Why do we need a unified theory?

[ChrisPeace]

Hello everyone!


I'm wondering why everyone feels its important that we have one unified theory of physics. Why must all things play by the same set of rules?

I play baseball, or I play basketball, they are sports, but their rules are unique and that makes them special right?

Why can't quantum mechanics and relativity just be separate and equally beautiful?

Does there HAVE TO BE a unified theory...and if so, explain why.


Chris

 

[Nick89]

Quantum mechanics and relativity 'play by the same rules' when they are describing the same phenomenon. If they would not play by the same rules, obviously one theory (at least) would be wrong!

If you throw a ball up in the air to test two different theories (A and B), and theory A says that the ball will come down, while theory B says the ball will stop in mid air and never return to earth. As soon as you throw the ball you will see that the ball falls back, thus theory B is absolutely wrong. Theory A is right as far as we can see, but it might not be right exactly.

That's also what happened with classical mechanics and special relativity for example: classical mechanics predict behavior correctly as long as the speeds involved are much smaller than the speed of light. Because moving things near the speed of light is very difficult (especially in the days that classical mechanics was developed), it was assumed that classical mechanics is exactly right. But then came along Einstein and said it was wrong! The fact is that classical mechanics is only very wrong when it comes to the speed of light: in low-speed phenomena the theories of classical mechanics seem to be exactly right, but as we know now, they are not. The 'error' however is so small that it can be regarded as correct for low speeds.

The fact remains that any new theory must at least be compatible with old theories that have been confirmed with experiments. Special relativity is compatible with classical mechanics because if you 'take the limit' for the speed to zero, the laws of special relativity approaches the same laws as classical mechanics.
In quantum mechanics, something similar exists that makes it compatible with classical mechanics (in the limit of large quantum numbers or something, don't know exactly...).So, quantum mechanics and relativity are not quite the same. They describe (mostly, as far as I know) different aspects of our universe, but the aspects that they share must at least agree in both theories, otherwise on would obviously be false.

I don't know if a unified theory HAS TO BE found, but it would be amazing if we could fit all the different theories (all describing different things) into one large theory that describes everything we know.

In the end, I think you must accept that all of our theories are just models of reality. We will probably never know what happens exactly because most models are only approximations or make unpractical assumptions that can never be 100% true in practice.

 

[ChrisPeace]

Its my understanding that relativity can only be related to larger systems, whereas quantum mechanics gets to the subatomic level.

From what I've watched and read, relativity cannot be applied to subatomic particles, and quantum mechanics cannot be applied to galaxies.

Its been my understanding that science sees this as an ugly phenomenon.

If the balls falls to the ground or floats it can be said that it will react different ways in different situations, which a theory cannot support.

 

[Naty1]

Quantum mechanics and general relativity both fail ( become indeterminate) , at some of the most interesting conditions imaginable: near the center of black holes and approaching the initial singularity of the universe: the big bang. Consequently we do not have a complete theory of either; each fails on its own. Every time such a "disconnect' has occurred in the past and been solved new insights and understanding have emerged...it's believed that will happen again when a theory is able to combine quantum and relativity theories.

In a perfect solution it may also happen that the fundamental constants of the standard model of particle physics also emerge. Currently those are experimentally determined (measured) parameters and we have no theory that correctly predicts them. It would be nice to know why key dimensionless quantities have the values we observe. In other words, for example, did such basic parameters of our universe emerge by chance or did the conditions at the big bang mandate precise values...or something in between??

 

[Dale]

I agree with Naty1, the two main motivations for a TOE are first to have a theory that covers the big bang or other situations with huge energies in small spaces, and second to hopefully reduce the number of fundamental parameters.

 

[Xezlec]

Its my understanding that relativity can only be related to larger systems, whereas quantum mechanics gets to the subatomic level.

From what I've watched and read, relativity cannot be applied to subatomic particles, and quantum mechanics cannot be applied to galaxies.

No, that is false. First of all, relativity is very important in subatomic particle physics. Effects such as time dilation and mass-energy equivalence are quite applicable there. Only certain aspects of relativity fall apart in quantum circumstances, namely things that have to do with gravity (in other words, general relativity). Second, saying "cannot be applied" isn't quite right. They can be applied. It's just that there isn't a known, correct way to combine the two in certain circumstances to produce meaningful results.

