Why The Big Bang banged and didn’t collapsed into a black hole?

[PAllen]

I'm not sure I understand. In eternal inflation models, there is no "start of inflation" anywhere. What creates our "bubble" is the end of inflation within that bubble--inflation ending is what "separates" our universe from the rest of the eternally inflating spacetime.

Actually, even in inflation models without eternal inflation, I'm not clear about what "the start of inflation" means. The spacetime geometry during the inflation epoch is basically de Sitter (since the inflaton field in this epoch has the same stress-energy tensor as a cosmological constant), and de Sitter spacetime has no initial singularity.

It doesn't matter whether there is a singularity or not. We could even consider Hawking's no boundary universe. In any model (manifold) that includes the current observable universe, there are a set of events causally connected to any part of the currently observable universe. FLRW solution provides a standard time slicing back to the end of inflation. One can then pick some isotropic slicing before then to cover all that is causally connected to our the observable universe. That may mean it is better to think of the beginning as t -> -∞, rather than t=0. So what?

 

[PeterDonis]

We could even consider Hawking's no boundary universe.

And what, in your view, would the term "Big Bang" signify in such a universe?

That may mean it is better to think of the beginning as t -> -∞, rather than t=0. So what?

If the "beginning" is $t \rightarrow - \infty$ (for example, if everything prior to the end of inflation is a de Sitter spacetime region), what, in your view, would the term "Big Bang" signify in such a universe?

 

[PAllen]

And what, in your view, would the term "Big Bang" signify in such a universe?
If the "beginning" is $t \rightarrow - \infty$ (for example, if everything prior to the end of inflation is a de Sitter spacetime region), what, in your view, would the term "Big Bang" signify in such a universe?

The earliest history covered by the model, whatever its coordinate value is in some standard coordinates. For de sitter, negative infinite is the most natural, but of course it is trivial to introduce coordinates that scale this to a finite value (at the cost of hiding some of hiding natural time symmetry).

 

[PeterDonis]

The earliest history covered by the model

I don't understand. Isn't the model supposed to cover everything? Isn't that what we mean when we say the "beginning" is $t \rightarrow - \infty$? Or in the Hawking no boundary case, isn't the model supposed to include the no boundary region?

 

[PAllen]

I don't understand. Isn't the model supposed to cover everything? Isn't that what we mean when we say the "beginning" is $t \rightarrow - \infty$? Or in the Hawking no boundary case, isn't the model supposed to include the no boundary region?

Not sure what our miscommunication is. I am including everything. I said so very explicitly. Earliest doesn’t have to be a particular moment, as I mean it. I literally mean all that is past causally connected to the observable universe. That would exclude other bubbles in an eternal inflation, but that is all. In standard coordinates, the Big Bang would be asymptotically minus infinity in a de sitter universe. In other coordinates you could make it whatever finite value you wanted, i.e. t‘ > 5 would equivalent to t > -∞ .

 

[PeterDonis]

In standard coordinates, the Big Bang would be asymptotically minus infinity in a de sitter universe.

But that is not part of the universe; there is no event or spacelike hypersurface at "asymptotically minus infinity".

It is true that the initial singularity in non-inflation FRW models is also not part of the universe, but it does have the important property of geodesic incompleteness; timelike and null godesics cannot be extended into the past at or beyond a finite value of their affine parameter. That is what justifies the usual interpretation of the initial singularity as "the beginning of the universe" in these idealized models, and that in turn motivates the use of the term "Big Bang" to refer to it. de Sitter spacetime does not have this property, so I don't think "asymptotically minus infinity" is a reasonable referent for the term "Big Bang" in such a spacetime.

In other coordinates you could make it whatever finite value you wanted

But that finite value would not label part of the actual spacetime; it would label some boundary that is only present in an extension, as the boundaries in Penrose coordinates do.

In any case, I am not concerned with coordinate choices or coordinate-dependent properties; I am assuming that any reasonable interpretation of the term "Big Bang" will refer to something that is picked out by physics, not just a coordinate choice.

