Thursday, May 11, 2017

Why the "testability" criticisms of inflation are silly

Because the inflation revolution is analogous to having eaten a forbidden apple

In February, three critics of inflation Ijjas, Steinhardt, Loeb (ILS) published a diatribe in Scientific American titled "Cosmic Inflation Theory Faces Challenges". They tried to defend the seemingly indefensible – the claim that there exist reasons to abandon the inflationary cosmology. They combined various unflatteringly sounding, mostly irrational sentences about the experimental status of inflation as well as its theoretical underpinnings.

As Sean Carroll and Peter W*it have mentioned, a day or two ago, dozens of authors signed under the response (also) in Scientific American named "A Cosmic Controversy". Note that ILS's title already tried to summarize their opinion while the title of the pro-inflation article doesn't make it clear that it's pro-inflation. This pattern can be seen repeatedly: Wrong statements often appear as titles but correct ones almost never do. Why is it so? I think that the journalists believe that more readers are attracted when the title is a wrong proposition.

The new pro-inflation text was penned by folks like Guth, Linde, Kaiser, Nomura (GLKN) but also by famous folks like Hawking, Witten, Maldacena, Susskind, occasional TRF guest bloggers Randall, Silverstein, but also by Sean Carroll, among others.

Most cosmologists would agree that the inflationary cosmology is a vital fundamental building block in most of the thinking about cosmology in the modern era. In its rather general form, the theory of inflation says that the Universe has undergone a period of intense, approximately exponential expansion driven by a scalar field that was away from the minimum where it's sitting now. The previous sentence is a huge insight but it's not a complete theory so most of the detailed questions may remain – and indeed do remain – unanswered even if you deduce all derivable consequences from the paradigm that I have already described.

When you deduce some basic universal consequences of the paradigm and/or the simplest models, you will make predictions that seem extremely encouraging, to say the least, and in some cases, they precisely agree with some highly nontrivial observations.

The basic predictions are that the Universe is very large, almost smooth, has a nearly uniform temperature of the cosmic microwave background, but also has some non-uniformities whose structure may be largely predicted, and complex curves predicted by the theory are exactly observed. The success is rather stunning and even though cosmology used to be much less hard, more philosophy-like scientific discipline, unlike particle physics, many of these predictions, work with possibilities, and experimental confirmations are totally analogous to those in particle physics.

If I were asked to formulate a concise argument in favor of inflation that I find most relevant, I wouldn't discuss any details about what are the simple models of inflation, whether we work with simple models or all models, or other fuzzy questions. Instead, I would tell you:
What happened when cosmologists began to study inflation and talk about inflation every day? Well, Alan Guth and pals have basically made the following simple observation: Scalar fields that appear in theories from the same class as the Standard Model may also have local maxima or places away from the minima where the scalars may be located for reasonably long periods of time.

When a scalar field spends some time near the local maximum, the potential energy \(V(\Phi)\) affects the cosmology just like a constant energy density of the vacuum – the cosmological constant (however, numerically a huge one) – and that makes the Universe expand just like a de Sitter space.

And when you study what happens to the Universe with a scalar field in such a state, you will find out that it was expanding and it was doing so in a way that seems strikingly similar to what is needed to get the large, almost smooth Universe around us, with some galactic structures born into it, and so on.
That's it. The point is that Guth, Linde, and others have realized that all physicists and cosmologists before them were overlooking a type of behavior that is very likely to take place in theories of particle physics that are very analogous to the Standard Model – theories with scalar fields. They were overlooking some interesting and almost unavoidable behavior of the scalar fields near the local maxima – and this behavior may be immediately seen to be very helpful to explain some features of the Universe around us.

Many important overall features seem to be explained well almost immediately and that's a reason why you should never overlook inflation – this expansionist behavior of the scalar field away from the minima – again! Instead, you should always take it into account. You can't be told in advance what will be all the conclusions you will be led to – but that's correct. You should go wherever the evidence will lead you! It's not the purpose of science to create prophesies about all detailed twists that will happen in the future. The real point is that since the late 1970s, it looks like an obvious fallacy to keep on ignoring inflation.

The most colorful, precise enough analogy for my claim is that Guth and Linde etc. became the Adam and Eve who just found and ate the apple in the Garden of Eden. This apple also gave them some new skill – it was a recipe for coitus or something like that, sorry, I don't have a full-blown education in Judeo-Christianity ;-) – and they haven't forgotten the lesson even since.

So the researchers who are using inflation are not fanatics who would resist a clearly even better theory in the hypothetical case that such a theory is proposed. However, some speculations about future discoveries shouldn't distract you from the current state of cosmology and the current state is that inflation seems to be the best explanation for the basic features of the Universe we know and even from the perspective of quantum field theory, one should agree that something like inflation is very likely to influence the Universe at some point.