If you cannot combine the two, then whenever you want to ask a question that involves both the quantum and relativistic aspects of nature, the only answer will be "we don't know". Since physics is supposed to answer questions, the failure to be able to combine the two theories would be a failure of physics.

Relativity and quantum mechanics are not two different, complete descriptions of the laws of nature, where the universe just switches from one to the other when some kind of event happens. They describe two different aspects of the behavior of the universe. GR deals with the fabric of spacetime, what it looks like, and how gravity interacts with it. QM deals with particles, units of information, and how that information can change and move through the fabric described by GR.

The reason quantum mechanics is less relevant for large objects is that its effects become very weak for large things comprised of many particles. Weird fluctuations in particles and the fundamental units of information become "averaged out" because there are so many of them in the big object. The reason general relativity becomes less relevant for small objects is that general relativity deals with gravitation, and the masses of small objects are small, so their gravitational interactions are weak. Weak does not mean zero, it just means "hard to measure".

When looking at incredibly extreme situations, such as black holes and big bangs, both quantum mechanics and general relativity become significant. A black hole has a bunch of mass squeezed to a tiny enough particle (and its event horizon is a thin enough region) that its quantum and information theoretical characteristics matter, but it also has enough gravity that its general relativistic characteristics matter.

 

[Nabeshin]

I'm no expert in either QM or GR, but I have heard this interesting explanation before about the applicability of GR to small-scale phenomenon:

When describing quantum phenomenon the universe on such scales is often described as foamy, which I guess means that unpredictable energy fluctuations are constantly changing the topography of spacetime at these small scales (really, at all scales, but noticeably here). At high enough energies, it is even possible that spacetime could puncture and tear (any other creative words) at these small scalse. Well, since GR (so the argument goes) is founded on the assumption that the universe is smooth, and QM clearly shows it is not, then GR cannot be right. It is only when we try to apply GR to quantum phenomenon that the theory fails to work because of this.

This is the argument I have heard (read), in the way that I remember it. I worry that it might be clouded in popular science termonolgy and not rigorous enough, but it seems to make sense to me. Now, if this were correct, it seems that GR is, much like Newtonian mechanics, only an approximation (clearly, it cannot be correct if the assumption upon which it is based is incorrect). Therefore, it would seem, QM is on stronger footing for describing all phenomena (at least in that it's assumptions have not been proven false).

Thoughts? Seems to me to be an interesting way of looking at the conflict when GR enters the realm of QM.

 

[LURCH]

All these posts are correct, but I'd like to re-emphasize this:

Quantum mechanics and relativity 'play by the same rules' when they are describing the same phenomenon. If they would not play by the same rules, obviously one theory (at least) would be wrong!

If you throw a ball up in the air to test two different theories (A and B), and theory A says that the ball will come down, while theory B says the ball will stop in mid air and never return to earth. As soon as you throw the ball you will see that the ball falls back, thus theory B is absolutely wrong. Theory A is right as far as we can see, but it might not be right exactly.

Aristotle's Law of Non-contradiction states that, "two mutually exclusive statements cannot both be true." And that's what we have with GR and QM. There are certain situations in which GR predicts one thing will happen, and QM predicts something else. They might both be wrong, but they can't both be right.

The flip side of this law is that for any situation, there is some statement that will be true. This is the statement researchers are looking for when they search for a Unified Theory.

 

[Varnick]

To, possibly needlessly, re-use the sporting analogy:

The universe is our sports game, but team A are playing by the rules of general relativity and team B are playing by the Rules of Quantum Mechanics. Clearly, the rules for a game should be the same, so we should see a "Unified theory", one ruleset that is applicable to all players.

Less sportingly: If two theories describe the same thing, they shouldn't disagree.

 

[Andy Resnick]

I'm going to take the Devil's advocate side, and state that none of the responses have been particularly satisfying. For example:

1) We may want to have a TOE to satisfy human notions of elegance and completeness, but that is very different from *requiring* that a TOE *must* exist due to physical principles. The question is not "Why should we try to discover a TOE?", but rather "Why must a TOE exist? "

2) What garuntee do we have that a TOE, if written down, is useful? For example, if I wanted to make a map of a region, the only way that map can contain all the information of the original is if the map is a 1:1 scale mapping (unintentional, honest!) of the original, which makes for a very useless and unwieldy map. Thus, it could be stated that a complete TOE will be likely unwieldy and useless.

