Timeless Beauty
Followup to: Timeless Physics
One of the great surprises of humanity’s early study of physics was that there were universal laws, that the heavens were governed by the same order as the Earth: Laws that hold in all times, in all places, without known exception. Sometimes we discover a seeming exception to the old law, like Mercury’s precession, but soon it turns out to perfectly obey a still deeper law, that once again is universal as far as the eye can see.
Every known law of fundamental physics is perfectly global. We know no law of fundamental physics that applies on Tuesdays but not Wednesdays, or that applies in the Northern hemisphere but not the Southern.
In classical physics, the laws are universal; but there are also other entities that are neither perfectly global nor perfectly local. Like the case I discussed yesterday, of an entity called “the lamp” where “the lamp” is OFF at 7:00am but ON at 7:02am; the lamp entity extends through time, and has different values at different times. The little billiard balls are like that in classical physics; a classical billiard ball is (alleged to be) a fundamentally existent entity, but it has a world-line, not a world-point.
In timeless physics, everything that exists is either perfectly global or perfectly local. The laws are perfectly global. The configurations are perfectly local—every possible arrangement of particles has a single complex amplitude assigned to it, which never changes from one time to another. Each configuration only affects, and is affected by, its immediate neighbors. Each actually existent thing is perfectly unique, as a mathematical entity.
Newton, first to combine the Heavens and the Earth with a truly universal generalization, saw a clockwork universe of moving billiard balls and their world-lines, governed by perfect exceptionless laws. Newton was the first to look upon a greater beauty than any mere religion had ever dreamed.
But the beauty of classical physics doesn’t begin to compare to the beauty of timeless quantum physics.
Timeful quantum physics is pretty, but it’s not all that much prettier than classical physics. In timeful physics the “same configuration” can still have different values at different times, its own little world-line, like a lamp switching from OFF to ON. There’s that ugly t complicating the equations.
You can see the beauty of timeless quantum physics by noticing how much easier it is to mess up the perfection, if you try to tamper with Platonia.
Consider the collapse interpretation of quantum mechanics. To people raised on timeful quantum physics, “the collapse of the wavefunction” sounds like it might be a plausible physical mechanism.
If you step back and look upon the timeless mist over the entire configuration space, all dynamics manifest in its perfectly local relations, then the “pruning” process of collapse suddenly shows up as a hugely ugly discontinuity in the timeless object. Instead of a continuous mist, we have something that looks like a maimed tree with branches hacked off and sap-bleeding stumps left behind. The perfect locality is ruined, because whole branches are hacked off in one operation. Likewise, collapse destroys the perfect global uniformity of the laws that relate each configuration to its neighborhood; sometimes we have the usual relation of amplitude flow, and then sometimes we have the collapsing-relation instead.
This is the power of beauty: The more beautiful something is, the more obvious it becomes when you mess it up.
I was surprised that many of yesterday’s commenters seemed to think that Barbour’s timeless physics was nothing new, relative to the older idea of a Block Universe. 3+1D Minkowskian spacetime has no privileged space of simultaneity, which, in its own way, seems to require you to throw out the concept of a global now. From Minkowskian 3+1, I had the idea of “time as a single perfect 4D crystal”—I didn’t know the phrase “Block Universe”, but seemed evident enough.
Nonetheless, I did not really get timelessness until I read Barbour. Saying that the t coordinate was just another coordinate, didn’t have nearly the same impact on me as tossing the t coordinate out the window.
Special Relativity is widely accepted, but that doesn’t stop people from talking about “nonlocal collapse” or “retrocausation”—relativistic timeful QM isn’t beautiful enough to protect itself from complication.
Shane Legg’s reaction is the effect I was looking for:
“Stop it! If I intuitively took on board your timeless MWI view of the world… well, I’m worried that this might endanger my illusion of consciousness. Thinking about it is already making me feel a bit weird.”
I wish I knew whether the unimpressed commenters got what Shane Legg did, just from hearing about Special Relativity; or if they still haven’t gotten it yet from reading my brief summary of Barbour.
But in any case, let me talk in principle about why it helps to toss out the t coordinate:
To reduce a thing, you must reduce it to something that does not itself have the property you want to explain.
In old-school Artificial Intelligence, a researcher wonders where the meaning of a word like “apple” comes from. They want to get knowledge about “apples” into their beloved AI system, so they create a LISP token named apple. They realize that if they claim the token is meaningful of itself, they have not really reduced the nature of meaning… So they assert that “the apple token is not meaningful by itself”, and then go on to say, “The meaning of the apple token emerges from its network of connections to other tokens.” This is not true reductionism. It is wrapping up your confusion in a gift-box.
To reduce time, you must reduce it to something that is not time. It is not enough to take the t coordinate, and say that it is “just another dimension”. So long as the t coordinate is there, it acts as a mental sponge that can soak up all the time-ness that you want to explain. If you toss out the t coordinate, you are forced to see time as something else, and not just see time as “time”.
