general machinery will prove that in categories such as that defined in the top answer of

http://mathoverflow.net/questions/92206/what-properties-make-0-1-a-good-candidate-for-defining-fundamental-groups

there are terminal objects. I don't have time to really think it through though.

comment by Luke_A_Somers · 2014-11-03T18:45:02.635Z · LW(p) · GW(p)

I sense that you do not know much modern mathematics

... from what do you get this impression, and in what way is it relevant? Yes, there are many parts of modern mathematics I am not familiar with. However, nothing that had come up up to this point was defined in the last 100 years, let alone the last 50.

I have a PhD in physics. I know what an affine space is. If you were thrown off by my uses of basis changes to effect translations, which would signal ignorance since vector addition is not equivalent to change of basis... I did clarify that I was in a function space defined over time, and in the case of function spaces defined over vector fields, translations of the argument of the function are indeed changes of basis.

In physics, we set the origin to be whatever. All the time. This is because we need to do actual arithmetic with actual numbers, and number systems with no 0 are needlessly cumbersome to use. Moving 0 around all the time in context-dependent and arbitrary ways completely defuses the 'harm' of A-theory, as far as I can tell.

Replies from: JeremyHahn
comment by JeremyHahn · 2014-11-04T00:42:08.563Z · LW(p) · GW(p)

I apologize for the snipy remark, which was a product of my general frustrations with life at the moment.

I was trying to strongly stress the difference between (1) an abstract R-torsor (B-theory), and (2) R viewed as an R-torsor (your patch on A-theory).

Any R-torsor is isomorphic to R viewed as an R-torsor, but that isomorphism is not unique. My understanding is that physicists view such distinctions as useless pendantry, but mathematicians are for better or worse trained to respect them. I do not view an abstract R-torsor as having a basis that can be changed.

Replies from: Luke_A_Somers
comment by Luke_A_Somers · 2014-11-04T14:28:20.195Z · LW(p) · GW(p)

Indeed it wouldn't. A function space defined on an R-torsor would have a basis which you could change.

comment by Tyrrell_McAllister · 2014-11-01T18:18:41.885Z · LW(p) · GW(p)

I probably should have used the words "presentism" and "eternalism" instead of "A-theory" and "B-theory". I've added a footnote which I hope clarifies this.

The point is that these are theories about what exists, and not just theories about how we should talk about time. To say that all of time is even "there" to be coordinatized is, in essence, to sign onto the B-theory as opposed to the A-theory.

comment by TheAncientGeek · 2014-11-03T00:41:23.718Z · LW(p) · GW(p)

I suspect the tendency to favour A over B .or vice versa is related to cognitive styles which favour raw (ish) experience versus cognitive styles which favour theory and detachment.

ETA

Experience favours presentism, because the past and future are not "there" phenomenologically, and changing present moment is.

Theory favours eternalism because it is hard to represent change mathematically ... it gets lost in the translation.

comment by ChristianKl · 2014-11-01T12:06:36.183Z · LW(p) · GW(p)

From a constructivist standpoint, I can observe the present, so that's there. Than I can construct based on that the past and the future.

It seems like the nature of time should be emotionally neutral1.

I personally can say that I had a very strong belief that if I awake and the last day I remember is Tuesday it should be Wednesday and not Monday. It was quite a shock to learn that isn't true.

Fundamental beliefs often do come with strong emotions if they are challenged.

Some A-theorists report that a particular basic experience of "becoming" is the immediate reason for their attachment to the A-theory. But the story above doesn't really have anything to do with "becoming", at least not obviously. (This is because I can't make heads or tails of "becoming".)

Experience is a word that refers to qualia. The blind man can't easily make sense of someone speaking about experiencing seeing red.

Different perception can easily lead to different intuitions.

Replies from: Tyrrell_McAllister
comment by Tyrrell_McAllister · 2014-11-01T17:54:46.649Z · LW(p) · GW(p)

I personally can say that I had a very strong belief that if I awake and the last day I remember is Tuesday it should be Wednesday and not Monday. It was quite a shock to learn that isn't true.

Yes, it can be very distressing to have a strong intuition challenged. It would be very distressing to learn that arithmetic for large integers didn't work the way I assumed it did.

But in math, the intuition precedes the emotional significance. I don't start out being emotionally attached to a certain answer. Rather, I start out emotionally neutral about the question, I form an intuition about the answer, and then I am emotionally attached to being right.

So, the emotional significance can't be used to explain where the intuition came from. In the case of math, the universality of the intuitions can be explained by our common biological and cultural heritage, and by our common experience with how the enumeration of things works.

In the case of time, we have the same biological and cultural heritage, and we have the same experience with time itself, but we arrive at different intuitions about the relationship between time and ontology. This is what I find puzzling.

Different perception can easily lead to different intuitions.

What are the different perceptions that would explain the different intuitions about whether the future exists?

