# Julian Barbour on does time exist

Gold Member
A thread was started on how unreal is space and time...I recently came across the promotion of a talk by Julian Barbour - http://www.perimeterinstitute.ca/Outreach/Public_Lectures/Public_Lectures/ [Broken] (although Barbour has never held any academic position, he is accredited by the GR/LQG community for teaching them all that GR is a relational theory)

His talk was on

"Many attempts to create a unified theory of the universe using relativity and quantum mechanics suggest that time as we seem to experience it does not exist - it may be only a well-founded illusion. The idea of a timeless universe can be traced back to Plato and his insistence that only being is real, while becoming is an illusion. In this talk, Prof. Barbour will explore how the Wheeler-DeWitt equation of quantum gravity suggests the fundamentally timeless nature of the quantum universe. He will also raise unresolved mysteries of our conscious experiences, and why these might provide insight into how a fundamentally timeless universe may be perceived as intensely temporal. A key result of his proposal could be an explanation of the asymmetry between the past and the future."

Actually, I myself have had the thought for a very long time that the now' is ALL there is to reality and the past and the becoming are illusions - a thought I have found depressing, for example what if the moment' is when you have just got knocked out of the world cup on penalties? When I found out about the timeless nature of GR this thought reoccurred to me. It would seem, maybe I'm reading too much into the Plato reference, that this is what Julian Barbour is talking about, but in a much deeper way than me.

I prefer Rovelli's explanation of evolution from a timeless universe which has to do with how we have limited information about the world - less depressing perhaps as it leaves room for change? Like England winning the world cup.

Last edited by a moderator:

Gold Member
"I myself have had the thought for a very long time that..." - I realise that this statement contradicts the premise of the argument that only the now is real...

Gold Member
Dearly Missed
"I myself have had the thought for a very long time that..." - I realise that this statement contradicts the premise of the argument that only the now is real...

Just 5 mintues ago, on reading this for the first time, I thought I heard you chuckling quietly to yourself, but I realize this statement is contradicted by the premise that only the now is real.

I prefer Rovelli's explanation of evolution from a timeless universe which has to do with how we have limited information about the world ...

If you have a particularly clear passage by him where he explains that idea, I'd be glad for a pointer to it. Are you perhaps thinking of this recent paper?
4059419]http://arxiv.org/abs/1209.0065
General relativistic statistical mechanics
Carlo Rovelli
(Submitted on 1 Sep 2012)
Understanding thermodynamics and statistical mechanics in the full general relativistic context is an open problem. I give tentative definitions of equilibrium state, mean values, mean geometry, entropy and temperature, which reduce to the conventional ones in the non-relativistic limit, but remain valid for a general covariant theory. The formalism extends to quantum theory. The construction builds on the idea of thermal time, on a notion of locality for this time, and on the distinction between global and local temperature. The last is the temperature measured by a local thermometer, and is given by kT = [STRIKE]h[/STRIKE] dτ/ds, with k the Boltzmann constant, [STRIKE]h[/STRIKE] the Planck constant, ds proper time and dτ the equilibrium thermal time.
9 pages. A tentative second step in the thermal time direction, 10 years after the paper with Connes. The aim is the full thermodynamics of gravity. The language of the paper is a bit technical: look at the Appendix first

Last edited:
Gold Member
- less depressing perhaps as it leaves room for change? Like England winning the world cup.

Unlikely... :-D

Naty1
Marcus: Here is some perspective from Rovelli and I suspect you have seen it since I think I got it in these forums [I have no source] ; just in case you missed it:

1.1 Space
Many simple arguments indicate that lP may play the role of a minimal length, in the same sense in which c is the maximal velocity and ¯h the minimal exchanged action.....

It may have a quantum granularity at the Planck scale, analogous to the granularity of the energy in a quantum oscillator. This granularity of space is fully realized in certain quantum gravity theories, such as loop quantum gravity, and there are hints of it also in string theory. Since this is a quantum granularity, it escapes the traditional objections to the atomic nature of space.

1.2 Time

Time is affected even more radically by the quantization of gravity. In conventional QM, time is treated as an external parameter and transition probabilities change in time. In GR there is no external time parameter. Coordinate time is a gauge variable which is not observable, and the physical variable measured by a clock is a nontrivial function of the gravitational field. Fundamental equations of quantum gravity might therefore not be written as evolution equations in an observable time variable.

And in fact, in the quantum–gravity equation par excellence, the Wheeler-deWitt equation, there is no time variable t at all. Much has been written on the fact that the equations of nonperturbative quantum gravity do not contain the time variable t. This presentation of the “problem of time in quantum gravity”, however, is a bit misleading, since it mixes a problem of classical GR with a specific quantum gravity issue.

Indeed, classical GR as well can be entirely formulated in the Hamilton-Jacobi formalism, where no time variable appears either. In classical GR, indeed, the notion of time differs strongly from the one used in the special relativistic context. Before special relativity, one assumed that there is a universal physical variable t, measured by clocks, such that all physical phenomena can be described in terms of evolution equations
in the independent variable t. In special relativity, this notion of time is weakened. Clocks do not measure a universal time variable, but only the proper time elapsed along inertial trajectories. If we fix a Lorentz frame, nevertheless, we can still describe all physical phenomena in terms of evolution equations in the independent variable x0, even though this description hides the covariance of the system.