Tomorrow (if I can shake today’s cold) I’ll talk about one of my points of departure from Barbour: Namely, I have no problem with discarding time and keeping causality. The commenters who complained about Barbour grinding up the universe into disconnected slices, may be reassured: On this point, I think Barbour is trying too hard. We can discard t, and still keep causality within r.
I dare to disagree with Barbour, on this point, because it seems plausible that Barbour has not studied Judea Pearl and colleagues’ formulation of causality—
—which likewise makes no use of a t coordinate.
Pearl et. al.’s formulation of “causality” would not be anywhere near as enlightening, if they had to put t coordinates on everything for the math to make sense. Even if the authors insisted that t was “just another property” or “just another number”… well, if you’ve read Pearl, you see my point. It would correspond to a much weaker understanding.
Part of The Quantum Physics Sequence
Next post: “Timeless Causality”
Previous post: “Timeless Physics”
- Causal Universes by (29 Nov 2012 4:08 UTC; 125 points)
- The Quantum Physics Sequence by (11 Jun 2008 3:42 UTC; 65 points)
- Timeless Identity by (3 Jun 2008 8:16 UTC; 54 points)
- Timeless Physics by (27 May 2008 9:09 UTC; 54 points)
- Timeless Causality by (29 May 2008 6:45 UTC; 46 points)
- Timeless Control by (7 Jun 2008 5:16 UTC; 41 points)
- Quantum Mechanics and Personal Identity by (12 Jun 2008 7:13 UTC; 12 points)
- [SEQ RERUN] Timeless Beauty by (19 May 2012 19:11 UTC; 6 points)
- 's comment on Einstein’s Superpowers by (31 May 2008 2:35 UTC; 2 points)
- 's comment on Timeless Beauty by (28 May 2008 18:17 UTC; 0 points)
Okay, now explain how you can have CP-violation but not CPT violation in a timeless physics. ;)
“I wish I knew whether the unimpressed commenters got what Shane Legg did, just from hearing about Special Relativity; or if they still haven’t gotten it yet from reading my brief summary of Barbour.”
Hard to say. I don’t really see the difference between “time is ‘just’ a coördinate in 3+1-dimensional spacetime” and “time really doesn’t exist.” Even if we can get rid of the t in our equations (because we never personally observe a t out there in the world, but infer it from our memories and clocks and such), something still has to account for our memories, and clocks, and the apparent changes in what we perceive: for things to be otherwise would be a violation of Egan’s Law. I don’t see why it matters whether we call this whatever-it-is “causal relations within configuration space” or whether we give it its own coördinate and call it time.
According to Barbour:
Platonia isn’t 3+1It’s a branching, .MW type structure.
Our memories aren’t causal traces.
There’s no reason we should remember coherent causal histories, rather than Alice in Wonderland, but we do. (Well, he handwaves about consciousNess of the gaps at that point)
I have also argued in the past that time is not a 4th dimension. Things exist and they move. We watch that movement and mentally discard aspects such as it’s path (straight line, arc) or type (swimming, flying), leaving only the progression aspect, and over all the entities we call this “time.”
Because it is the progression that indicates the passage of time and not the positions at any given “moment,” if someone snapped their fingers and instantly moved all objects to their 100 years hence positions, it would not be the future, because no progression of motion took place. If they “fast-forwarded” it then yes, it would be the future.
I beg to differ. Everybody would remember the motion having taken place; the history of that 100 years would be recorded. There is no way in principle to experimentally distinguish this occurrence from the normal progression of time by 100 years, so I claim they are the same.
There is a nice philosophical thought-experiment that points to the opposite position:
Suppose that some phenomena occurs (purple light is the original I believe) now and then in various places in the universe. Whereever it happens, a day later that region is “frozen in time” for an “external” week. The phenomenon then occurs everywhere. There is no way to tell whether a week of time has passed or not, but it’s a shorter theory, a better compressed description, that says that it has been frozen for an external week in this circumstance, that a week of time passed that no one experienced.
The similarities with snapping fingers to move objects to their future positions should be clear.
Even including Harry Potter and his sudden ability to move particular objects discontinuously 100 years into the future by snapping his fingers, my claim stands. The point is regarding the instantaneous movement of every part of the universe to its future position, in which case inhabitants of the universe will see the signal (fingers snapping) and see nothing out of the ordinary happen. These observers will even continue to observe what happens throughout the next 100 years, or at least it will be indicated as such with 100% complete consistency in any and all records present at the end of those 100 years, including the memories of every living being. The only difference when including Harry in the picture is that our fundamental description of the physical laws change; when the whole universe is moved, not a single one of their consequences is distinguishable from time progressing normally, thus they are still equivalent statements. By introducing unphysical Harry, we develop a way to distinguish the two explanations, but this is irrelevant to our reality.
Peeve: it’s “et al.”
In Barbour’s philosophy, if only a single Now existed, could it be conscious? Or would it not be conscious until there were enough to put in a row along something time-ish?