Replies from: ChristianKl
comment by ChristianKl · 2014-11-01T19:05:41.899Z · LW(p) · GW(p)

What are the different perceptions that would explain the different intuitions about whether the future exists?

You don't have a qualia for "becoming" therefore it's no important concept for you. Other people perceive a qualia for that concept.

I'm not 100% sure on that but I know that it's a common stumbling block when talking about the experience of being. Some people do have a qualia for that others don't.

In the case of time, we have the same biological and cultural heritage, and we have the same experience with time itself

I don't think that's the case. Typical mind fallacy makes you believe that other people have the same experience with time itself.

If I would guess than I would think that people who are strongly associated with their bodies are more likely to prefer A-theory while nerds without a relationship to their bodies prefer B-theory.

For that matter I personally don't feel a strong preference for either of the two.

Replies from: buybuydandavis, Tyrrell_McAllister
comment by buybuydandavis · 2014-11-01T21:51:45.419Z · LW(p) · GW(p)

You don't have a qualia for "becoming" therefore it's no important concept for you. Other people perceive a qualia for that concept.

What do you claim to know, and how do you claim to know it?

Replies from: ChristianKl
comment by ChristianKl · 2014-11-02T17:53:43.012Z · LW(p) · GW(p)

What do you claim to know, and how do you claim to know it?

Fair question.

In the framework of Danis Bois perceptive pedagogy (PP) there's a qualia called "feeling of existence/being".

It took me roughly a year to get a handle around that concept and identify it for myself. I went from a state where I didn't know what a person giving a lecture about PP meant with "feeling of existence/being" as opposed to "feeling present" to having an my own phenomenological references sorted (there's also a related qualia of "globality" and one of "depth" in the PP-framework).

I can say whether I feel that qualia of "feeling of existence/being" more or less, in a quite similar way than I can say whether something looks more or less red. In addition to teaching PP Danis Bois also went to a French university and become something like visiting professor of philosophy. In that role he gave lectures about phenomenologically and existentialism and according to him it's next to impossible to get across to his audience what he means when he speaks about "sense of being".

I would guess that as far as "becoming" goes, things are similar. The semantics that the OP quotes that talk about "particular basic experience of "becoming" is the immediate reason for their attachment to the A-theory" suggest this. "Basic experience" is what people say when they mean qualia.

Unfortunately I'm not good at distinguishing change qualia. When reading Feldenkrais and going to his list of perception test I can do the things that most people can't do according to Feldenkrais. Then Feldenkrais lists things that can be perceived and among them is Rhythm. I have never really thought of rhythm as something that can be perceived of a basic level as a qualia. It triggers I'm me my "notice confusion trigger".

Rhythm might be in the same box of strange time qualia as "becoming", but it's something where I have to speculate at the moment. In a year my thoughts on the matter might be more clear.

That's basically my thoughts from a PP perspective.

I also have experience with time perception within NLP's timeline therapy. If you tell people to remember something, you can ask them where those that they remember come from. Some people perceive the past as something behind them. Others as something right to them, others in the front.

You can mess around with that representation and produce psychological effects with it. It can be quite intensive work and I have multiple reference experiences with it. Those experiences and a bit of theory about timeline therapy lead to intuitions about how things are. It's however not easy to summarize which specific experiences lead to which specific beliefs. Unfortunately NLP is also no subject with well structured theory.

There's an unfortunate paradox. Things in which I'm confident are usually hard to argue. I believe them because the are consistent with a wide array of observations I made in different contexts. They are not condensable to a short text.

I know the effort it took me to get a handle around "feeling to be/exist" from my own experience and I know about the research on phenomenological primitives and how hard it is to learn new ones.

Replies from: buybuydandavis
comment by buybuydandavis · 2014-11-08T03:54:26.088Z · LW(p) · GW(p)

I went from a state where I didn't know what [a] a person giving a lecture about PP meant with "feeling of existence/being" as opposed to "feeling present" to [b] having an my own phenomenological references sorted

[c] I can say whether I feel that qualia of "feeling of existence/being" more or less in a quite similar way than I can say whether something looks more or less red.

It's the jump from [b] to [c] that I question. I'll take it as given that you have some consistent phenomenological referent for what you call a "feeling of existence/being".

The difference with Red is that you can point and verify with other people that you both call the same things red. I don't see any similar verification method with "feeling of existence/being".

I would guess that as far as "becoming" goes, things are similar.

So would I, but from my perspective, people would be better off labeling their feelings as "feelings of gloob1", "feelings of gloob2", to avoid smuggling in conceptual connotations to the feelings they're having, and avoid identifying them with the feelings other have, without evidence of any relation between them.

Do you have such evidence?

When reading Feldenkrais and going to his list of perception test

I've read much of Feldenkrais, but that doesn't ring a bell (it's been a decade now). What book is that from? Is that what it was called?

I'm familiar with NLP and it's theories on the manipulation of sensory modalities. Interesting and plausible to me. I've never put them to the test, however. I like Jonathan Haidt's various moral modalities. And I have my own theory on truth modalities.