 

[PAllen]

But that is not part of the universe; there is no event or spacelike hypersurface at "asymptotically minus infinity".

It is true that the initial singularity in non-inflation FRW models is also not part of the universe, but it does have the important property of geodesic incompleteness; timelike and null godesics cannot be extended into the past at or beyond a finite value of their affine parameter. That is what justifies the usual interpretation of the initial singularity as "the beginning of the universe" in these idealized models, and that in turn motivates the use of the term "Big Bang" to refer to it. de Sitter spacetime does not have this property, so I don't think "asymptotically minus infinity" is a reasonable referent for the term "Big Bang" in such a spacetime.
But that finite value would not label part of the actual spacetime; it would label some boundary that is only present in an extension, as the boundaries in Penrose coordinates do.

In any case, I am not concerned with coordinate choices or coordinate-dependent properties; I am assuming that any reasonable interpretation of the term "Big Bang" will refer to something that is picked out by physics, not just a coordinate choice.

And I have said several times that I don’t view the big bang as being an event or specific hypersurface. I don’t disagree with what you said, I just don’t find it relevant to my definition.

 

[PeterDonis]

I don’t disagree with what you said, I just don’t find it relevant to my definition.

Fair enough. At this point we've given the OP more than enough to chew on, anyway.

 

[timmdeeg]

Whereas we know that there was a hot, dense, rapidly expanding state that is the earliest one for which we have good evidence; the question is what came before it, inflation or something else. Calling that earliest known state the "Big Bang" is the only use of that term that is model independent.

Ok and as I understand you most cosmologists would agree with that.

We have some indirect evidence that inflation happened but it seems not possible to talk about a "state" of the universe before inflation in terms of physical properties like density, temperature or the like. I would still prefer to identify the Big Bang with the very "begin" (to replace $t = 0$) of the universe but from the above this has caveats. Thanks for your explanations.

 

[PeterDonis]

We have some indirect evidence that inflation happened but it seems not possible to talk about a "state" of the universe before inflation in terms of physical properties like density, temperature or the like.

In the "eternal" inflation models, which seem to me to be the ones mostly favored now, there is no "before" inflation; inflation extends indefinitely into the past ("asymptotically to minus infinity" as @PAllen put it). The state of the universe in any inflation region of spacetime is pretty boring: the inflaton field is in its "false vacuum" state (which is what causes inflation), and nothing else is present.

 

[timmdeeg]

In the "eternal" inflation models, which seem to me to be the ones mostly favored now, there is no "before" inflation; inflation extends indefinitely into the past ("asymptotically to minus infinity" as @PAllen put it).

But you don't say the de Sitter like state of that patch of the universe which created our bubble "extends indefinitely into the past", right? I have a problem with "inflation extends indefinitely ..." because as I understand it inflation means exponential expansion which was extremely short.

Well after all I tend to see the Big Bang after inflation which also fits to the solution of the horizon and flatness problem.

 

[PeterDonis]

you don't say the de Sitter like state of that patch of the universe which created our bubble "extends indefinitely into the past", right?

In eternal inflation models, there is only one de Sitter like patch, the patch that is eternally inflating. As I said before, each "bubble" gets created when inflation ends in a small pocket and that pocket experiences a "Big Bang" (in my preferred meaning of the term--a hot, dense, rapidly expanding state with lots of energy in the Standard Model fields, the energy having just come from a "reheating" event in that small pocket when inflation ended there). The de Sitter like patch extends indefinitely into the past.

I have a problem with "inflation extends indefinitely ..." because as I understand it inflation means exponential expansion which was extremely short.

In eternal inflation models, inflation is eternal, not extremely short.

 

[timmdeeg]

In eternal inflation models, there is only one de Sitter like patch, the patch that is eternally inflating. As I said before, each "bubble" gets created when inflation ends in a small pocket and that pocket experiences a "Big Bang" (in my preferred meaning of the term--a hot, dense, rapidly expanding state with lots of energy in the Standard Model fields, the energy having just come from a "reheating" event in that small pocket when inflation ended there). The de Sitter like patch extends indefinitely into the past.