In general relativity, when we describe the dynamics of the gravitational field (not to be confused with the dynamics of matter in a given gravitational field), there is no external time variable that can play the role of observable independent evolution variable. The field equations are written in terms of an evolution parameter, which is the time coordinate x0, but this coordinate, does not correspond to anything directly observable. The proper time _ along spacetime trajectories cannot be used as an independent variable either, as _ is a complicated non-local function of the gravitational field itself.

Therefore, properly speaking, GR does not admit a description as a system evolving in terms of an observable time variable. This does not mean that GR lacks predictivity. Simply put, what GR predicts are relations between (partial) observables, which in general cannot be represented as the evolution of dependent variables on a preferred independent time variable.
This weakening of the notion of time in classical GR is rarely emphasized: After all, in classical GR we may disregard the full dynamical structure of the theory and consider only individual solutions of its equations of motion. A single solution of the GR equations of motion determines “a spacetime”,
where a notion of proper time is associated to each timelike worldline.
But in the quantum context a single solution of the dynamical equation is like a single “trajectory” of a quantum particle: in quantum theory there are no physical individual trajectories: there are only
transition probabilities between observable eigenvalues. Therefore in quantum gravity it is likely to be impossible to describe the world in terms of a spacetime, in the same sense in which the motion of a quantum electron cannot be described in terms of a single trajectory.
To make sense of the world at the Planck scale, and to find a consistent conceptual framework for GR and QM, we might have to give up the notion of time altogether, and learn ways to describe the world in atemporal terms. Time might be a useful concept only within an approximate description of the physical reality.

1.3 Conceptual issues
The key difficulty of quantum gravity may therefore be to find a way to understand the physical world in the absence of the familiar stage of space and time. What might be needed is to free ourselves from the prejudices associated with the habit of thinking of the world as “inhabiting space” and “evolving in time”.
Technically, this means that the quantum states of the gravitational field cannot be interpreted like the n-particle states of conventional QFT as living on a given spacetime. Rather, these quantum states must themselves determine and define a spacetime —in the manner in which the classical solutions of GR do.
Conceptually, the key question is whether or not it is logically possible to understand the world in the absence of fundamental notions of time and time evolution, and whether or not this is consistent with our experience of the world.....

Rovelli: Unfinished revolution
Introductive chapter of a book on Quantum Gravity, edited by Daniele Oriti,
to appear with Cambridge University Press
Carlo Rovelli
Centre de Physique Th´eorique de Luminy_, case 907, F-13288 Marseille, EU
February 3, 2008

Gold Member
Dearly Missed
It is an excellent overview essay on the whole QG effort. History, motivation, contextual setting in the rest of physics, conceptual requirements, goals. It's worth re-reading despite having been written 6 years ago back in 2006. Thanks for recalling and quoting!

kindlin
Ever since I watched Imagining The Tenth Dimension, I've basically assumed that what we view as time is completely wrong. Now, what 'the proper way to think of time' is I have only half guesses, but the idea that time is always and permeates everything and every possible conscious being feels time the same, I just don't believe.

The basic idea behind my lack of trust in our perception is based on how ITTD explains the formulaic way that the dimensions build on top of one another. My main example is thus: a line(1D) is a cross section of a square(2D) which is then just a cross section of a cube (3D) which is then (and here's the key) just one cross section of the time (4D). So for us third dimensioners, we view our 3D world changing slowly through the fourth dimension.

So, what if our universe only blew up with two physical dimensions and the 3rd to the 10th were all rolled up? I would assume that the second dimensioners would experience their version of time by going through cross sections of a greater third dimensional whole. What would the 3rd dimension we see every day look like for the universe that these second dimensioners live in? I theorize it would be a kind of movie but each frame is placed on top of the other, in the same way that 1000 sheets of paper (the square universe) can be stacked to form a cube.

What about the other possibility? What if the universe expanded with 4 physical dimensions? Would it be just like our world in 3D, but simply 4D? (whatever that is) and then their 5th dimension is their version of time? Or, would they see our 3rd dimensional world extending in two directions (along the 4th dimensional 'timeline' which would be figuratively and quite literally a line designating time for the 3rd D'ers). From this point it would make sense to think that they could alter the 4th D just like we can move a couch across a room (which would make as little since to us as trying to explaining that same situation to a 2nd D'er). Along this same vein the 4th D'ers would experience time through the 5th dimension, and the current viewable timeline (their world) would be one slice of the next higher dimension. So that means that the entire timeline could be changing over time. So then what would the people experiencing that timeline view?

TL:DR
Trying to combine higher dimensions (what some ToE's currently predict) with our experience of time leaves a lot of unanswered questions. It could either make perfect sense, or no sense at all, but something is up with our ideas about how we view time.

Gold Member
I can buy the idea that time is not fundamental, rather, it is an emergent property of the universe. It makes no sense, however, to question the obvious reality of time in the current universe. If it is an illusion, it is so extraordinarily clever it raises even more troubling questions than the ones it would resolve.

Gold Member
Dearly Missed
I think that Julian Barbour would agree with the idea that time is a REAL emergent feature of our experience. "Time does not exist" is an attention-getting headline and simply means what you suggest, namely that it is not FUNDAMENTAL. Let's bear in mind that Julian Barbour was giving a public lecture as part of outreach program. Some suggestive hyperventilating may have been called for.