I'm open to the possibility of such explanations, but also worry that they can too easily answer all questions. "Well, he just uses different sensory modalities." "Well, he just has different qualia referents than you do."

Going back to the original question that you explained away with qualia:

What are the different perceptions that would explain the different intuitions about whether the future exists?

The phrase "the future exists" simply violates what we mean in english by the words. "Exists" is a verb in the present tense, referring to state in the present, and whatever you want to say about the future, it's pretty well agreed that it aint in the present.

I'm fine with treating time as another dimension, and talking about objects 4 dimensional spacetime objects. All sorts of "future" objects would "exist" in that model. But most people don't have that model.

Me, I think the people who identify exists_everydaymode with exists_spacetimemodel are just conceptually confused by their high falutin ideas. Exists_everydaymode didn't cease to exist when we got our fancy new spacetime model to play with, and it's relevance and functionality didn't cease to exist either. "I have cancer" is really distinguishable in important ways to us from "I had cancer."

comment by Tyrrell_McAllister · 2014-11-01T21:41:34.352Z · LW(p) · GW(p)

For that matter I personally don't feel a strong preference for either of the two.

Do you have this experience of "becoming"?

If so, then it doesn't seem to lead you to support one theory strongly over the other. Why then do you think that it explains the split in intuitions between different people?

If you do not experience "becoming", why do you think that it is a distinct quale that only some people have, rather than a (mis)interpretation of qualia that we all share?

comment by Gunnar_Zarncke · 2014-11-01T09:55:43.312Z · LW(p) · GW(p)

When reading the first paragraph I stopped to think what my intuition about 'orderings of events in time' is. Before being primed by actual proposals (I luckily didn't know A vs. B beforehand and the naming doesn't give anything away). I thought about some events today and yesterday, their ordering and was mindful of how I phrased it and the tempos used. I didn't came to any clear theory of time this way.

To be specific I used phrases like "yesterday I did A and then I did B", "today I first did C and now do D", "later I will do E and then I will do F". During this mental speech I observed it and noticed that almost always two events are put in an X-before-Y or X-after-Y relation.

Then I went back to the wiki-article and read is sentence by sentence to avoid revealing more than absolutely needed to arrive at a forking of my mental model in two by the differentiations provided.

The result? I didn't see any contradiction between both theories and actually don't find any more compelling. Rather they provide different views on the phenomenon and perception of time. Both aspects were present in my model and mental process.

Maybe this view results from the process applied. Maybe it is because my mental model was rich enough initially. And maybe it is as in calefs comment that people with a computational background are used to dealing with both views.

Replies from: Tyrrell_McAllister, TheAncientGeek, TheAncientGeek
comment by Tyrrell_McAllister · 2014-11-01T17:59:09.334Z · LW(p) · GW(p)

Thanks for this data point.

In the post, I unthinkingly used "A-theory" and "B-theory" as code for "presentism" and "eternalism". I'd be curious to know how you react to the positions in these articles.

Replies from: Gunnar_Zarncke
comment by Gunnar_Zarncke · 2014-11-01T22:31:17.253Z · LW(p) · GW(p)

In kind of the same way as calef again: It is a question of definitions or what is meant by 'exist'. A psychological artifact or a reality modelling artifact. There is not really a contradiction. I think philosophy sometimes beats these things to death. They should rather right these wrong questions.

comment by TheAncientGeek · 2014-11-03T00:26:38.417Z · LW(p) · GW(p)

Do you think there can be an equivalence between a theory that says time passes,and one that says it doesn't?

Replies from: Gunnar_Zarncke, torekp
comment by Gunnar_Zarncke · 2014-11-03T06:51:20.642Z · LW(p) · GW(p)

There are lots of dual theories of things. Esp. in math. Think about geometry: In triagulation you can describe areas by their borders or by the centers. Voronoi triangulation converts vetween the 'views'. Implementations of both approaches have different performance characteristics.

Replies from: TheAncientGeek
comment by TheAncientGeek · 2014-11-03T10:10:16.803Z · LW(p) · GW(p)

How is that applicable to this particular case?

Replies from: Gunnar_Zarncke
comment by Gunnar_Zarncke · 2014-11-03T12:12:16.583Z · LW(p) · GW(p)

A theory of time which models time as changing will use entities to represent 'now' and 'change'-events, whereas a static-time theory will designate entities to points in time. The former is better suited to answer questions about now (and implementations built upon that will be faster on this kind of query) whereas the latter is better suited to answer questions about fixed points in time or compare these (and implementations based on this will be faster on these operations).

Replies from: TheAncientGeek
comment by TheAncientGeek · 2014-11-03T12:49:29.049Z · LW(p) · GW(p)

But that isn't a duality in the mathematical sense, because there is no translation of change tfrom the dynamic scheme to the static scheme: it's "horses for courses".

comment by torekp · 2014-11-03T01:53:40.486Z · LW(p) · GW(p)

Yes, in a vaguely comparable way that a different "theory" says the tree in the forest makes no sound, and another says that it does. The verbal problem in your example would center on "passes".