Getting rid of time seems to require a global space of simultaneity to slice along, while a block universe doesn’t; but I suppose Barbour addresses this, as well as the trouble of eliminating time but not space from GR.
I’d also like to hear something about CP-violation, even if only “Barbour answers it” or “nobody knows”.
I did; or, rather, I think I internalized timelessness gradually (unlike Shane’s shock), from a more psychological than physical starting point, before ever hearing of Barbour or his ideas. I didn’t see the aesthetic argument against collapse until this post, but it actually seems to depend on a block-universe rather than unordered-pile-of-timeslices view.
Z.M.Davis: maybe I’m misreading (and I know I’m not yet mathematical enough to follow this properly) but the difference seems to be that epiphenomenal time goes away. “X is a full description of the configuration of the universe and t is the time” can be shortened to just X, because X is unique.
I’m trying to find a way to explain why this freaks me out so much...
If time is just another dimension in our model then that’s no big deal: here we all are cruising along the time dimension, t = 76 and then one second later t = 77 and so on. Saying that time is the t parameter which is a dimension is just mathematical notation; it doesn’t actually get rid of time. Intuitively, we can think of the value of t increasing over time, and thus we haven’t actually defined time at all, we have only represented it by t.
If we switch to this “static crystal” view of things, time is now just a relationship between things along one of the dimensions. There is no “now”, nor a “future”, nor a “past” and time doesn’t “flow”. We can’t even talk about “moving” along the time dimension because movement would be a change over time. We have stepped outside of time and completely defined it within our model. Inside the model it feels like we are moving through time because of the properties of the time dimension relation: one effect of which is that the “past” is partly represented in the current state, for example as memory, as well as the expected “future”. It all simply and naturally comes from the relation on the structure of the crystal.
I’m no physicist, however when something mysterious can be brought into a model in a way that is simple and makes it no longer mysterious, well it has a certain ring of truth to it.
But to accept this at an intuitive level, for example, the way in which the mystery of the present “now” can be seen as a natural illusion deriving from a simple relationship… I’m not joking when I say that thinking about this makes me feel a bit weird.
You make an interesting point that I hadn’t even considered. In my “current” mind, I have representations of past configurations stored as “memories” and representations of future configurations stored as “expectations”. Both memory and prediction are incredibly imprecise so… if it weren’t for us constantly updating our mental models of past configurations from external sources (observations of the effects of “past” causes), there wouldn’t be all /that/ much difference experientially if: snap causality is now reversed, time now moves backwards, arrow of time *= −1, however you want to describe it.
I wonder why Eliezer wants to keep causality. Is there some detectable difference if causality exists in the world, compared to the situation where it doesn’t exist?
I might imagine that causality exists in my conscious experience: each configuration of particles in my brain “leads to” the next.
Or I might imagine that each configuration of particles in my brain exists as a momentary fluctuation in a universe already subject to heat death: yet each of the configurations is such that it seems to remember another configuration (which did not in fact produce it), and such that it seems to expect another configuration (which it will not in fact produce.
Obviously, there is in principle no way for me to detect a difference between these two situations. And if I extend the situation to include others, there will be no difference that anyone at all can ever detect, even in principle. By Eliezer’s principles, should we not invent a new physics where the two possibilities are one and the same?
If you throw away time and .causality, what do experiments mean? See Dowkers critique of Barbour.
Steven: Wrong question. “nows” aren’t slices of “Block Universe” such that there could coherently be “only a single now”. Nows are mathematical relationships in configuration space, which as a large set of mathematical relationships connected by internal inferential flow couldn’t be other than how it is.
Shane: I would say that Special Relativity imposed locality and thus destroyed the Newtonian “Static Crystal Configuration Space” view of quantum mechanics and that Barbour is trying to restore it with a relational quantum mechanics.
I’m still trying to wrap my non-physicist brain around this.
Okay, so t is redundant, mathematically speaking. It would be as if you had an infinite series of numbers, and you were counting from the beginning. The definition of the series is recursive, and defined as such that (barring new revelations in number theory) you can guarantee it will never repeat. As a trivial example, { t, i } = { 1, 1.1 }, { 2, 1.21 }, { 3, 1.4641 }.… t is redundant, in the sense that you don’t need it there to calculate the next item in the series, and subtracting it makes the definition of the series simpler.
I also keep thinking back to Conway’s game of life—the time parameter (or generation, at least in xlife) is superfluous to a description of the “universe”. The cells automatize (?) identically regardless of the generation. It’s only the actual description of the playfield, combined with the rules for creating the next generation, that could be said to “exist” (at least for a glider physicist).
But in both those things there’s still a concept of “successive states”, and a “before” and “after” situation. It’s the (allegedly) objective label for progress through successive states that’s redundant. The idea of a block or crystal universe being “real” seems like a map/territory confusion, at least the way I’m understanding it—that you could statically load up all the states of the universe into the memory of a sufficiently powerful n-dimensional computer, and that alone (with no processing per se) would be sufficient to create our experiences of existing inside the universe.