Things can be real but not fundamental--the example often given is the temperature of a system---the individual molecules do not have temperature so it is not fundamental at the microscope level of physical reality. But temperature emerges importantly at a collective level.

Rovelli's recent paper carries this a step further---I guess you could consider him a follower of Barbour in this deep investigation of time (as opposed to naively geometrizing it as just another spatial dimension.)

I'd be fascinated to know what you and (other) Julian make of the paper.
======================

I think (if I can crudely oversimplify) it goes like this. Obviously every observer has his own PROPER (meaning his own personal) time. That's familiar to everybody from (general) Relativity. Every observer going thru life has a clock.
But of course the different proper times of different observers differ. Denote the local observer time by letter s.

OK. Now rovelli adds another factor in. He says that associated to THE STATE OF THE UNIVERSE there is also another time. Unless I'm mistaken, any physically realizable state defines a flow on the space of states according to which it is at equilibrium. So the state of the universe, whatever it is, defines its OWN time. Call it tau.

So then there is a local temperature which includes the idea of the temperature of the GEOMETRY. d tau/ds.
We have to have a temperature of the geometry if we are going to do thermodynamics of the geometry of the world. And there do seem to be hints from people like Tolman and Jacobson that geometry can have a temperature and a thermodynamics.

So here is this giddy possibly profound possibly wrong and frankly risky/unintuitive thing a local geometry temperature dτ/ds.

My hunch is that IF this leads to an interesting thermodynamics of geometry THEN it will turn out that tau is a good kind of emergent time. that would be a fine thing. So I'm curious what other people think when they've taken a look at the 0065 paper: 1209.0065 that I linked to in post #3.

Last edited:
kindlin
I was liking your post until I got to "naively geometrizing [time] as just another spatial dimension."

Care to explain why this view (detailed conceptually in my previous post) is not a good way to think of things? It is not a theory in itself, it is a prediction when you have multiple extra dimensions.

Gold Member
Dearly Missed
The Foundational Questions Institute abbreviated FQXI had an essay contest 3 or 4 years ago about the nature of time. Famous people George Ellis, Julian Barbour, Carlo Rovelli, Claus Kiefer,... all submitted these very thoughtful often rather original essays. They had a panel of physicists judging and they awarded prizes etc etc. They pretty much all had reasons why "geometrizing" time is a bad idea. the "block universe" is physically and logically invalid.
George Ellis had an elementary thought experiment to demonstrate the latter. Involving radioactive decay and a trolleycar zooming back and forth on a length of track. Sort of vaguely like Schroedinger Cat , but not.

You might find it fun to read some of the prizewinning essays. Maybe others too.

http://arxiv.org/abs/gr-qc/0604045 (Unfinished Revolution)
and http://arxiv.org/abs/0903.3832 (Forget Time)
They are both short fairly easy reading for the most part. You can skip any hard parts and still get the idea.

George Ellis was Stephen Hawking's co-author of the classic book The Large Scale Structure of Space-Time back when Hawking was doing majorly important science. Ellis is what you'd call an expert on fundamental questions about time and space and he nixes the block spacetime and drives the point home with his trolleycar. As I recall that's a fun one too, at least the first few pages. I don't have the link though.

Last edited:
kindlin
I wish I understood half of what is written in these articles. I'm not even sure what I don't understand, just that after reading a paragraph, I'm like, "what?"

I've tried looking up each individual things I don't understand, but they each reference half the other things that I don't understand. Besides that, these are all advanced topics so half the time the things I read are approximations of what is actually being discussed, and I don't have enough mathematical background to even teach myself these higher levels of math needed to actually understand this stuff (Hamiltonian, wave function, Lorentz invariant, etc etc etc).

I wish I had the time to take multiple high level college courses in this stuff, to at least give myself a basic understanding to build off of. I find these topics SOOOOO interesting, I'll read them for hours, and when I'm done reading, I'll have learned nothing. /sigh

audioloop
A thread was started on how unreal is space and time...I recently came across the of a talk by Julian Barbour -http://www.perimeterinstitute.ca/Outreach/Public_Lectures/Public_Lectures/ [Broken] (although Barbour has never held any academic position, he is accredited by the GR/LQG community for teaching them all that GR is a relational theory)

His talk was on

"Many attempts to create a unified theory of the universe using relativity and quantum mechanics suggest that time as we seem to experience it does not exist - it may be only a well-founded illusion. The idea of a timeless universe can be traced back to Plato and his insistence that only being is real, while becoming is an illusion. In this talk, Prof. Barbour will explore how the Wheeler-DeWitt equation of quantum gravity suggests the fundamentally timeless of the quantum universe. He will also raise unresolved mysteries of our conscious experiences, and why these might provide insight into how a fundamentally timeless universe may be perceived as intensely temporal. A result of his proposal could be an of the asymmetry between the past and the future."

Actually, I myself have had the thought for a very long time that the now' is ALL there is to reality and the past and the becoming are illusions - a thought I have found depressing, for example what if the moment' is when you have just got knocked out of the on penalties? When I found out about the timeless nature of GR this thought reoccurred to me. It would seem, maybe I'm reading too much into the Plato reference, that this is what Julian Barbour is talking about, but in a much deeper way than me.