Replies from: TheAncientGeek
comment by TheAncientGeek · 2014-11-03T10:15:01.560Z · LW(p) · GW(p)

Do you think there is an equivalence in this case?

Replies from: torekp
comment by torekp · 2014-11-04T01:10:12.142Z · LW(p) · GW(p)

I dunno, are we talking about A-theory, or presentism, and which version? I'm ready to agree that there are at least some ways of formulating A-theory that don't have a B-equivalent.

comment by TheAncientGeek · 2014-11-03T00:26:21.285Z · LW(p) · GW(p)

Do you think there can be an equivalence between a theory that says time passes,and one that says it doesn't?

comment by shminux · 2014-11-01T03:23:37.306Z · LW(p) · GW(p)

I dislike how people call this vague A- (or B-) intuition a theory, given how it is untestable even in principle. It's no more a "theory" than counting the proverbial angels on the head of a pin. The term "true" does not apply in this case.

Replies from: pragmatist, Tyrrell_McAllister, TheAncientGeek
comment by pragmatist · 2014-11-01T06:25:24.588Z · LW(p) · GW(p)

Except it is testable, I think. If the A-theory of time is true, we would expect our best theory of space-time to contain an objectively definable notion of "present". However, our best theory of space-time contains no such notion, and in fact actively militates against it. It is, of course, possible to posit an undetectable preferred foliation of space-time or some such, but this just shows that you have to complicate your physical theory in order to sustain A-theory.* Plain relativity, interpeted naturally, is simply incompatible with it. So the experimental success of the general theory of relativity is strong evidence against the A-theory.

In addition, the fact that our experience of the passage of time can be adequately explained within B-theory in terms of the Second Law of Thermodynamics means that pretty much the only argument for A-theory (the supposed inability of B-theory to account for our experience of time) fails. So A-theory is a theory that requires us to complicate our best physical understanding of the world for no perceivable explanatory benefit. That is a bad theory in scientific, not just philosophical, terms.

* I see this move as a slightly more respectable version of protecting biblical creationism from empirical refutation by saying that God created the universe 6000 years ago but made it look exactly as if it was billions of years old.

Replies from: shminux, TheAncientGeek
comment by shminux · 2014-11-01T19:11:24.024Z · LW(p) · GW(p)

Except it is testable, I think. If the A-theory of time is true, we would expect our best theory of space-time to contain an objectively definable notion of "present". However, our best theory of space-time contains no such notion, and in fact actively militates against it.

It is true that the original formulation of GR is covariant, i.e. has no time evolution built in, only a "block" spacetime manifold whose curvature is precisely its matter content. Similarly, classical EM, though originally formulated as an initial value problem, also looks better in a "timeless" form, where second derivative of the 4-vector potential is charge-current density.

It is, of course, possible to posit an undetectable preferred foliation of space-time or some such, but this just shows that you have to complicate your physical theory in order to sustain A-theory.*

I disagree. You have to recast GR into an initial value problem and then pick a foliation to model interesting physical phenomena, like stellar collapse and black hole collision. Completely independent of any underlying ontology. There is no intent to "sustain A-theory", that's just silly. You want to know how to detect the dying cry of a star torn apart by a supermassive black hole in the center of a galaxy, not whether to pick A or B from some book.

Plain relativity, interpreted naturally, is simply incompatible with it. So the experimental success of the general theory of relativity is strong evidence against the A-theory.

Are you saying that this A-theory predicts that there is a preferred foliation? By that logic, wouldn't B-theory predict that no foliation is possible at all? Or that all foliations are equal, whether they are timelike, null or spacelike? If so, the B-theory has been clearly falsified (if you can ever falsify anything in philosophy of physics).

the fact that our experience of the passage of time can be adequately explained within B-theory in terms of the Second Law of Thermodynamics

This seems like a major category error to me, mixing qualia ("experience of the passage of time") with statistical mechanics. They are about a dozen of abstraction and energy levels removed from each other. I can't take arguments like this seriously.

A-theory is a theory that requires us to complicate our best physical understanding of the world for no perceivable explanatory benefit. That is a bad theory in scientific, not just philosophical, terms.

What requires us to "complicate our best physical understanding of the world", such as recast the beautiful Einstein equation into an ugly ADM form, is the drive to explain and predict what we see or will see. The ontological narrative is a byproduct.

  • I see this move as a slightly more respectable version of protecting biblical creationism from empirical refutation by saying that God created the universe 6000 years ago but made it look exactly as if it was billions of years old.

This was almost verbatim the Hoyle's criticism of the Big Bang model, wasn't it?