3) The current research focus of a TOE, to reconcile GR and QM in the high energy, small distance regimes, does nothing to explain numerous phenomena we already know about and cannot explain from elementary principles- friction, turbulence, Hofmeister series, colligative properties, constitutive relations, to name a few. Why try to understand black holes when we don't yet understand water? Claiming to understand 'the big bang' is specious- we cannot observe and perform controlled experiments, thus the result is little better than an educated guess.

For example, it was stated that "There are certain situations in which GR predicts one thing will happen, and QM predicts something else." What are these situations, what experiments have been performed, and what are the results? Lacking that, we are simply demanding that the universe conform to a particular human desire for understanding rather than trying to gain understanding of the actual universe.

 

[Bourbaki1123]

I wonder, since physics is principally a mathematical system, even though it is subject to physical phenomena, the unification is more logical/mathematical than physical so is it possible that this system is analogous to a logical system that is necessarily consistent but that we wish to make complete? If so, there will never be a unified theory. Certainly, as there cannot be a unified theory of mathematics there cannot be one of everything, so if the mind is physical, so too is mathematics... in a sense, so surely there could not be a unified theory of physics?

Also, since the mind is physical, and the laws which we use to describe and predict physical phenomena come from the mind, wouldn't they be produced by themselves? Then also wouldn't consciousness be a physical property? Unless of course you do not believe that the mind is necessarily physical.

 

[Xezlec]

OK, this is just getting silly.

1) We may want to have a TOE to satisfy human notions of elegance and completeness, but that is very different from *requiring* that a TOE *must* exist due to physical principles. The question is not "Why should we try to discover a TOE?", but rather "Why must a TOE exist? "

A theory of everything is just a complete description of as much of the universe as is visible to us. If you were to make a list of every detail of event humans will ever observe throughout all time and space, you would have a TOE, in a certain sense. In such a raw, uncompressed form, this list would be extremely unwieldy, but nonetheless useful. I'd love to know what next week's lottery numbers are.

Saying that such a raw set of facts does not exist just doesn't make sense to me. Are you denying that there is any reality?

2) What garuntee do we have that a TOE, if written down, is useful? For example, if I wanted to make a map of a region, the only way that map can contain all the information of the original is if the map is a 1:1 scale mapping (unintentional, honest!) of the original, which makes for a very useless and unwieldy map. Thus, it could be stated that a complete TOE will be likely unwieldy and useless.

Compression works when information contains at least some patterns somewhere. If you truly believe that the events in the universe cannot be reduced at all, and therefore that they contain absolutely no patterns whatsoever, and are therefore totally random, think about what that would mean. Everything as it is in exactly this instant in time has zero relationship to everything in the next instant. There are no places, no things, no stillness, no motion, no rules, just pure white randomness throughout the universe. A particle existing now has virtually no chance of existing in one nanosecond. There are no forces. No bodies which continue to hold the same shape across multiple instants in time. No processes. Certainly no stars and planets, let alone life forms, let alone language, let alone computers, let alone the patterns of letters and words displayed upon the patterns of display pixels you are looking at right now.

Of course the universe has patterns. And another name for a pattern is a rule, which may also be called a law. We already know there are laws. We're just working out some of the finer details. In fact, our existing set of rules is much smaller than the whole universe and already seems to correctly predict an awful lot of what's out there.

As to your example, you're not correct. Maps are not random. Look at a map, then generate a plot of random pixels. See the difference? You can certainly compress map data. There are lots of patterns. Things tend to be similar to things near themselves. Go make a very detailed map out of a bunch of pixels and then turn it into a GIF. Congratulations, you just made a perfectly accurate map of that original map, with fewer bytes than the original. And it's obviously useful!