I prefer Rovelli's explanation of from a timeless universe which has with how we have limited information about the world - less depressing perhaps as it leaves room for change? Like the world cup.

is the sort of the strange things, i.e. similar propositions that can solve in principle fundamental problems in physics

http://fqxi.org/data/forum-attachments/DFTTrieste_talk.pdf
http://arxiv.org/pdf/0912.2845v3.pdf
...The concept of time evolution is of course central to any dynamical theory, and in particular to quantum mechanics. In standard quantum mechanics time, and spacetime, are taken as given. But the presence of time in the theory is an indicator of a fundamental incompleteness in our understanding, as we now elaborate. Time cannot be deﬁned without an external gravitational ﬁeld [this could be ﬂat Minkowski spacetime, or a curved spacetime]. The gravitational ﬁeld is of course classical. Thus the picture is that an external spacetime manifold and an overlying gravitational ﬁeld must be given, before one can deﬁne time evolution in quantum theory...

...There are four reasons why our present knowledge and understanding of quantum mechanics could be regarded as incomplete. Firstly, the principle of linear superposition has not been experimentally tested for position eigenstates of objects having more than about a thousand atoms. Secondly, there is no universally agreed upon explanation for the process of quantum measurement...

and from the Two-State-Vector-Formalism
http://arxiv.org/ftp/arxiv/papers/1207/1207.0667.pdf
Coexistence of Past and Future Measurements’ Effects,
Predicted by the Two-State-Vector-Formalism and Revealed
by Weak Measurement

block models, cramers and so on....

Last edited by a moderator:
Gold Member
Time does not exist, provided that
1. All time evolution in the Universe is governed by the Schrodinger equation only.
2. The state of the Universe is a state with definite total energy (e.g., zero in canonical quantum gravity).

However, 1. is correct only in the many-world interpretation of quantum mechanics (QM). All other interpretations introduce some additional time dependence. The Bohmian formulation of QM provides a particularly natural origin of time:
http://arxiv.org/abs/1209.5196

Gold Member
If you have a particularly clear passage by him where he explains that idea, I'd be glad for a pointer to it. Are you perhaps thinking of this recent paper?
4059419]http://arxiv.org/abs/1209.0065
General relativistic statistical mechanics
Carlo Rovelli
(Submitted on 1 Sep 2012)

It is in Rovelli's paper "Forget time" http://arxiv.org/pdf/0903.3832.pdf he talks about it:

"The time of our experience is associated with a number of peculiar features that make it a very special physical variable. Intuitively (and imprecisely) speaking, time “flows”, we can never “go back in time”, we remember the past but not the future, and so on. Where do all these very peculiar features of the time variable come from?

I think that these features are not mechanical. Rather they emerge at the thermodynamical level. More precisely, these are all features that emerge when we give an approximate statistical description of a system with a large number of degrees of freedom. We represent our incomplete knowledge and assumptions in terms of a statistical state..."

Yes this is related to his new paper. After posting the Barbour's intro I came across Rovelli's new paper. Exciting to see if there is progress as I remember the paper he wrote with Connes http://arxiv.org/pdf/gr-qc/9406019.pdf and finding it very interesting but that was a while ago. I'm having a look at them both now. Thanks for outlining the what he says in new paper Marcus.

HomogenousCow
Too much metaphysics for my taste

Gold Member
Dearly Missed
Too much metaphysics for my taste

Dear Cow,
Statistical mechanics is not metaphysics and General Relativity is straightforward hard physics as well, though possibly not to your taste
It is an urgent unsolved problem how to do Thermodynamics in a GR context.
So take another look at the September 2012 paper (0065) on this topic before you make dismissive noises.

==1209.0065 page 1 excerpt==
Thermodynamics and statistical mechanics are powerful and vastly general tools. But their usual formulation works only in the non-general-relativistic limit. Can they be extended to fully general relativistic systems?

The problem can be posed in physical terms: we do not know the position of each molecule of a gas, or the value of the electromagnetic field at each point in a hot cavity, as these fluctuate thermally, but we can give a statistical description of their properties. For the same reason, we do not know the exact value of the gravitational field, which is to say the exact form of the spacetime geometry around us, since nothing forbids it from fluctuating like any other field to which it is coupled. Is there a theoretical tool for describing these fluctuations?

The problem should not be confused with thermodynamics and statistical mechanics on curved spacetime. The difference is the same as the distinction between the dynamics of matter on a given curved geometry versus the dynamics of geometry itself, or the dynamics of charged particles versus dynamics of the electromagnetic field. Thermodynamics on curved spacetime is well understood (see the classic [1]) and statistical mechanics on curved spacetimes is an interesting domain (for a recent intriguing perspective see [2]). The problem is also distinct from “stochastic gravity” [3, 4], where metric fluctuations are generated by a Einstein-Langevin equation and related to semiclassical effects of quantum theory. Here, instead, the problem is the just the thermal behavior of conventional gravity.1
A number of puzzling relations between gravity and thermodynamics (or gravity, thermodynamics and quantum theory) have been extensively discussed in the literature [5–14]. Among the most intriguing are probably Jacobson’s celebrated derivation of the Einstein equations from the entropy-area relation [15, 16], and Penrose Weil-curvature hypothesis [17, 18]. These are very suggestive, but perhaps their significance cannot be evaluated until we better understand standard general covariant thermodynamics.
==endquote==

Last edited:
Gold Member
Dearly Missed
Since we just turned the page, I'll quote a key comment Julian made earlier and the abstract of the paper on this topic that I just referred to.
I prefer Rovelli's explanation of evolution from a timeless universe which has to do with how we have limited information about the world...