Replies from: Tyrrell_McAllister, pragmatist
comment by Tyrrell_McAllister · 2014-11-01T22:00:35.074Z · LW(p) · GW(p)

Are you saying that this A-theory predicts that there is a preferred foliation?

Not to speak for pragmatist, but, yes, that is my understanding. But, importantly, the foliation isn't just preferred by some distinguishing physical characteristic (the way a preferred reference frame would be, for example). Rather, the foliation is preferred in a more ontologically fundamental sense: When one leaf exists, no other leaves of the foliation exist at all, nor do the parts of spacetime that they would "foliate". For the presentist/A-theorist, at this moment, a completely exhaustive ontology of the world contains nothing that is not in the present leaf.

By that logic, wouldn't B-theory predict that no foliation is possible at all? Or that all foliations are equal, whether they are timelike, null or spacelike? If so, the B-theory has been clearly falsified (if you can ever falsify anything in philosophy of physics).

The B-theory allows foliations to be different from one another in physically real ways. The B-theory doesn't allow that leaves of one special foliation "pass into and out of existence", which is what the presentist/A-theoretic approach requires.

(That is my understanding of what a presentist would say, anyway. But, as I said, I can't really make sense of presentism, so I might not be portraying the view accurately.)

comment by pragmatist · 2014-11-02T04:53:06.543Z · LW(p) · GW(p)

I disagree. You have to recast GR into an initial value problem and then pick a foliation to model interesting physical phenomena, like stellar collapse and black hole collision. Completely independent of any underlying ontology. There is no intent to "sustain A-theory", that's just silly. You want to know how to detect the dying cry of a star torn apart by a supermassive black hole in the center of a galaxy, not whether to pick A or B from some book.

None of these amounts to picking a single foliation and stating, "This (and no other) is the correct foliation of space-time." A-theory requires a single privileged foliation. The fact that we often use foliations when modeling physical phenomena has nothing to do with sustaining A-theory, you're right, but I didn't say that any use of a foliation would have that role.

Are you saying that this A-theory predicts that there is a preferred foliation? By that logic, wouldn't B-theory predict that no foliation is possible at all? Or that all foliations are equal, whether they are timelike, null or spacelike?

No. The B-theory predicts that there is no single preferred foliation. That is not equivalent to saying that no foliation is possible. Nor is it equivalent to saying that all foliations are equal.

This seems like a major category error to me, mixing qualia ("experience of the passage of time") with statistical mechanics. They are about a dozen of abstraction and energy levels removed from each other.

There is, of course, a mystery about how (or even if) particular qualia are produced by physical processes. I don't claim that statistical mechanics can answer that mystery, but that is not a mystery that A-theory claims to answer, either. However, if you grant (as I think you should) that our experience of the passage of time is related to the way in which our brain performs various computations, then stat. mech. becomes immediately relevant, and isn't dozens of levels removed. There is a rich literature applying statistical mechanics to understand constraints on computational processes.

The beauty of stat. mech. methods is that they are not constrained to a particular energy level. They can be applied to understand the behavior of molecules in a gas, but also to understand the behavior of galaxies in a supercluster. In any case, my mention of stat. mech. in this context wasn't just a throwaway. Part of my dissertation was about understanding the experience of time direction (in particular, the fact that cognitive systems record memories in one temporal direction and intervene in the opposite temporal direction) in statistical mechanical terms. I'd be happy to summarize the argument if you'd like, when I have the time.

What requires us to "complicate our best physical understanding of the world", such as recast the beautiful Einstein equation into an ugly ADM form, is the drive to explain and predict what we see or will see.

As I said before, the motivations and consequences of the ADM formalism (at least when applied to obtain numerical solutions to initial value problems) are quite distinct from those of the A-theory. Now it may turn out that in our ultimate theory of quantum gravity, we do have to specify a single preferred foliation of space-time, in which case I will readily admit that this particular objection to A-theory no longer holds. But that just highlights how this debate is responsive to empirical confirmation and disconfirmation.

This was almost verbatim the Hoyle's criticism of the Big Bang model, wasn't it?

Perhaps it was, I don't know. But if it was, then he was wrong, because the criticism certainly doesn't seem to apply to the Big Bang model as we know it. Just because the criticism can be misapplied doesn't mean it's never valid.

Replies from: torekp, shminux
comment by torekp · 2014-11-03T01:46:47.267Z · LW(p) · GW(p)

Part of my dissertation was about understanding the experience of time direction (in particular, the fact that cognitive systems record memories in one temporal direction and intervene in the opposite temporal direction) in statistical mechanical terms. I'd be happy to summarize the argument if you'd like, when I have the time.

Drool - I would very much like to get a copy of this part of your dissertation, or the whole thing, or a summary, whatever and whenever is convenient to you. I think I sorta get it on an intuitive level, but to fill in more of the physics would be wonderful.