3) The current research focus of a TOE, to reconcile GR and QM in the high energy, small distance regimes,

I think the current state of unification is to finish unifying the electromagnetic and weak forces, and then to try to pull in the strong force. I think pulling in gravity (GR), while certainly a goal, is much, much further off. Correct me if I'm wrong.

does nothing to explain numerous phenomena we already know about and cannot explain from elementary principles- friction, turbulence, Hofmeister series, colligative properties, constitutive relations, to name a few. Why try to understand black holes when we don't yet understand water?

I was under the impression that at least turbulence and friction were well known to arise from elementary principles, and that both appear spontaneously in simulations of more fundamental laws. If you're just saying that we haven't found easier mathematical tools to describe them than such brute force simulation, I fail to see your point. I don't see why the difficulty of dealing with some of the results of the fundamental laws means that we shouldn't finish figuring out what the rest of those fundamental laws are and finding out what else they have to teach us. If we had waited for better mathematical tools for modeling friction before trying to understand electromagnetics, we wouldn't have computers.

Claiming to understand 'the big bang' is specious- we cannot observe and perform controlled experiments, thus the result is little better than an educated guess.

Claiming to understand what my computer will do in 10 minutes is specious- I cannot observe and perform controlled experiments on the future until it gets here, thus the result is little better than an educated guess.

Oh, wait, no, that's silly. I know that most rules tend to hold true through some contiguous region of time and space, so it's a reasonable assumption that my computer will behave the same way in 10 minutes -- or 10 miles away -- that it does here and now. It is also reasonable to assume we know what Mars' surface looks like, even though man has never been there. Yes, we're assuming that the laws of physics upon which the rovers are based continue to operate in the Martian environment, even though it is very different from our environment. But it turns out that's not a terrible assumption, even if it isn't perfect.

GR and QM have successfully predicted things that happened very far away. No doubt one or both gradually changes as you get near crazy things like black holes, but just like we've done in the past, we can probably use those situations to fine tune the laws to make them even more general. And no doubt this will reveal all kinds of cool (and maybe even useful) stuff, just as it has in the past.

For example, it was stated that "There are certain situations in which GR predicts one thing will happen, and QM predicts something else." What are these situations, what experiments have been performed, and what are the results?

I mentioned one such situation: black holes. I believe there are examples in particle physics as well, beyond our current accuracy threshold. If experiments had been performed to resolve these problems, then they would have been resolved by now. The reason these things are unknown is that the necessary experiments are beyond our current ability to conduct.

Many things have been predicted by theorists before they were found by experiment. Einstein found a way to reconcile the observed constancy of the speed of light with observed everyday-scale mechanics. The result he came up with was relativity. Astronomers later confirmed most of this theory by experiment. So why would you be opposed to today's physicists attempting to reconcile the observed laws of quantum mechanics with the observed behavior of gravitation?

Lacking that, we are simply demanding that the universe conform to a particular human desire for understanding rather than trying to gain understanding of the actual universe.

What? I would think that "trying to gain understanding" is the result of "human desire for understanding". I don't see how those things are different. In any case, the best we can do is to come up with the simplest rule that explains everything we know at any given time, and make it more complex as more is discovered. And that's what everyone is trying to do.

 

[Andy Resnick]

A theory of everything is just a complete description of as much of the universe as is visible to us. <snip>
Saying that such a raw set of facts does not exist just doesn't make sense to me. Are you denying that there is any reality?

I never claimed facts do not exist. But a (useful) theory is much more than just a set of facts. A good physical theory both systematizes the facts into comprehendable order, suggests/predicts new facts that have not yet been determined, and, the essential part, reduces observed complexity into it's simplified and abstracted components. Thus Hooke's law (for example) is a highly useful theory of linear elastic materials, even though no real material precisely follows Hooke's law.

Compression works when information contains at least some patterns somewhere. If you truly believe that the events in the universe cannot be reduced at all, and therefore that they contain absolutely no patterns whatsoever, and are therefore totally random, think about what that would mean. <snip>

Again, I never claimed that. My claim is that for a theory to precisely and perfectly reproduce the entire universe would be akin to have a 1:1 scale map. Maps do not *precisely* reproduce the actual terrain. If you like, maps are a 'coarse grained' theory.