... Are you perhaps thinking of this recent paper?
http://arxiv.org/abs/1209.0065
General relativistic statistical mechanics
Carlo Rovelli
(Submitted on 1 Sep 2012)
Understanding thermodynamics and statistical mechanics in the full general relativistic context is an open problem. I give tentative definitions of equilibrium state, mean values, mean geometry, entropy and temperature, which reduce to the conventional ones in the non-relativistic limit, but remain valid for a general covariant theory. The formalism extends to quantum theory. The construction builds on the idea of thermal time, on a notion of locality for this time, and on the distinction between global and local temperature. The last is the temperature measured by a local thermometer, and is given by kT = [STRIKE]h[/STRIKE] dτ/ds, with k the Boltzmann constant, [STRIKE]h[/STRIKE] the Planck constant, ds proper time and dτ the equilibrium thermal time.
9 pages. A tentative second step in the thermal time direction, 10 years after the paper with Connes. The aim is the full thermodynamics of gravity. The language of the paper is a bit technical: look at the Appendix first

BTW there is a frank and insightful 8-page overview/outlook article on Superstring by Gerard 't Hooft that just came out:

Last edited:
HomogenousCow
I apologize, I was not referring to the paper mentioned earlier.
My problem is instead with those "time does not exist" theories, I believe that physics is based on both theoretical and empirical research, without time, how does one perform an experiment to verify the theory?

Gold Member
Dearly Missed
I apologize, I was not referring to the paper mentioned earlier.
My problem is instead with those "time does not exist" theories, I believe that physics is based on both theoretical and empirical research, without time, how does one perform an experiment to verify the theory?

I see. Well I imagine that "time does not exist" is a attention-getting headline for a popular lecture or wide-audience essay. It does not accurately represent Barbour's own view. His prizewinning FQXi essay is a brilliant exposition of how time arises from observing the motions of a many-body system.

To respond directly to your question "...without time, how does one perform an experiment ...?" I would say that of course we always have time---in reality as modeled by GR we have the observer's own PROPER time.

However an observer's own (proper=individual personal) time is not universally applicable---it does not coordinatize a 4D block universe. In GR the fourth coordinate is unphysical: it is a mathematical convenience but not observable/measureable---does not correspond to any existing clock. One has different observers and their different experiences of time. So there remain some unresolved problems concerned with thermodynamics and the like.

In cosmology, happily enough, folks make the unrealistic but extremely useful assumption of uniformly distributed matter and homogeneous isotropic geometry (which anyone can see is not true, after all we have black holes, neutron stars, GPS satellites, Domino's pizzas etc all inhomogeneous as the dickens). But in cosmology we therefore acquire this beautiful universal time roughly linked to the nearly isotropic Cosmic Microwave Background. I suspect you know all about that.

But that is not quite good enough. So we get papers like Rovelli's September 2012 exploring how to do General Covariant thermodynamics and stat mech. How to treat time in that context. It seems to me it's something that makes sense for people to work on. They'll get it (I think.)

Don't want to sound like I think I'm an expert in these matters, which I'm not! I hope if I'm mistaken about any of this someone more knowledgeable will correct me on it. But anyway this is how I see it (as interested member of the audience.)

Last edited:
HomogenousCow
Hmmm but why is thermodynamics carried over to general relativity?
Shouldn't the two be naturally incompatible? I mean how do you provide things like a velocity distribution when the bare notion of an average is undefined?
(What I mean by this, is that we can't even compare four velocities that are not at the same point, how do we talk about a distribution?)

Gold Member
Dearly Missed
Hmmm but why is thermodynamics carried over to general relativity?
Shouldn't the two be naturally incompatible? I mean how do you provide things like a velocity distribution when the bare notion of an average is undefined?
(What I mean by this, is that we can't even compare four velocities that are not at the same point, how do we talk about a distribution?)

Smart questions! You should probably look at the article and some of the references to pre-existing work, on page 1.

First, stat mech has been developed on curved spacetimes earlier by other people and their papers, if you choose to look them up, should answer your questions somewhat there.

Second, what this paper is talking about is not the velocities of particles moving IN spacetime, it is talking about the fluctuations of geometry ITSELF.

Tolman already some 80 years ago found that geometry had temperature*, then later of course Hawking and Unruh found temperature associated with geometry. Jacobson derived the equation of GR from some thermodynamic assumptions about entropy. There are various hints that geometry itself (not the particles zooming around inside it) has thermodynamics.
So geometry is DYNAMIC of itself and one can ask what degrees of freedom are fluctuating? what are the "molecules" in this case? Areas? Angles? Volumes? Bits of shape? What's shaking? And of course geometry interacts with matter. So whatever is shaking is somehow able to feel matter moving around too. So there is an interesting statistical mechanics to be discovered. Ted Jacobson is someone to watch on this. Also Than Padmanabhan, also Rovelli. Erik Verlinde did some work 2 or 3 years back that attracted a lot of attention but the noise about that has quieted down. Geometry had entropy so it could exert an entropic force etc. He got a huge grant and gave a lot of talks. But that was just one try, one attempt. It is an ongoing story involving a number of different research approaches.

