PeterDonis said:
Not at all. We can both work backwards from observations, using some particular model, and work forwards from an assumed initial condition, using some particular model.
Of course different theorists will have different preferences; your preference appears to favor the "backwards" mode. But your preference is not the same as an absolute requirement.We don't know the complete laws of physics. So there is no way of enforcing your requirement. We can say what our currently known laws of physics predict, but our currently known laws of physics are not the same as "the" laws of physics.Since, as you have already said, this domain only goes back so far, we cannot apply this prescription further back than that. See next comment.You have already said that the Standard Model can only take us backwards so far. We can't apply this prescription further back than that, and "further back than that" includes the end of inflation in inflation models.If this is true, then, as I have already said, pronouncements about what is "ruled out" or about what does or does not have something to back it up are not justified. So you should stop making such pronouncements.
There is some point, it looks like t=10
-12, but maybe somebody thinks it is really t=10 minutes, where we are at energies in which the only thing we have to assume is that the laws of physics at a given energy are the same as they are today, at which we have reasonable certainty. We can deduce that there is asymmetry by that point in time . We can't have Big Bang Nucleosynthesis, for example, at a state of matter-antimatter asymmetry.
We can say what our currently known laws of physics predict, but our currently known laws of physics are not the same as "the" laws of physics.
We can't be sure of them up to infinite energy scales, but we can be pretty darn comfortable about them in their proven domain of applicability.
And., I think
@kimbyd is way too pessimistic about what we know already. We know enough to dramatically curtail the magnitude and nature of what we don't know. If we've seen absolutely no evidence of BSM physics of some kind at one energy scale, there won't be overwhelming evidence of it "just around the corner" at slightly higher energy levels. High energy physics is too interconnected for that to happen. So, the cutoff point is necessarily fuzzy. The bigger the change from New Physics, the higher the energies at which it has to occur.
Yes, there is a gap of t=0 to this cutoff point at which all bets are off, we have no idea what the laws of physics are, we have no direct observations.
All we know is that the New Physics from t=0 to this cutoff point in the very early universe needs due to the early Universe physics, old or new, need to dovetail to the known physics a little bit later. If inflation has consequences, those consequences need to have manifested by that point. If there is new source of matter-antimatter asymmetry, it needs to have manifested by then.
Now, as long as you can come up with New Physics that gets you from t=0 to this cutoff with the right conditions at the cutoff, you're all good.
f your proposed New Physics at ultra high early Universe energies doesn't get you to where you get from solid ground, however, you have a problem.
But it's not what you get when you take the condition at the end of inflation and apply the known Standard Model to the reheating process. So we are clearly missing something somewhere.
In order words, something about the conditions at the end of inflation or the beyond proven range of applicability Standard Model reheating process is wrong. That's what "we are clearly missing something somewhere" means. We may not be sure how it is wrong, but we know we need to keep looking for physics that can meet our minimum requirements.
Oh nonsense. It is perfectly reasonable to work forward from theoretical initial conditions and see what you're model gets you, even if that model gets you something that doesn't make sense at the time. The only time you're wrong is when you try to say that your model is correct and it can't have been another way. No one is saying that about any theory or model current in use by science. Even a grossly incorrect model or theory is useful in the sense of telling scientists, "This probably isn't the right way of going forward."
If you work forward from theoretical initial conditions until you get to the point where we have observations and physical laws within their domain of applicability, yes, your theory really is wrong. Maybe you weren't trying to get it right. But it really doesn't seems so outlandish to think that we should devote more attention to models that work or come close to working than to models that don't come remotely close. It's like the vacuum energy paradox. We know that something about the way that is formulates is grossly out of touch with reality so we need to be thinking about the problem very differently rather than continuing down that path.
We do have a wildly under-constrained theory space of possible New Physics.
We have almost no experimental data to guide us. We have assumed as an axiom that the laws of nature that we know and love might not work, maybe even probably won't work. But, we can
at least start discounting the theories that don't get us where we need to get and pay more attention to the ones that do.
Is there any other legitimate, scientific, way to prefer one option over another?
So, for example, if we think we need inflation and we are just convinced for no really defensible reason that t=0 has equal amounts of matter and antimatter, we should be much more favorably inclined to an inflaton that gives rise to matter-antimatter asymmetry, than to one that does not, because the CP violation in the CKM matrix and PMNS matrix isn't going to cut it to accomplish those ends.
We can take "no new physics" back well beyond the experimentally proven domain of applicability of the Standard Model and see what initial conditions that will give us. This is something we can do. It doesn't carry the same proven certainty, but it is hard to see any other better baseline to compare other theories to.
We could also take the position that we ought to look for the keys we lost under the streetlight, because that's the best we can do at the moment.
Or, maybe we ought to be able to acknowledge, in much the same way that we do about what lies beyond the event horizon in a black hole, or what exists beyond the observable Universe, that there are somethings that are just intrinsically unknowable.
The central point is all of this, however, is that no axioms uncorroborated by data are entirely to any deference just because we think that they are beautiful. If we get axioms that produce good results and they are also beautiful, swell. But, right now, we have some quite ugly theories that get the job done and there is no good reason to believe that the provable discoverable scientific truth will ever get prettier.
Ideas like "naturalness", "fine tuning", and resorts to Platonic ideals that we pull out of our butts, are not science. The idea that we have to start at matter-antimatter balance at t=0 has nothing to back it up. The idea that there should be strong force CP violation has nothing to back it up. We shouldn't presume to know what the laws of nature should look like. Going down that road is a waste of time and effort that has failed over and over and over again for decades now.