Unknown—surely it can’t be a coincidence that we feel like we experience things happening in an order that implies causality?
Shane—get The End Of Time, it’s very well written. At first I thought ‘get out’, but the more you read, the more you think, the more it stands up. And yes, scary stuff.
iwdw—I can’t remember which, but one of Brian Greene’s books had a line that convinced me that all the configurations do exist simultaneously: “The total loaf exists”. How can anything that crazy-sounding not be right?
Ben, actually as soon as I posted I realized that I had contradicted my argument against the dust theory: in other words, according to this, there would be a reason that we seem to have a continuous conscious experience passing through time, but there would be no reason for us not to see strange events like cars turning into elephants and so on, as long as it did not kill us. And since there would be vastly more such possibilities than ordinary possibilities, we could expect such things with a high degree of certainty, even in the next ten seconds.
So actually it turns out that Eliezer is probably right about causality.
iwdw: there has been some thinking about the universe as an actual game of life, Steven Wolfram’s New Kind of Science is the one that comes to mind, but I’m sure there are more reputable sources that he stole the idea from. I believe that this thinking runs into trouble with special relativity.
Speaking of which, has anyone ever attempted to actually model space as a graph of relationships between points, in a computer program? Something like the distance-configuration-space in the last post? It occurs to me that this could actually be a more robust representation for some purposes than just storing the xyz coordinates.
Eliezer: I actually have been getting the insights you speak of repeatedly throughout this series, and it’s one of the reasons why I find it helpful to post comments—because it forces me to think through the ideas well enough to get their occasional mind-bendingness. It’s also why I have continued reading despite all the what-is-Science business.
But I still think that the subjective time-like-ness of time, as well as the concept of causality, are all caused (ha-ha) by the universe starting out in a low-entropy state. So if you had a toy block universe in your hands, you would still see a direction in the block corresponding to time. There is no way to assign a meaningful distance in that direction for the whole universe because of the locality of physics, but the direction is global, isn’t it?
What is it about this subject that makes it impossible for you people to grasp its implications properly? I am unable to comprehend your incomprehension.
And why haven’t you noticed that the premises of Dust are incontrovertible?
Michael V, I agree with “all mathematical structures exist” type of ideas, but it still seems to me the question is meaningful; it seems like I could represent all the data for one Now into a computer without doing the same thing for all Nows.
Caledonian: why isn’t that true?
Actually ignore the words “I … but” in my previous post, they’re not relevant.
I can’t remember which, but one of Brian Greene’s books had a line that convinced me that all the configurations do exist simultaneously: “The total loaf exists”. How can anything that crazy-sounding not be right?
I’m not sure that taking the crazy-sound of a given statement as positively correlated with it’s truth is a useful strategy (in isolation). :-)
I guess I’m not sure what “exists” even means in this context. Is this in the general sense that “all mathematical objects exist”? I don’t know what sin(435 rad) is offhand, but I know that it’s defined (i.e. that it exists). But that’s very different from actually instantiating it in the memory of an HP48G.
I accept that all states of the universe are defined, in the mathematical sense (in that, given the parameters of existence, they couldn’t be different than they are—given that F(0)=0 and F(1)=1, F(20) can’t not be 6765). But the mathematical definition of something, and the instantiation of it (or “real existence”) seem to be distinctly different things.
Caledonian: I too have kinda the same issue with the dust hypothesis. My natural instinct was to go “yes, of course the dust hypothesis is true, it seems impossible it could be otherwise!”
But then, how do I explain that even though there’re far more chaotic configurations that contain me than apparently consistent configurations that contain me, I seem to find myself in more ordered state?
That is, I seem to note that the bits of my perceptions corresponding to recent memory and bits of my perception corresponding to what I’m observing in front of me seem to usually fit together. Further, I notice that my perception of my memories themselves seem to have a bit more structure then one’d otherwise expect if the vanilla dust hypothesis were true.
It would seem that, like it or not, there’s definately something missing from the “dust hypothesis” picture of reality.
The question about the Born statistics seems to be a similar type of question. Perhaps it’s pretty much the same question, in fact. The vanilla Dust Hypothesis would seem to completely and utterly reject things like the Born rule.
You have to look not only at the individual, infinitesimal moments, but the connections between the slices through time. You only exist as a process, and a process only exists along a timelike axis.
Any coherent perspective has an infinite number of ordered levels of implementation below it. Any disordered level has no levels above it.
Thants not-dust. If consciousness supervenes on coherent structures of multiple Now’s, that might solve the Alice in Wonderland problem.
Caledonian, you are completely wrong. Even if having an experience requires more than a single instant, it surely does not require more than a second: and a brain could arise and persist even for a full second, with sufficient luck, within a heat dead universe (or in other chaotic circumstances.) So you can look at a sequence of brains: brain 1, having experience A; brain 2, remembering experience A and having experience B; brain 3, remembering experience A & B and having experience C; and so on. There will be many, many, more sequences of such brains that remember incredibly disordered events than sequences that remember ordered events.