What ILS – the critics of inflation – are doing is to ask us to forget about the apple in the Garden of Eden once again. They tell us to uneat the apple, unlearn the skill of coitus, and start overlooking the expansionist behavior of the scalar fields once again. We should forget that scalar fields have points away from the minima in the configuration space and we should forget the lessons what these fields are doing to the Universe when they are located over there. Apologies but the lesson of the eaten apple is irreversible. You can't make the researchers forget about such important things once again. It would be utterly irrational and unscientific to do so. Unlike religions, science simply isn't compatible with the commandment "you shall forget about all your previous gods that the ILS prophets have declared blasphemous". In science, "not to overlook" is superior in comparison with "to overlook". Period.

You can only forget about detailed proposals that have actually been ruled out. So many very specific models of inflation have been shown to produce predictions that disagree with the experimental data. It's enough to abandon the detailed models but it is not enough to abandon the inflationary framework. It's common sense. The garden analogy exists. In some countries, some forms of sex – like one with children – may be banned but that's different from banning sex altogether. People more or less creatively probe various types of sex, while taking various desires, pressures, and laws into account, but they don't generally forget the lesson from the apple in the garden of Eden.

Whether the framework is easily falsifiable has nothing whatever to do with the question whether it's correct. Only imbeciles are confusing these two things. And researchers should primarily search for theories that are correct, not those that are easy to work with. Some tasks awaiting a scientist may be hard or messy or they may take a long time but that doesn't prove that the direction of that journey is incorrect.

To be specific, let me pick a sentence from the ILS rant that talks about the alleged untestability of inflation:
In other words, scale invariance is possible but so is a large deviation from scale invariance and everything in between, depending on the details of the inflationary energy density one assumes. Thus, the arrangement Planck saw cannot be taken as confirmation of inflation.
The problem aren't these statements by themselves. The problem is that ILS think that the sentences imply something bad about inflation. Well, they don't. They're completely analogous e.g. to the following statements about Darwin's theory:
Darwin's natural selection implies that the survival of large animals is possible but so is the survival of small animals and animals in between. Therefore, the sizes of observed animal species cannot be taken as a confirmation of Darwin's theory.
Well, right, Darwin's theory implies that it may be very clever to be a big animal in some situations or a small one or a medium one. The theory doesn't make a unique clearcut recommendation for the animals how big they should be. But that doesn't mean that there is something wrong with the theory. Even more importantly, it doesn't mean that there is nothing correct about the theory. The fact that one particular would-be piece of data isn't positive evidence in favor of the theory doesn't mean that there is no positive evidence. Other things are the positive evidence – both in the case of Darwin's and inflationary theories. There is no obligation for a theory to give some unambiguous recommendations of this kind for every quantity that may be discussed within the theory. Some theories combined with some selection of the data and constraints simply allow numerous answers to a question. There's nothing wrong about it. A scientific theory predicts things but especially if you only pick some rough summary of the scientific theory, it just doesn't predict everything.

What I say about these matters sounds like pure common sense to me. I can't believe that someone is incapable of understanding this simple point. Nevertheless, all the people who whine about "testability" seem to misunderstand it. And that's why I think that all of them are just plain idiots.

Once the cosmologists have eaten the apple of inflation, they may study various questions such as:
  • Can they determine the precise selection of the required inflaton field (or fields) and their potential that fits the detailed data or that is derivable from a deeper (string) theory?
  • Can they derive the inflaton as an effective field describing some more detailed microscopic physics (such as the distance between two branes)?
  • Can they deduce some predictions for so far undiscovered phenomena (such as the primordial gravitational waves)?
  • Does the theory predict a large multiverse which connects inequivalent environments and do the observations constrain such scenarios? The more anthropic a paper about the multiverse is, the more irrational it seems to be, but the very existence of a multiverse may be predicted and it's just foolish to try to ban it just because it looks too large and therefore untestable to you.
  • Is it possible to say about the vacuum selection in string theory or particle physics in general by extending the cosmological considerations?
  • Can one design theories where the successes of inflation are explained truly differently?
And others. They're detailed questions and there's obviously some disagreement about every new theory that tries to say something rather specific about these questions – and much finer, more technical questions that cosmologists actually study.

But the problem with ILS is that they don't want to rationally discuss particular questions. They want to ban inflation as a blasphemy and use this ban to place some of their pet theories – cyclic or ekpyrotic Universe or whatever they like now – to the place of the inflation. However, they don't actually have any scientifically valid arguments for that which is why the people who keep on thinking as scientists won't do it.