*[1] R. C. Tolman, Relativity, Thermodynamics and Cosmology. Oxford University Press, London, 1934.

Last edited:
ImaLooser
It seems to me that GR showed that the traditional concept of time was not correct. But all of physics was built on it, and anything better has yet to appear.

The cleverest idea I've seen was Feynman-Kac, where particles moved with a fractal path and time was defined as the standard deviation of the fractal. Or something like that. The particle travels an infinite distance. Unfortunately the original article is stashed behind a wall of money and I can't find the version that I read. The idea was never picked up and taken anywhere.

Gold Member
Dearly Missed
Looser, it's good to be reminded about trajectories---that we don't expect a particle to necessarily have a smooth trajectory. Maybe we only know its position at a finite number of points and not what it did in between. Maybe smooth trajectories do not exist at small scale, and only approximately do at large.

A classical spacetime is in a certain way like a smooth "trajectory" thru the realm of possible spatial geometries. So by the same token maybe spacetimes do not exist. We can only make a finite number of geometrical measurements, perhaps, that establish something about the course of evolution taken by geometry.

Like establishing that the particle went thru this slit and this and this, but we do not know what it did inbetweentimes.

I suppose it is possible for geometry to be "jittery" at small scale, and for its evolution to be poorly defined---perhaps comprised of a multitude of degrees of freedom, all able to interact with whatever matter is present.

So that perhaps the Einstein Field Equation is analogous to the Gas Law---PV = nkT. In a vague sense.

I don't have any ideas to offer myself, but I want to argue that people who are trying to understand the Thermodynamics of Geometry (and related things, time, entropy, temperature of general covariant systems, how to model the micro geometric d.o.f. ) that they are doing real physics. There are real unsolved physical problems. It is not just metaphysics or philosophy. Not always anyway. And it is a bunch of people. Ted Jacobson and Carlo Rovelli are just two out of 8 or 10 that come to mind.

Chronos (longtime PF contributor) just mentioned a recent paper by Padmanabhan, I think favorably, which may be related. Chronos may even have started a thread about the paper, I don't recall. I'll take a look and see if it fits into this pattern:
http://arxiv.org/abs/1207.0505
Emergent perspective of Gravity and Dark Energy
(Submitted on 2 Jul 2012)
There is sufficient amount of internal evidence in the nature of gravitational theories to indicate that gravity is an emergent phenomenon like, e.g, elasticity. Such an emergent nature is most apparent in the structure of gravitational dynamics. It is, however, possible to go beyond the field equations and study the space itself as emergent in a well-defined manner in (and possibly only in) the context of cosmology. In the first part of this review, I describe various pieces of evidence which show that gravitational field equations are emergent. In the second part, I describe a novel way of studying cosmology in which I interpret the expansion of the universe as equivalent to the emergence of space itself. In such an approach, the dynamics evolves towards a state of holographic equipartition, characterized by the equality of number of bulk and surface degrees of freedom in a region bounded by the Hubble radius. This principle correctly reproduces the standard evolution of a Friedmann universe. Further, (a) it demands the existence of an early inflationary phase as well as late time acceleration for its successful implementation and (b) allows us to link the value of late time cosmological constant to the e-folding factor during inflation.
38 pages; 5 figures

The first 16 pages are full of the thermodynamics of geometry. Also the conclusions section. I can't say much about the second half of the paper--which seems more than usually speculative, it may make you want to scream or just stop reading.

Last edited:
HomogenousCow
I really do not know much about the current state of theoretical physics, but are there any large gaping holes that have been found empirically? As in the procession of mercury for our time.

Gold Member
To respond directly to your question "...without time, how does one perform an experiment ...?" I would say that of course we always have time---in reality as modeled by GR we have the observer's own PROPER time.

However an observer's own (proper=individual personal) time is not universally applicable---it does not coordinatize a 4D block universe. In GR the fourth coordinate is unphysical: it is a mathematical convenience but not observable/measureable---does not correspond to any existing clock.

What about the early quantum dominated epoch of the universe; there were no classical degrees of freedom that could have played the role of a classical clock measuring some proper time.

Maybe there is no time at the fundamental level? But maybe this OK because none of us were around just after the big bang?

HomogenousCow
I hope to see these issues resolved in my life time, if not, then shame.
I suspect that after the four interactions are reconciled, some new problem will arise.

Gold Member
In the paper http://arxiv.org/pdf/gr-qc/0304074.pdf Ashtekar, Bojowald, Lewandowski say something similar

"For instance, the question of whether the universe had a beginning at a finite time is now ‘transcended’. At first, the answer seems to be ‘no’ in the sense that the quantum evolution does not stop at the big bang. However, since space-time geometry ‘dissolves’
near the big-bang, there is no longer a notion of time, or of ‘before’ or ‘after’ in the familiar sense. Therefore, strictly, the question is no longer meaningful. The paradigm has changed and meaningful questions must now be phrased differently, without using notions tied to classical space-times."

Gold Member
Dearly Missed
Hi Julian, you cite a 2003 paper by ABL. At that time B was the main guiding light, the main person developing LQC and the other two were helping him get the math right. Around 2006 2007 there was a change. Improved dynamics, basically LQC was reformulated. A became a central figure. The most active people now, besides Ashtekar, are younger folk who have been collaborating primarily with Ashtekar recently ( Sloan, Henderson, Nelson, Agullo, Singh, Corichi, Wilson-Ewing,...)
The impression I get from recent papers Ashtekar et al post 2010 is quite different from the somewhat mysterious quote from the 2003. More straightforward evolution thru the bounce. Hundreds of numerical simulations. Conditions specified on the space-like hypersurface of the universe at the moment of bounce---treated as physically meaningful.