The fact that we do not have such experiences shows that the dust theory is false, whether you like it or not, and whether or not you think its premises are incontrovertible.
Upvoted for correctness. Consciousness does not appear to supervene on subjectively long periods, but subjectively short periods are still objectively long, ie trillions of microphysical events can occur during the 0.1 s or so that constitute a subjective Now (specious present).
Related, I’ve been wondering something else.
Given our current level of technology (TL7 going on 8), is it even possible to simulate a universe computationally (the configuration space of the universe, whatever)?
If the wave equation is the distribution over a configuration space with respect to an arbitrary reference frame (i.e. “time”), then what “really exists” (again, not clear on what that means in this context) is an underlying configuration space. Do we know enough about that to represent one to the extent that we could create a minuscule universe that behaves structurally like our own?
May 28, 2008 at 02:15 PM was me. Typekey lied to me.
Z. M. Davis:
Hard to say. I don’t really see the difference between “time is ‘just’ a coördinate in 3+1-dimensional spacetime” and “time really doesn’t exist.” … something still has to account for our memories, and clocks, and the apparent changes in what we perceive: for things to be otherwise would be a violation of Egan’s Law.
That something is the Second Law of Thermodynamics. The only arrow of time which is not explained by thermodynamics ( + quantum decoherence) + anthropic principle is a CP violation in certain subatomic interactions. Hopefully Eliezer’s next post will explain this more clearly.
I am unable to understand why you can’t understand this incredibly simple point. Do you imagine that the nature of the implementing hardware changes the behavior of an algorithm? Do you think a program run on a ‘virtual’ computer behaves differently than one run on an equivalent ‘real’ computer?
If I run a Java program that implements Conway’s Game of Life, do you think it will behave any differently when I encode a computational system in, let’s say, a hive of bees, and use it to implement that program?
You don’t have any evidence that conscious experience supervenes on objectively instantaneous moments.
You don’t have any evidence that conscious experience supervenes on algorithms rather than underlying physical activity.
The two claims are incompatible, since alalgorithms take some time to run.
edit PaleMoon lost original reply. I will try to recreate it :(
Not saying you’re incorrect in criticizing the above(the two claims do seem incompatible), but isn’t it the case that algorithms are just structures and that only they take time only to run? What I mean is that within the block-universe view there would be structures that we would be in ignorance of their nature and in order for us to learn about them we might have to count them (and since we are living in a timeline with computers that operate per cycle our accounting of them would take some time to complete), but that it’s only ignorance to an observer like us that necessitates this? That if you knew some property of some local region of the block-universe you could use it to estimate some other property via the algorithm that represents their (mutual) structure, but that the algorithm describing their structure merely is. There’s plenty of times when choosing algorithms to describe mathematical objects that we choose algorithms that fall along a space-time tradeoff, so it stands to reason that there should be a ‘all-space’ choice that only encodes the answer we seek in structure alone.
I don’t see how that relates to the supervenience of experiential states on instantaneous brain states.
The parent made 3 claims(the 3rd one was snuck into the conclusion). I only addressed 2 and 3. 1 is a credible point that stands on its own merit. Without points 2 and 3 however with 1 it’s no longer a sound argument.
Caledonian, once again, if you have algorithms for all possible sequences of transitions on a grid like that of Conway’s game of life, there will be far, far more sequences that are incredibly disordered and which do not follow any obvious rule, than sequences that follow Conway’s rules.
Well, duh.
Conscious beings do not exist in those rule systems, Unknown. Your objection is inane.
I thought the dust theory was that conscious beings exist in every system.
Eliezer, if you believe all of this, why do you care so much about saving the world from “future” ravenous AIs? The paperclip universes just are and the non-paperclip-universes just are. Go to the beach, man! Chill out. You can’t change anyting; there is nothing to change.
His caring just is..
Actually that is excessively glib. Regrets and fears are posited on things being able to be different, being capable of change. The irony is that single universe plus indetermimistic collapse theories support ES moral attitudes much better than EYs ontology.
Anonymous, I think I sort of get how thermodynamics explains why the arrow of time points the direction it does, but why is there (why does there seem to be) a dimension for the arrow to exist in? Okay, so we can axe the t from the equations, and describe the development of the wavefunction solely in terms of r, but how can the wavefunction develop without something-like-time, whether we decide to call it time or internal relations? Probably I’m just not seeing the reduction yet and future posts will blow my mind. Or I will need to study more maths.
Caledonian, let me guess: you didn’t actually finish Permutation City. See questions 4 and 5 of the “Dust Theory FAQ.”
Bambi, even though the future is “already” determined, it’s still logically dependent on what we choose to do now. The information that there is a determined fact of the matter of whether or not Eliezer will go to the beach (or in what proportion of Everett branches he will do so), won’t help him when he’s trying to decide whether or not to go to the beach. “It all adds up to …”
bambi: I think this would be related to Newcomb’s Problem? Just because the future is fixed relative to your current state (or decision making strategy, or whatever), doesn’t mean that a successful rational agent should not try to optimize it’s current state (or decision making strategy) so that it comes out on the desired side of future probabilities.