Replies from: pragmatist
comment by pragmatist · 2014-11-03T12:16:10.441Z · LW(p) · GW(p)

I've been meaning to make a couple of posts summarizing my dissertation, since it is rationality-relevant. This is an added impetus. I'll get on it.

comment by shminux · 2014-11-02T19:15:51.484Z · LW(p) · GW(p)

Well, I guess I sort of agree that "B-theory" is not as constraining, even if both use the term "exist" in a way I disagree with. In any case, they seem to be more of an inspirational value, like the Mach principle (wrong if quantified) was for Einstein.

I am not clear on how one can use stat.mech to explain our time perception, feel free to elaborate some time. Maybe in a separate post, if the argument is long. I agree that humans only experiencing the present is not an argument against "static" block spacetime.

Anyway, I get frustrated by these discussions online, too much is left unsaid, the back-and-forth is slow and selective, only bits and pieces of the argument and motivation get expressed. Could be just my lack of communication skills, of course. If you are ever in Vancouver and feel like meeting up, lunch/dinner/drinks are on me.

Replies from: pragmatist
comment by pragmatist · 2014-11-03T12:25:11.459Z · LW(p) · GW(p)

I appreciate the offer. The next time I'm in the vicinity of Vancouver (I currently live over 10,000 km away, unfortunately) I'd love to take you up on it.

comment by TheAncientGeek · 2014-11-03T00:14:58.182Z · LW(p) · GW(p)

Thermodynamics explains the arrow or direction of time, not the passing-ness. You can have a static, eternal, arrow.

Replies from: pragmatist
comment by pragmatist · 2014-11-03T12:37:41.491Z · LW(p) · GW(p)

My point was that thermodynamics explains (most of) the experience of passing-ness. Or, more precisely, it explains why our cognition is set up to process time dynamically, with memories accumulating as we grow older, and with the future appearing to be "open". I don't believe there's any objective phenomenon of passing-ness, so that's not something that needs to be explained.

Replies from: TheAncientGeek
comment by TheAncientGeek · 2014-11-03T14:37:27.866Z · LW(p) · GW(p)

Even if passimgness is not an objective reality, it is a subjecrive illusion, and that is no easier to explain.

An eternalist theory can't explain why we process dynamically, since it is exclusive of dynamism.

Eternalism can account for an arrow of time , and therefore why I don't have memories of time T4 at moment T3. What it can't account for is why I don't have simultaneous direct awareness of T2 and T3 and T4. Saying that that is because they are different times answer the question easily under the A series theory, because T4 does not exist at T3. But if all times co exist, why don't I have consciousness of all of them? If I am a 4D entity, why don't I have a 4D consciousness ?

Replies from: pragmatist
comment by pragmatist · 2014-11-03T16:36:21.045Z · LW(p) · GW(p)

Consciousness and other brain states are not properties of you, the 4-D entity. They are properties of particular time slices of that 4-D entity, and different time slices can instantiate different properties. So it is a category error to ask "Why don't I have consciousness of all times?" if by "I" you're referring to a 4-D entity and not an individual time slice. 4-D entities are not the sorts of things that are conscious. Time-slices of (certain) 4-D entities are. I don't even know how to answer your final question, because I have no idea what you mean by "4-D consciousness".

Now maybe we can re-word your question like so: "If all times co-exist, why don't my time-slices have consciousness of all of them?" But the answer to this question is simple: Your time-slices do have consciousness of all of them (at least, all of the ones between your birth and death). It's just that different time-slices are conscious of different times; each one is conscious of the time at which it is located. It is (currently) true that my time-slice from two hours ago (who is located at a time two hours in the past of the time I'm typing this) is conscious of that time.

An analogy: Suppose you invent a time machine. You use the time machine to travel three years into the past. As a consequence, there are now two spatially separated "yous" located at the same time -- the "you" from three years ago and the "you" who just time-traveled. Let's say the former is in England and the latter is in Greece. Surely you wouldn't expect each one of the "yous" to have some sort of combination of England and Greece experiences. You'd expect the one in England to have England experiences and the one in Greece to have Greece experiences. For similar reasons, the eternalist should expect each time-slice to have experiences specific to its own (space-time) location, not some combination of the experiences of all time-slices. And, indeed, that is what we observe. So I don't think there's any deep mystery here.

To explain the phenomenology of passingness, then, reduces to a problem of explaining why each time slice identifies strongly with other time slices, and the particular structure of that identification. The way in which my current time-slice's psychological states are related to those of past time-slices is very different from the way in which they are related to those of future time-slices. Also, the way they are related to proximal past time-slices is different from the way in which they are related to those of distal time-slices. I think it is analyzing the nuances of these cross-temporal relationships that will get us farthest towards understanding the phenomenology of temporal passage (at least, as far as we can get without a solution to the hard problem of consciousness). A-theory and presentism, by contrast, do not help me understand passage at all, so I really don't see their advantage in this regard.