So I see only a possibility for semantic disagreement and or semantic confusion. There are various kinds of time used in various departments of science. The word does not have a universal unique meaning. Different models run on different kinds of time-variable.

Ashtekar's computer does not seem to be bothered by the fact that no classical observer and no clock could be imagined to exist at the moment of bounce--it just keeps computing and churns on thru from collapse to rebound to inflation and so on. Bounce-time is a well-defined time, at which conditions and parameter values can be and ARE specified.

So some appropriate concept of "time" needs to be worked out, I suppose. An interesting open problem.

My guess is that Rovelli's THERMAL TIME tau would not extend back thru the bounce into the contracting phase. but that is just an uneducated guess. I haven't thought about it.
It seems to me that if it did then his temperature number would blow up.
proper time s is not defined since no observer can exist continuously thru the bounce (I assume). So d tau/ds would fail to be defined.
Someone could even ask Rovelli about that. Or one might ask Hossenfelder, who speculates you could have hbar -> 0 at extreme energy density or temperature.
Rovelli's temperature of geometry is T = (hbar/k)dτ/ds.
So if hbar -> 0 (ala Hossenfelder) that might control the blowup of dτ/ds and Rovelli's temperature T might actually have a definite finite value at bounce-time, which would be curious and delightful, but quite unexpected.

So I think that pragmatically just looking at the behavior of the experts and at the current research there IS a notion of time near the start of expansion, but it is an interesting open problem how to invent a precise definition corresponding to that notion. I'm pretty sure one will be invented. It may already be lurking in the literature somewhere unbeknownst to me

Last edited:
==1209.0065 page 1 excerpt==
Thermodynamics and statistical mechanics are powerful and vastly general tools. But their usual formulation works only in the non-general-relativistic limit. Can they be extended to fully general relativistic systems?
The problem can be posed in physical terms: we do not know the position of each molecule of a gas, or the value of the electromagnetic field at each point in a hot cavity, as these fluctuate thermally, but we can give a statistical description of their properties. For the same reason, we do not know the exact value of the gravitational field, which is to say the exact form of the spacetime geometry around us, since nothing forbids it from fluctuating like any other field to which it is coupled. Is there a theoretical tool for describing these fluctuations?
==endquote==

What can it mean to say "spacetime fluctuates" unless one introduces an additional temporal dimension?

Gold Member
Dearly Missed
Hi Ruta, thanks for joining the conversation. Nice to hear from you! You quoted Rovelli's September 2012 paper, in which he defines and uses several types of TIME.
What can it mean to say "spacetime fluctuates"...?
Well you are quoting Rovelli so your question is what does HE mean, so you could write and ask him. But I will venture to suggest that what he means is the spacetime geometry cbanges with the passage of time.

I had better copy the abstract and the passage to give context to what you were quoting. Certainly one does not have to have a 4D block universe with a physically meaningful time coordinate being one of the dimensions merely in order to model change with the passage of time. You in particular would be expected to know this better than many others, including myself. But here in this paper we have no lack of times: proper time, and thermal time, and a local version involving a local hamiltonian.

You have many choices for what fluctuation can mean, of course, since you have several opportunities to describe things as changing with the passage of time.

I think in that page 1 paragraph it is meant in a general sense without specifying the particular time-evolution. But anyway I will copy the material to get it all together where we can look at it:
http://arxiv.org/abs/1209.0065
General relativistic statistical mechanics
Carlo Rovelli
(Submitted on 1 Sep 2012)
Understanding thermodynamics and statistical mechanics in the full general relativistic context is an open problem. I give tentative definitions of equilibrium state, mean values, mean geometry, entropy and temperature, which reduce to the conventional ones in the non-relativistic limit, but remain valid for a general covariant theory. The formalism extends to quantum theory. The construction builds on the idea of thermal time, on a notion of locality for this time, and on the distinction between global and local temperature. The last is the temperature measured by a local thermometer, and is given by kT = [STRIKE]h[/STRIKE] dτ/ds, with k the Boltzmann constant, [STRIKE]h[/STRIKE] the Planck constant, ds proper time and dτ the equilibrium thermal time.
9 pages. A tentative second step in the thermal time direction, 10 years after the paper with Connes. The aim is the full thermodynamics of gravity. The language of the paper is a bit technical: look at the Appendix first.

==1209.0065 page 1 excerpt==
Thermodynamics and statistical mechanics are powerful and vastly general tools. But their usual formulation works only in the non-general-relativistic limit. Can they be extended to fully general relativistic systems?

The problem can be posed in physical terms: we do not know the position of each molecule of a gas, or the value of the electromagnetic field at each point in a hot cavity, as these fluctuate thermally, but we can give a statistical description of their properties. For the same reason, we do not know the exact value of the gravitational field, which is to say the exact form of the spacetime geometry around us, since nothing forbids it from fluctuating like any other field to which it is coupled. Is there a theoretical tool for describing these fluctuations?