I have spent many and many hours by reading papers and listening to talks about the cyclic cosmology etc. I think that the percentage of indisputably wrong statements – e.g. about the second law of thermodynamics – is high among the people proposing some things. Some versions of these theories could work but I would view them just as generalizations of inflation (e.g. with some extra-dimensional interpretation of the inflaton, e.g. as the distance between branes in the direction of a new dimension of space).

Concerning the cyclic character itself, I haven't seen any argument or calculation that would persuade me that the cyclic cosmology is likely. But even if it were true, the last cycle from the "Bang" is clearly the most important one for explanations of what we can see – the previous history has been scrambled, melted, and washed away. So the obsession of ILS with convincing the reader that even the "Bang" in the term "Big Bang" is semi-blasphemous seems utterly irrational. Even if there were a pre-history, the recent history will still be more important and will often be talked about separately. It's foolish to expect otherwise.

I have also spent many and many hours with some smart and nice folks who were convincing me e.g. that their string/brane gas cosmology is a totally viable – and maybe superior – competitor to inflation. Well, I am still not 100% sure about these matters but I simply couldn't quite find the "corresponding features" in the string gas cosmology that would demonstrably reproduce similar successes as those that make me believe in inflation.

Most cosmologists find themselves in a rather similar condition. They are somewhat open-minded about explanations that are very different. But those don't quite make as much sense to them as inflation which is why these alternative paradigms aren't substituted as the fundamental building blocks in these people's thinking about matters. ILS haven't actually brought us any rational reason to do so. Screaming that some paradigm is blasphemous because it's not easily falsifiable just won't make it. People don't have any rigorous proof that the human society may live without sex (the counterpart of the "falsification") but that doesn't mean that the people will stop practicing it.

And again, we have many very different reasons to take inflation seriously. Even if Cumrun or Robert or Ali would persuade me that string/brane gas cosmologies may do all the good things for cosmology to replace inflation, I would still think that a viable field theory – or string vacuum – will probably have some scalars with some local maxima or other inflation-like pieces of the configuration space. So even if the job of inflation could be done by someone else, there would still be reasons to think that some inflation is there, anyway.

Just to be sure, I wouldn't be quite happy about all comments in the pro-inflation GLKN text. For example, we learn
According to the high-energy physics database INSPIRE, there are now more than 14,000 papers in the scientific literature, written by over 9,000 distinct scientists, that use the word “inflation” or “inflationary” in their titles or abstracts.
OK, that's nice that the paper has 33 authors and they mention that some 9,000 others are likely to be on their side but that still doesn't prove that they're right. Instead, critics like ILS are surely willing to say – and they are already weakly saying – that all these 9,000 people are indeed deluded and similar claims are in principle possible. And yes, the public could be willing to buy the opinion that some 9,000 people are a cabal of hacks who have hired each other – similar statements hold for some other groups of 9,000 people. If ILS had an actual valid argument, they could strengthen the case that the 9,000 people are just wrong.

(Don't make me wrong: I think it is good that GLKN plus 30 others have published the pro-inflation article. I just think that such pro-inflation and pro-string etc. semipopular articles should have been written some 10 years ago or so, too.)

But ILS haven't really provided us with any valid argument against inflation. So the article hasn't changed an epsilon about the status quo and masses of cosmologists simply won't be leaving inflation after they have read the ILS rant.

Inflation seems to be a previously overlooked type of behavior of rather generic field theories – those with scalar fields – and what the papers mentioning inflation have in common is that their authors no longer want to make the mistake of overlooking the inflationary regime. Guth, Linde, and perhaps a few others have eaten the apple in the garden of Eden and they can't uneat it again. At most, they may defecate it but some clear falsification – and/or some better food than apples – would be needed for that.

P.S.: Inflation critics ILS have written another "rebuttal" of the defense of inflation. They repeat many things and write e.g. that inflation is highly sensitive to initial conditions because of some new research. I think that it's just not true. Also, we read points like
What has changed about the inflationary theory? Originally scientists thought that the outcome of inflation (a smooth, flat universe with a certain spectrum of density fluctuations and gravitational waves) was generic. Now we know it is not.
If it's not generic (and this adjective seems rather hard to be precisely defined), then it's not generic. Things may be hard. But a smooth Universe with a realistic spectrum of density fluctuations is still much more likely with some random inflation model than in it without inflation – without inflation, it seems almost impossible – so considerations linked to these observed features still dramatically increase the probability ratio of inflation vs non-inflation.

They repeat the comments that there are many kinds of inflation and many vacua and some specific models have been ruled out. But none of these facts actually reduces the probability that the inflationary cosmology is needed to understand the Universe and their opinion to the contrary proves that they are irrational or worse.

No comments:

Post a Comment