I think the current state of unification is to finish unifying the electromagnetic and weak forces, and then to try to pull in the strong force. I think pulling in gravity (GR), while certainly a goal, is much, much further off. Correct me if I'm wrong.

The electroweak theory is very mature, AFAIK. It is true that QCD and the electroweak theories have not yet been unified, but again, AFAIK, string theory, loop QG, etc. are all (putative) theories of quantum gravity

<snip>
Claiming to understand what my computer will do in 10 minutes is specious- I cannot observe and perform controlled experiments on the future until it gets here, thus the result is little better than an educated guess.

That's again, not what I was saying. As you correctly point out, our physical laws are likely to apply 10 minutes from now, so we feel comfortable making predictions. Also, our laws of physics are believed to have applied 10 minutes ago, so we feel comfortable making *postdictions*. However, our initial postulate is that our laws of physics *do not* hold near the initial singular event of the universe, so how can you confidently make *any* statement about the validity of a inherently untestable theory?

Many things have been predicted by theorists before they were found by experiment.

Actually, that's very backwards and not how the overwhelming amount of scientific progress has occurred. In general, first comes observation of a phenomenon, then comes experiment and manipulation of the phenomenon, *then* comes theory. Which is one reason why today's theorists are having so many problems making progress- there's nothing to compare results to.

What? I would think that "trying to gain understanding" is the result of "human desire for understanding". I don't see how those things are different. In any case, the best we can do is to come up with the simplest rule that explains everything we know at any given time, and make it more complex as more is discovered. And that's what everyone is trying to do.

Again, that's backwards. The purpose of a good theory is to simplify, not to add complexity.

 

[Xezlec]

I never claimed facts do not exist. But a (useful) theory is much more than just a set of facts. A good physical theory both systematizes the facts into comprehendable order, suggests/predicts new facts that have not yet been determined, and, the essential part, reduces observed complexity into it's simplified and abstracted components.

Which is all just another way of saying that a good theory uses patterns to simplify the data rather than a list of raw, uncompressed data points. What I don't understand is why you would claim that there might be no unified theory. GR is a pattern. QM is a pattern. In the past, when such otherwise generally applicable patterns have conflicted, there has turned out to be a greater pattern underlying both of them, providing more accuracy, more generality, and more insight.

Hooke's Law is known to derive from more fundamental laws. This is good because we don't have to guess; when you need to know just how far Hooke's Law will take you before it stops working, you can find that out (and get better predictions) by going down to the next level of detail of theory. Nonideal springs can be handled. What's more, those more fundamental laws turn out to describe a whole lot more stuff than just springs.

Again, I never claimed that. My claim is that for a theory to precisely and perfectly reproduce the entire universe would be akin to have a 1:1 scale map.

Can you support this claim? I believe I've shown that the information in the universe can be precisely and perfectly reproduced without a 1:1 scale map, if by that you mean a description the same size as the universe. Some things in the universe have patterns, and information with patterns can be compressed (which just means "described completely by simpler information"). If that isn't what you meant by a 1:1 scale map, then I don't know what you mean. In that case, can we drop the cartographic analogy and speak literally?

The electroweak theory is very mature, AFAIK. It is true that QCD and the electroweak theories have not yet been unified, but again, AFAIK, string theory, loop QG, etc. are all (putative) theories of quantum gravity

It occurs to me that there may be more than one definition of "unification" at play here. Do QCD and QEW actually contradict one another, or is their unification just a "mathematical purity" sort of goal? If the latter, then I suppose grand unification is not part of what I was assuming we were talking about.

That's again, not what I was saying. As you correctly point out, our physical laws are likely to apply 10 minutes from now, so we feel comfortable making predictions. Also, our laws of physics are believed to have applied 10 minutes ago, so we feel comfortable making *postdictions*. However, our initial postulate is that our laws of physics *do not* hold near the initial singular event of the universe, so how can you confidently make *any* statement about the validity of a inherently untestable theory?

That sure isn't my initial postulate. I postulate that a pattern holds except when and where I find reason to believe that it does not. We don't believe these laws hold, as stated, at the big bang, but that isn't a postulate, it's the result of careful reasoning which shows that our known laws are not consistent with the big bang.