It all sorts itself out in the end, of course—if you’re the kind of agent that gets paralyzed when presented with a deterministic universe, then you’re not going to be as successful as your consciousness moves to a different part of the configuration as agents that act as if they can change the future.
And for those of us who haven’t read Permutation City at all, here’s an explanation of this whole “dust theory” thing they’re talking about.
(The FAQ Z.M.Davis points to has answers to several good questions about dust theory, but not the question “what is it?”)
Borges had a better understanding. Not anywhere near Egan’s grasp of mathematics, of course—but the argument isn’t being made in mathematics. I suspect this is a large part of the problem, since some of you keep making an invalid argument.
Assuming that dust theory or the block universe or Barbourian timelessness are true… I fail to see how it matters to any of us.
Presumably, we are all timeful beings. I know I am (cogito, ergo tempus fugit), and I assume the rest of you are, too. Whether I and my memories and my perception of time passing only exist as collections of block slices or as neighboring nodes in the static quantum foam in configuration space or as relationships between specks of dust… or even as time-slices in a computer simulation, or as integers in MathSpace which is the only thing that really exists… it doesn’t matter. I still perceive time. And I bet you do, too.
If physics experiments and solid reasoning lead us inexorably to conclude that time, identity, and consciousness are mere illusions… well, they also lead us (or lead me, anyway) to conclude that those illusions are impenetrable. It’s impossible for me not to perceive time, to not perceive myself as myself, to not perceive my own consciousness.
What basis, then, is there for saying that time is not “real”?
What value does the concept of dust, blocks, or Barbour bring to our intellectual discourse, other than making for entertaining conversation among stoners?
The problem with dust theory is the part where somebody confuses a simulation with reality. Everything else follows from that initial erroneous step.
Ok, it looks to me like these answers (invoking the future over and over after accepting that there is no ‘t’) are admissions that this type of physics thinking is just playfulness—no consequences whatsoever, to our own actions or to any observable aspect of the universe.
That’s cool, I misunderstood is all. Maybe life is just a dream, eh?
The problem with your understanding of dust theory is the part where you don’t realize simulation and reality are relative to perspective. There is no objective distinction between them.
yes there is: a real world can contain simulations,but a simulation cannot contain a reality.
Okay, I’ve heard people claim that a person and a simulated person could have indistinguishable experiences, but I’ve never had someone claim a person and a computer or a person and a collection of random particles are objectively the same thing. That’s rather like stating that there’s no objective distinction between cars and bananas.
Caledonian, if you cannot manage to post only the substance of your comments and keep the insults and taunts out of them, I will go through and delete them even if others have already responded to them. Consider yourself warned.
Caledonian, if your position is that observers do not exist in sequences that would produce disordered observations, I would have to say that the evidence would lead me to agree with you, at least in the sense that there must exist more sequences where observations are ordered, than sequences where observations are disordered, precisely for the reasons that I have been giving. I don’t if anyone else would consider this to be consistent with dust theory. Certainly it isn’t consistent with my understanding of the theory, or Egan’s for that matter, as Z.M. Davis pointed out.
You’ve never been so intoxicated that you “lose time”, and woken up wondering who you threw up on the previous night? You’ve never done any kind of hallucinogenic drug? You don’t … sleep?
Those things you listed are only true for a fairly narrow range of operational paramaters of the human brain. It’s very possible to not do those things, and we stop doing them every night.
The sensation of time passing only seems to exist because we have short term memory to compare new input against. Disrupt short term memory formation—by, say, getting extremley drunk, or getting a head injury—and you lose the sensation of time passing.
Observing entities do not exist in walks through the Library in which ordered perceptions of themselves do not exist. Postulating an observer therefore requires postulating a minimal amount of order, which is considerably greater than what would arise from a randomly-chosen walk on average.
Order is not sufficient, of course—all walks containing observers are highly ordered, but not all highly-ordered walks are observed.
If you use an even more abstract sense of ‘observer’ and ‘observation’, such as the way they are traditionally used when discussing certain properties of quantum mechanics, then this argument doesn’t hold—but neither does the ability of the argument to address the matter of orderliness in the first place.
Steven: All the data for one now is necessary and sufficient for every other now.
Caledonian: Pure anthropics may explain what we see, and what you are calling dust theory, as you claim, but you are not giving us any evidence, or even an argument, that dust doesn’t generate observers with only partially ordered experiences. I can see the outline of such an argument, but it doesn’t appear to me to be likely to be sound.
You’ve never been so intoxicated that you “lose time”, and woken up wondering who you threw up on the previous night? You’ve never done any kind of hallucinogenic drug? You don’t … sleep?
I have in fact done at least two of the above three. (Perhaps if I slept I wouldn’t need to take drugs so often...)