Replies from: TheAncientGeek
comment by TheAncientGeek · 2014-11-05T12:56:46.519Z · LW(p) · GW(p)

Consciousness and other brain states are not properties of you, the 4-D entity. They are properties of particular time slices of that 4-D entity, and different time slices can instantiate different properties. So it is a category error to ask "Why don't I have consciousness of all times?" if by "I" you're referring to a 4-D entity and not an individual time slice. 4-D entities are not the sorts of things that are conscious. Time-slices of (certain) 4-D entities are. I don't even know how to answer your final question, because I have no idea what you mean by "4-D consciousness".

A 4D consciousness is what you would naturally expect a 4D entity to passed, and a 4D entity is what you would naturally expect to find in a 4D universe.

Empirical, entities seem to be 3D, and consciousness, where present, seems to supervene on 3D states...seems empirically that is. Fully fledged empiricism would not only suggest 3D consciousness, but presentism ... there us no direct evidence of past and future start. Fully fledged 4Dism would suggest 4D consciousness. What you have put forward is a compromise. 3D consciousness is a natural consequence of presentism, because there is only a 3D slice to available to supervene on. 3D consciousness is not a natural consequence of Eternalism.

It's not the only compromise position either...the growing block universe is also a compromises.

Replies from: pragmatist
comment by pragmatist · 2014-11-05T18:25:02.076Z · LW(p) · GW(p)

On reflection, I disagree that consciousness seems to supervene on 3D states. Consciousness seems to me to be a necessarily temporally extended phenomenon. I don't know what it would mean for an object that only existed for a single instant to be conscious.

Consider this thought experiment: Suppose you were able to freeze me right now, so that all processes in my body halted. In particular, my brain remained frozen in its current state, with nothing happening in it. Would this frozen version of me be experiencing things continuously? It seems to me that both science and intuition strongly suggest that the answer is "no", that the frozen me would be unconscious. Yet if you believe that consciousness supervenes on 3-D states, then you would have to say that the answer is "yes". Each 3-D slice of that frozen person is identical to my current 3-D slice (or the 3-D slice of me a few moments ago, to be more accurate), and there's no denying that I am currently experiencing things. If those experiences supervene on my 3-D state, then they must be shared by the frozen version of me, but that seems pretty counter-intuitive to me.

So in a sense I agree that consciousness is a property of 4-D entities. Not the entire 4-D entity corresponding to a person's space-time worm, but of thin (but not infinitesimally thin) slices of that worm.

Also when you say that empirical entities seem to be 3-D, you are presumably talking about prima facie appearance. Our most sophisticated empirical understanding of the world suggests that those entities are 4-D, since presentism (and even endurantism) has huge problems reconciling itself with relativity. It would be a poor empiricist who placed more credence in prima facie empirical evidence than the carefully refined empirical evidence provided by our best science.

Replies from: TheAncientGeek, Jiro
comment by TheAncientGeek · 2014-11-09T17:36:47.127Z · LW(p) · GW(p)

Relativity, however, is a bad match for quantum mechanics. The growing block universe is, on the other hand, a good match for the objective reduction. I agree with Rovelli that each major physical theory has a different picture of time.

I also agree that we should expect consciousness to supervene on small but nonzero stretches of activity.

http://lesswrong.com/lw/1jm/getting_over_dust_theory/88rs

However, this still isn't a hold match for full strength Eternalism.

comment by Jiro · 2014-11-05T21:41:17.077Z · LW(p) · GW(p)

Each 3-D slice of that frozen person is identical to my current 3-D slice

The numerical value of each part of that slice is equivalent to the numerical value of the same part of your current slice, but the derivative is not.

comment by Tyrrell_McAllister · 2014-11-01T17:37:24.311Z · LW(p) · GW(p)

It is true that the word "theory" is used with different meanings in different contexts.

Replies from: shminux
comment by shminux · 2014-11-01T19:12:24.631Z · LW(p) · GW(p)

What is the meaning of "theory" in philosophy?

Replies from: Tyrrell_McAllister
comment by Tyrrell_McAllister · 2014-11-01T19:18:47.166Z · LW(p) · GW(p)

I don't know how to give an intensional definition. But the use of the word "theory" in discussions of the A-theory vs B-theory and Eternalism vs. Presentism gives an ostensive definition.

comment by TheAncientGeek · 2014-11-03T00:15:57.283Z · LW(p) · GW(p)

If the b theory is true, and can't even explain even an illusion of passing time then there's your test.

comment by MaximumLiberty · 2014-11-01T14:49:20.836Z · LW(p) · GW(p)

I'm not sure why, but I use A-series for epistemology and B-series for metaphysics. That's probably deeply wrong somehow, but it fits with a strong belief in the fallibility of both memory and prediction.

comment by bramflakes · 2014-11-01T01:25:12.205Z · LW(p) · GW(p)

I had to re-read this several times before I understood the point of what you were saying. It has a lot of important things missing, in particular:

Why bring time into this? Intuitions 1 and 2 collide in the same way regardless of whether the two mind-states are causally connected (e.g. I'd still feel that a sufficiently-similar-to-me simulation in some place outside my light cone is still me, somehow, even though I don't have any of his qualia).