What other basis do you have for drawing this distinction? Why would you otherwise claim that the immediate past or future be more likely to obey the laws of physics as we know them than the big bang would? I can think of only one response: the big bang is further away from us than the 10-minutes past or future. However, we now have a set of rules that seems to hold true for a pretty long time and a pretty good distance. In the past, these laws have been extended by finding problems at those kinds of edges, extrema where they don't agree, and then trying to find the simplest way to reconcile those contradictions. Currently, they don't agree at the big bang. The simplest way to reconcile them there may be reasonably expected to lead yet again to a better, more general theory.

Actually, that's very backwards and not how the overwhelming amount of scientific progress has occurred. In general, first comes observation of a phenomenon, then comes experiment and manipulation of the phenomenon, *then* comes theory. Which is one reason why today's theorists are having so many problems making progress- there's nothing to compare results to.

Are you claiming that relativity is a bad example? Or do you believe that relativity was somehow not a theoretical prediction that preceded experimental validation?

I don't see how you think what I said is "backwards". The number of experiments that culminate in a theory is usually dwarfed by the number of future experiments whose outcomes that theory correctly predicts. Theory fills in huge experimental gaps. Someone didn't test Hooke's Law at every possible displacement of every possible spring at every point in the universe. They just tried it at a few points, and made a theory to describe and generalize an observed pattern. The theory predicted what would happen at all points in between, and that turned out to hold remarkably well.

Relativity is still a good example. The phenomenon observed experimentally was the constancy of the speed of light. That's what I'm calling a contradiction with an existing theory. The new, better theory was formed by reconciling that with the other laws known at the time. That sounds like just what you described. Theory certainly does proceed from "experiment," it's just that the definition of an "experiment" can be flexible. For the purposes of producing improvements on existing theory, an experiment is really anything that illustrates a contradiction with a previous theory. In this case, the big bang and black holes could be seen as "experiments". That's where we differ. You say we can't perform experiments there. I say, insofar as we care about it (that is, insofar as it conflicts with existing theory), these things are the experiments.

The problem is, as you say, that we have fewer experiments now. Every contradiction of existing theories is like an experiment, but we can't just go out and perform more, because these are a different kind of experiment. We only have what we have. So, when more than one theory explains an existing contradiction, it can be hard to find other things to confirm that one of them explains more than the others. You would need to find more contradictions, either through thought experiments or giant machines.

Again, that's backwards. The purpose of a good theory is to simplify, not to add complexity.

It is a simplification of the previous theory plus some accumulated body of contradictions or exceptions to it, but it is an increase in complexity over the previous theory by itself. The universe is complicated, and therefore theory gets more complicated as it gets more accurate. Remember, we want to keep things "as simple as possible, but no simpler".

 

[Fra]

Why can't quantum mechanics and relativity just be separate and equally beautiful?

Does there HAVE TO BE a unified theory...and if so, explain why.

I'd like to question the meaning with "to be" here. I'll speak from my point of view, either I know of a unified theory or I don't. This morning I knew of noone. And it seems no one else has one either. Still life goes on. So in this sense the answer to your question is does there have to be a unified theory is No.

I think the question is really why we are LOOKING FOR a unified theory, or at least looks for FURTHER unification, even if not the ultimate one.

The answer to this appears to me as one of the basic traits of life and organisation of the world. Life thrives on trying to take advantage of it's environment in order to get an advantage and survive. Part of this, I see is self-organisation and seeing simplicity in what was previously complex. It's closely related to progress. The ambition to make progress is almost inseparable to the desired of survival. Without this ambition, we probably wouldn't be here. I think this ambition not only applies to humans, but also to physical systems, if you extend survival of biological organisms to survival and self-preservation of an arbitrary physical systems as in stability.

Somehow, to make sense and see persistent a simplicity in the big, is the basis for stability of an observer. Those who fail to do so, won't stick around. Those who do, will. That's how I see it, as an element of evolution.

/Fredrik

 

[Ivan Seeking]

The justification for the pursuit of a unified theory is the same one used to justify any fundamantal reasearch - the potential for discovery. The fact that we don't have a complete model tells me that we might yet have some very important things to learn. There is no way to know the signficance of what might be discovered, but history tells us that a relatively small investment in science can vastly improve the human condition.