But you’re taking my words too literally and missing my point. Indeed, it is very possible for me to fail to perceive time; I’ve done it before, and at some point I’ll do it forever. But the very fact that I can sit here, now, and talk about “before” and “forever” and “now” (and “I”) shows that I must be perceiving time. It is not possible that I am not perceiving time—unless I’m a zombie and not perceiving anything. But I’m pretty sure I’m not. And I don’t think you are, either, although I can’t prove it.
The sensation of time passing only seems to exist because we have short term memory to compare new input against.
That’s not the reason the sensation of time seems to exist—it’s the reason the sensation of time does exist. It is the very definition of the perception of time. As I said, this sensation may be an illusion, but it is also indisputably real, and it seems pointless (or rather, I don’t yet see the point) to say there’s no such thing as time simply because we can imagine a block universe or Barbour manifold or what have you.
Flags flap. Wind blows. Minds change. Time moves. We remember. It’s all the same thing.
People who have been in accidents, especially those that are physically shocking (like car accidents that involve sudden, violent stops) tend to lose all recollection not only of the moments right after the blow, but the time before it. The data is lost without being processed or moved to long-term storage. And there are no ‘experiences’ associated with them. It doesn’t take very much disorder to disrupt the processes of conscious experience. And then there’s no observer there! A great deal of continuity is required for a conscious observer to exist along a particular axis of time. It takes very little to violate that continuity.
The complexity of a system necessary to have experiences is far, far greater than complexity (or lack thereof) of the data the system can pick up on in any brief period of time. If we’re going to consider a system that’s partly ordered and partly disordered, the vast majority of them won’t have functional minds perceiving disorderly inputs—there are far more ways the mind can be improperly constructed than the incoming sense data can be scrambled.
And that’s not even addressing the deeper issue that algorithms do not somehow magically perceive the systems implementing them. I keep hearing the objection that dust-borne consciousnesses that might arise within high-entropy systems shouldn’t perceive orderly surroundings, but that’s just silly. The perceived surroundings have absolutely nothing to do with the environment the algorithm is being emulated within, and absolutely everything to do with the internal states of the algorithm. An algorithm that wasn’t processing highly orderly internal states wouldn’t fit the criteria necessary for it to qualify as an observer.
The natural-language arguments being presented have the same flaws, over and over. The claimed conclusions simply do not follow from the specified premises.
I see that my post explaining just why it’s valid is gone. Never mind that, then.
As for your ‘likely to be sound’, you should stop concerning yourself with what’s likely and focus on what’s logical. Soundness isn’t a useful concept here.
As I already said, you are welcome to repost the comment without the little dig at the end. I have already emailed you the text of it. Just put in the substantive arguments, and leave out the “Wrong!” or “How stupid!” or “You’re so illogical!” Otherwise, I may zap your comments whenever I get around to it.
I approve of Eliezer’s response to Caledonian’s insults and taunts because if Calendonian can continue the way he would like to continue then he will eventually attract others commenters who are habitually unecessarily combative, boastful or immoderate, which will eventually repel thoughtful readers from reading the comments. Now on to my comment.
I tentatively agree with Eliezer that taking the t out of Schroedinger’s equation makes it more beautiful although of course for my opinion to be worth very much I would have to spend at least a few months applying the equation to a large variety of physical situations and reflecting on that experience.
I wish Eliezer would start to apply the sophisticated sense of scientific aesthetics that he obviously has to not only models of reality but also systems of value because otherwise there is a good chance that the first engineered intelligence in the age of engineered intelligence will start out with a system of value approximately as ugly as a pre-Galilean theory of physics are ugly IMHO.
Although Eliezer’s plans include potent measures for refining the initial system of value, it would IMHO be less risky to start with a system of value of vastly better aesthetics—in the sense of the word aesthetics that we have been talking about for the last couple of days—than the systems of values that are popular or dominant in the human population at the present time. And I claim to know of systems of value of vastly better aesthetics. And yes, I see the significant risks in my thinking that my taste in values and morals is vastly better than most people’s!
In general, I wish I could persuade some of the people who are able to apply a sophisticated scientific sense of aesthetics to models of reality to drop their systematic unwillingness to apply the same aesthetic sensibility to moral principles and systems of valuing things. Although systems of value differ from models of reality in that there is nothing that could count as evidence for a system of values, considerations like parsimoniousness (Occam’s Razor) and the scientific sense of beauty IMHO apply to both.
I don’t buy this abolition of time at all, but this question of how CP violation appears in Barbour’s scheme seems like a good test of one’s understanding.
The abolition of a time coordinate in quantum gravity is not Barbour’s invention. The usual Schrodinger equation is Hψ = -i/hbar dψ/dt, where the H operator represents total energy (typically, sum of a potential and a kinetic term). But in general relativity, the total energy persistently shows up as zero (gravitational potential energy cancelling out against mass-energy, I believe; I confess I’m just relating this secondhand). So H=0, and there’s no time evolution, just the stipulation that ψ has an H-eigenvalue of zero. Barbour’s contribution is to offer an interpretation of configuration space as a set of “time capsules”, static configurations containing static observers experiencing an illusion of time due to their memories. (I don’t know what meaning Barbour ascribed to the amplitudes, and perhaps he has problems comparable to those suffered by the usual, timeful many worlds interpretation.) So that’s it: H=0 gives you a standing wave in configuration space, and Barbour proposes a timeless variant on many worlds, with variables internal to the cosmic configuration, like cosmological radius, acting as clock variables, proxies for time.