In eternal inflation models, inflation is eternal, not extremely short.

Thanks, very interesting stuff. I want to deepen this a little more and found 3. THE
INFLATIONARY UNIVERSE
This article written by Alan Guth doesn't seem to be pop-science of the sort which isn't recommendable. But if otherwise please give me a hint.

 

[PeterDonis]

This article written by Alan Guth doesn't seem to be pop-science of the sort which isn't recommendable. But if otherwise please give me a hint.

Here's a hint: is it a textbook or a peer-reviewed paper? If not, it isn't recommendable. That doesn't necessarily mean it's incomplete or misleading; it just means that, unless you yourself are an expert in the subject, you can't tell whether the scientist is (most likely unintentionally) saying something that is incomplete or misleading. Textbooks and peer-reviewed papers have other experts reviewing what is said to catch those things.

 

[timmdeeg]

Here's a hint: is it a textbook or a peer-reviewed paper? If not, it isn't recommendable.

So this rule is set in stone and and famous names and formulae don't change that.

That doesn't necessarily mean it's incomplete or misleading; it just means that, unless you yourself are an expert in the subject, you can't tell whether the scientist is (most likely unintentionally) saying something that is incomplete or misleading.

Or unless I ask an expert in the subject. Is this article incomplete or misleading?

 

[PeterDonis]

So this rule is set in stone and and famous names and formulae don't change that.

If you have a particular source that isn't a textbook or a peer-reviewed paper, and you are sure that it is reliable, you can always reference it, and the moderators can always weigh in if they need to.

But if you have to ask whether a particular source is reliable, that means you don't already know that it is. Then the rule I gave is the best rule I can give you.

Or unless I ask an expert in the subject.

Sure, but then be prepared to be told that no experts have the time to take a detailed look. None of us get paid for this.

Is this article incomplete or misleading?

I have no idea since I haven't had time to read it. If I have time I will take a look and give any feedback that I think is relevant.

 

[PAllen]

After look at this article, I would describe thus:

It is a presentation of Alan Guth‘s own views on cosmology circa 1997 at the reader level and quality of a better than average Scientific American article. Good pop sci by a highly qualified author, but not remotely a professional article. If Guth taught an undergraduate course on this material for astronomy majors, the level would be far more precise and rigorous than this, let alone in one of his many papers.

 

[Brad Watson - Miami]

Our current best models say that the universe is infinite in spatial extent, in which case it was always infinite.

That's false. The BIG Bang(-Bit Bang) is our best model for the beginning of the Universe 13.8 billion years ago. Our Universe is clearly not infinite in space or eternal in time in our past. Perhaps you mean perpetual.

 

[Nugatory]

Our current best models say that the universe is infinite in spatial extent, in which case it was always infinite.

That's false. The BIG Bang(-Bit Bang) is our best model for the beginning of the Universe 13.8 billion years ago. Our Universe is clearly not infinite in space or eternal in time in our past. Perhaps you mean perpetual.

You may be misunderstanding Ibix’s point. Our best models say that the universe is infinite in spatial extent, and always (that is, at the time of and ever since the Big Bang) has been.

 

[Ibix]

That's false. The BIG Bang(-Bit Bang) is our best model for the beginning of the Universe 13.8 billion years ago. Our Universe is clearly not infinite in space or eternal in time in our past. Perhaps you mean perpetual.

As Nugatory has already noted, you appear to be confused about what I was claiming. The universe, according to our current models, is infinite in spatial extent. Exactly what happened 13.9bn years ago is still a topic of research. Naive models can be read as suggesting that the universe started then, but I believe that opinions among more sophisticated models differ on what went on. But whether the universe has a finite or infinite past, if it is currently infinite in spatial extent then it can never have been finite in extent.