How does B-theory solve this problem?

Replies from: Tyrrell_McAllister
comment by Tyrrell_McAllister · 2014-11-01T17:48:48.881Z · LW(p) · GW(p)

Why bring time into this?

I'm not sure that I understand the question. The post started out being about time. Time wasn't "brought into it".

Intuitions 1 and 2 collide in the same way regardless of whether the two mind-states are causally connected. [...] How does B-theory solve this problem?

I think you're right that the intuitions collide regardless. In my experience, B-theorists reject Intuition 1. B-theorists incline to say that you are a temporally extended object, and that your present awareness is just one time-slice of this 4D object. That is, they allow that you contain, within your 4D extension, several self-aware parts that are not aware of one another. Two such self-aware parts are you-today and you-tomorrow.

(Here, I'm using "aware" to refer to the kind of immediate awareness that you have of your current experiences. Normally, even the most vivid memories lack this sense of immediateness and so are easily distinguished from present experience.)

Replies from: bramflakes
comment by bramflakes · 2014-11-01T21:07:37.210Z · LW(p) · GW(p)

B-theorists incline to say that you are a temporally extended object, and that your present awareness is just one time-slice of this 4D object. That is, they allow that you contain, within your 4D extension, several self-aware parts that are not aware of one another. Two such self-aware parts are you-today and you-tomorrow.

I'm an A-ist and that's what I think. I don't see how it's incompatible with A-ism.

Replies from: Tyrrell_McAllister
comment by Tyrrell_McAllister · 2014-11-01T21:32:33.687Z · LW(p) · GW(p)

I probably should have used the terminology of "presentism" and "eternalism" instead of "A-theory" and "B-theory". Do you consider yourself to be a presentist?

comment by buybuydandavis · 2014-11-01T03:43:56.920Z · LW(p) · GW(p)

If time were topologically a loop, wouldn't both A and B theories be inaccurate representations of time?

They'd still work locally, but not globally.

Replies from: pragmatist, shminux, TheAncientGeek
comment by pragmatist · 2014-11-01T06:27:49.378Z · LW(p) · GW(p)

I don't think B-theory, broadly construed, would be incompatible with that. B-theory is basically just treating space-time as an integrated entity, rather than having radically different ontologies for space and time. It doesn't require time to be partially ordered, any more than it requires spatial directions to be partially ordered.

Replies from: buybuydandavis
comment by buybuydandavis · 2014-11-01T21:47:16.700Z · LW(p) · GW(p)

It doesn't require time to be partially ordered

I would think it does. Time without ordering aint much like time to me. YMMV.

But that was my point. You don't get well defined global ordering if time loops.

comment by shminux · 2014-11-02T18:49:47.926Z · LW(p) · GW(p)

There are spacetimes, most notably the Godel Universe, where time is both a loop and not a loop. Some time directions (from the same spatial point) loop around, others go on forever, never getting close to the point of origin. You can still say, however, that it is compatible with the timeless B-like approach. There is no way to construct this spacetime by taking an A-like fixed spatial slice and evolving it forward in time using the Einstein equations. (There is a loophole, however, where one can lift the Godel universe from flat space using a twisted timelike fiber, but that's a different story.)

comment by TheAncientGeek · 2014-11-03T00:28:19.885Z · LW(p) · GW(p)

One has becoming. The other doesn't.

comment by SilentCal · 2014-11-05T22:41:38.170Z · LW(p) · GW(p)

I haven't actually spoken to a lot of people about their philosophy of time, but my best guess for why one would develop a strong emotion on the topic is that when one first encounters the distinction, one identifies one of the theories as "common sense" and the other as "counterintuitive philosophy", and they have a strong emotional disposition one way or the other regarding that dichotomy.

(I'm not sure there's a general answer to which theory is common sense, but I think it's likely one would make an identification one way or the other)

comment by buybuydandavis · 2014-11-01T21:58:08.362Z · LW(p) · GW(p)

I'm trying to understand those A-theorists who aren't bothered by the implications of the B-theory for free will.

You're trying to figure out what the payoff is in believing A-Theory, if it's not being used to solve some conceptual tizzy over free will, or are you trying to figure out how they manage not to have a tizzy over the implications of B-Theory for free will?

Replies from: Tyrrell_McAllister
comment by Tyrrell_McAllister · 2014-11-01T22:07:52.529Z · LW(p) · GW(p)

You're trying to figure out what the payoff is in believing A-Theory, if it's not being used to solve some conceptual tizzy over free will, or are you trying to figure out how they manage not to have a tizzy over the implications of B-Theory for free will?

The first option. The second issue isn't so mysterious to me, because the usual compatibilist arguments seem to be just as available to the A-theorist as to the B-theorist. (... and just as necessary, since there are still compelling physical arguments for determinism even if you're an A-theorist.)