Back to this CP violation problem. It is a theorem that quantum field theories have CPT symmetry, and it is an experimental result that CP symmetry is violated, i.e. that T symmetry is violated, in kaon-antikaon transitions. We actually have a quantum field theory implementing T-violation, the Standard Model, where it’s implemented in the quark mass matrix. The numbers in that matrix would be coefficients of interaction terms in the potential-energy part of H, connecting quark fields and weak-boson fields, I think. It is something like postulating that the amplitude for (quark disappears) (weak interaction happens) (antiquark disappears) is different from the amplitude for (antiquark disappears) (weak interaction happens) (quark appears). In a sense it is quite unmysterious, since there is no mathematical barrier to postulating such an asymmetry, though one would like to know a deeper reason for it; the CPT theorem only says that if you transform those two processes by CPT, the amplitudes for the corresponding processes had better have the same relation.
Now formally it is a straightforward thing to take a quantum field theory and couple it to general relativity. For example, you can just take the QFT’s old H, add a term for scalar curvature, and multiply the whole thing by “sqrt(determinant(metric))”. Set the new H equal to zero, and you now have what should be the equation for a Standard Model universe with Einstein gravity thrown in; and though you probably can’t solve that equation, you can still go ahead and follow Barbour’s procedure in the resulting configuration space.
It seems obvious(?) that the expression of CP violation in the timeless picture will have something to do with those quark/antiquark fields in the CP-violating terms, and I would point out that although my informal description of those terms might seem to make one the time reverse of the other, actually there is an algebraic difference. Algebraically, “X appears” means that you use a “creation operator”, while “X disappears” involves an “annihilation operator”. So understanding the implications of those operators for amplitude gradients in the timeless picture may be the key to figuring this out. Also, since other physical variables act as clocks in the time capsules—proxies for an actual time—kaons and antikaons ought to somehow have a different relationship to the clock variables. If I was seriously trying to figure this out, I’d be thinking at the intersection of those two approaches. (And I’d be doing it using the simplest T-violating QFT I could find, rather than with the full Standard Model.) This is an excellent question to think about, for anyone trying to understand Barbour’s interpretation in detail.
Erratum, I should have written “(quark disappears) (weak interaction happens) (antiquark appears)”. My point is that the algebraic reversal of that expression is different from the time reversal of it, so those two parts of the Hamiltonian do actually say different things about the amplitude gradients in configuration space, even when you think about it “timelessly”.
Bambi: The particular reason for blogging rather than rum is that the math says he blogs here and now. The future isn’t immune to our actions, it is what it is as the result of our actions, which likewise are what they are. We cause it to be in the same manner that the earlier states of a Turing Machine cause the later states to be.
The question that interests me, Michael, is whether a human being’s coming to believe that the future is already determined will make the human being less likely to write the blog post or to help build the spaceship that deflects the civilization-destroying asteroid.
Thinking about it objectively, Richard, it should do neither. If I want to drink fruity rum on the beach, I can. If I want to blog, I can do that too. The fact that the result of this choice already exists further along the block is neither here nor there. It shouldn’t make any difference at all. I’d very much like to hear a sensible counterargument.
‘Do you have free will?’ is the wrongest question of them all.
It should do neither, but people are weird that way. Any excuse to slack off....
Your criticism seems off the mark to me. These AI researchers were trying to automate reasoning. They turned to formal logic, which makes perfect sense—formal logic is just highly disciplined reasoning, so disciplined that a computer can check it. If you’re using formal logic, of course your symbols don’t mean anything of themselves; all the meaning is in the axioms, a.k.a. the symbols’ network of connections to each other.
A better criticism would be that they stuck to deductive logic—which only tells you when you can know something with absolute certainty—and took decades to realize that most of the useful real-world knowledge was probabilistic. One can’t formalize the concept of an apple because it is not well-defined, anymore than a blegg is.
“In timeful physics the same configuration can still have different values at different times, its own little world-line, like a lamp switching from OFF to ON.” And after that I understood it! The beauty of timeless in that we don’t have conception of CHANGE of the SAME object, it’s oxymoron that we delete. In classic physics how we define something ONE object/configuration if it DIFFERENT at different times, these definitions are arbitrary, classic physics create paradox of Theseus ship, timeless disappoint it. Before this moment I don’t understand why timeless physics is better, not just good alternative, but now I see that timeful physics looks as crutched as collapse postulate. How with multiverse it isn’t just interpretation that just looks better, it is more right model, because we don’t have fluids of same object. It’s how “no individual particles”, but about time, not about space. This post and all sequence should be more liked.