The problem should not be confused with thermodynamics and statistical mechanics on curved spacetime. The difference is the same as the distinction between the dynamics of matter on a given curved geometry versus the dynamics of geometry itself, or the dynamics of charged particles versus dynamics of the electromagnetic field. Thermodynamics on curved spacetime is well understood (see the classic [1]) and statistical mechanics on curved spacetimes is an interesting domain (for a recent intriguing perspective see [2]). The problem is also distinct from “stochastic gravity” [3, 4], where metric fluctuations are generated by a Einstein-Langevin equation and related to semiclassical effects of quantum theory. Here, instead, the problem is the just the thermal behavior of conventional gravity.1
A number of puzzling relations between gravity and thermodynamics (or gravity, thermodynamics and quantum theory) have been extensively discussed in the literature [5–14]. Among the most intriguing are probably Jacobson’s celebrated derivation of the Einstein equations from the entropy-area relation [15, 16], and Penrose Weil-curvature hypothesis [17, 18]. These are very suggestive, but perhaps their significance cannot be evaluated until we better understand standard general covariant thermodynamics.
==endquote==[/QUOTE]

Last edited:
Hi Ruta, thanks for joining the conversation. Nice to hear from you! You quoted Rovelli's September 2012 paper, in which he defines and uses several types of TIME.

Well you are quoting Rovelli so your question is what does HE mean, so you could write and ask him. But I will venture to suggest that what he means is the spacetime geometry changes with the passage of time.

Thanks for the reply, marcus. I don't see an answer in Rovelli's quote, so I'm hoping someone here can shed some light on the meaning of "spacetime fluctuations." I hear this phrase often so it must mean *something* to someone.

In your take, you say "spacetime geometry changes with the passage of time." Thus, we have different spacetime geometries which are ordered 'temporally' and you have introduced a 5th (and temporal) dimension. [Since spacetime contains all of Rovelli's definitions of time, the time of the temporal ordering is not one of them.] What does this mean to us empirically?

Last edited:
Gold Member
Dearly Missed
... I'm hoping someone here can shed some light on the meaning of "spacetime fluctuations." I hear this phrase often so it must mean *something* to someone.

I see! You hear the phrase often, from different people in different contexts. I don't suppose it always means the same, or is used the same way. So I don't see how I can set your mind at rest. One has to see how a given author translates the idea into mathematics.

Purely verbal description has a certain vagueness--the common-language meaning, i.e. the usage, "fluctuates" one might say So often, I imagine, one has read further into any given article to see mathematically what is intended.

Here is the passage which I think you are saying you do not understand. It is on page 1, early in the introduction of the article, where an author often speaks in general terms about what will be made mathematiclly precise later.

== http://arxiv.org/abs/1209.0065 page 1 excerpt==
Thermodynamics and statistical mechanics are powerful and vastly general tools. But their usual formulation works only in the non-general-relativistic limit. Can they be extended to fully general relativistic systems?

The problem can be posed in physical terms: we do not know the position of each molecule of a gas, or the value of the electromagnetic field at each point in a hot cavity, as these fluctuate thermally, but we can give a statistical description of their properties. For the same reason, we do not know the exact value of the gravitational field, which is to say the exact form of the spacetime geometry around us, since nothing forbids it from fluctuating like any other field to which it is coupled. Is there a theoretical tool for describing these fluctuations?
==endquote==

Now as it stands this bit of the introduction could be talking about different sorts of variation. Later on this particular article introduces some notions of temporal variation. Not of the "block universe" sort, or where there is a global foliation! But observer proper time is mentioned, and "many-fingered time" and local foliation (slicing) under specific conditions---so one can talk about temporal variation and make it precise in different ways, in the context of this paper.

But that is later on! What I think you are saying you do not understand (and want someone to give you a meaning) is this bit of the introduction. As it stands it could be talking about different sorts of variation, as I said. There could be variation in terms of SCALE, for instance. The field could appear to fluctuate randomly not with time but as you narrow down and zoom in. And that is not the only sort of non-temporal variation in geometry that the author might have had in mind! If it were Laurent Freidel, he might have been thinking of the variation of spacetime geometry depending on the observer's VANTAGE POINT. Freidel has argued that there does not exist a unique spacetime for all observers but according to "relative locality*" each has his own phase space and Freidel deduces empirically testable consequences from this. It is a bit exotic and I have forgotten the details but it came out last year and has not gone away. Anyway there can be different sorts of non-temporal variation of spacetime geometry---it can differ ala Freidel from observer to observer and according to the scale at which a given observer examines the spacetime geometry. As an expert you probably are aware of other types of non-temporal fluctuation that I can't think of at the moment.

But with Rovelli's introduction to his September 2012 paper, I think it is simpler than that. If you read on into the paper you will see, I imagine, that he is only talking about types of TEMPORAL variation, in a strictly 4D context (no 5th dimension ) of spacetime geometry, whether it be, say, as experienced by a single observer in one locale, or by a many-fingered multitude of observers, or defined according to this interesting "thermal time" concept. Temporal, in other words, but not assuming a fixed global foliation.

Please let me know if I am missing something in my reading of the paper.

* http://arxiv.org/abs/1106.0313/
Relative locality: A deepening of the relativity principle
Giovanni Amelino-Camelia, Laurent Freidel, Jerzy Kowalski-Glikman, Lee Smolin

Last edited: