I have received a free copy of "Why String Theory" by Joseph Conlon, a young Oxford string theorist who has done successful specialized work related either to the moduli stabilization of the flux vacua, or to the axions in string theory. (He's been behind the website whystringtheory.com, too.)
The 250-page-long paperback looks modern and tries to be more technical than popular books but less technical than string theory textbooks. Unfortunately, I often feel that "more technical than a popular book" mostly means that the book uses some kind of an intellectual jargon – but the nontrivial physics ideas aren't actually described more accurately than in the popular books.
From the beginning, one may see that the book differs from the typical books that are intensely focusing on the search for a theory of everything. Well, the dedication as well as the introduction to each chapter at the beginning of the book (and others) sort of shocked me.
The dedication remains the biggest shock for me: the book is dedicated to the U.K. taxpayers.
It's not just the dedication, however. In the preface, Conlon explains that he wants the "wonderful fellow citizens who support scientific research through their taxes" (no kidding!) to be the readers. He is very grateful for the money.
The preface has only reinforced my feeling that he is "in it for the money". And the theme has continued to reappear in the following chapters, too. It became a distraction I couldn't get rid of. In at least two sections, he mentions that the financial resources going to string theory are much smaller than those in the medical research and the latter funds are still a tiny portion of the budgets.
Great. But why would you repeat this thing twice in a book that is supposed to be about physics? The money going to pure science is modest because most taxpayers are simply not interested in pure science at all. They are interested in practical things. A minority of the people is interested in our pure knowledge of Nature and those would pay a much higher percentage of the budgets to string theory, too. The actual amounts (perhaps a billion of dollars in the U.S. every year?) are a compromise of a sort.
The idea that all taxpayers will be interested in such a book is silly (almost equivalently, it's silly to think that someone will read the book because he is a taxpayer whose money is partly spent for the research; most people don't read books about cheaper ways to hire janitors although this decides about billions of dollars a year, too) and it's hard for me to get rid of the feeling that Conlon's formulations are shaped by the gratitude to the taxpayers for the money – so he's sort of bribed which is incompatible with the scientific integrity. You may imagine that a sensitive reader such as myself reads the text and sees the impact of the "bribes" on various formulations (for example, Conlon's outrageous lie that all string critics are basically honest people is probably shaped by the financial considerations – because many of the string critics are taxpayers) but quite suddenly, the book counts the string theorists by the number of mortgages that people have because of some work that is linked to string theory. Is that serious? And does a majority of string theorists have a mortgage? Whether it's right or not, why should such things matter?
The obsession with the financial aspects of Conlon's job has distracted me way too often. It's totally OK when some people are considering string theory research to be just another job – but it is just not too interesting to read about it. We don't read books about the dependence of other occupations on wages, either. And for a person who is interested in physics sufficiently to buy the book, the money circulating in string theory research is surely a negligible part of the story.
And this financial theme kept on penetrating way too many things. The first regular chapter, "The Long Wait", starts in June 1973. I honestly wouldn't know what event deserving to start the book occurred in June 1973. It turned out that it was the date when the papers on QCD were submitted and in the book, that event is "special" from the today's viewpoint because these were the newest theoretical physics papers as of today that were awarded by the Nobel prize. It seems technically true – Veltman and 't Hooft did their Nobel-prize-winning work in 1971, Kobayashi and Maskawa earlier in 1973, and so on.
But is this factoid important enough to be given the full first chapter of a book on string theory? I don't think so. The fact that no Nobel prizes came to theoretical physicists for their more recent discoveries isn't really important – except for those who are only doing physics because of the money, of course. But even when it comes to the money, numerous people (especially around string theory and inflation) got greater prizes for much newer insights. There are various reasons why the Nobel prizes aren't being given to theoretical physicists for more recent discoveries but these reasons don't imply that breathtakingly important discoveries haven't been made. This focus on June 1973 is just a totally flawed way to think about the importance in theoretical physics – an unfortunate way to start a semipopular book on theoretical physics.
I knew that the following chapter was about scales in physics which is why I was like "WTF" when I saw the first words of that chapter: "As-Salaam-Alaikum". What? ;-) This Arabic greeting means "peace be upon you". What does it have to do with scales in physics? Even when you add the following exchange from the desert that Conlon added, "where are you coming from and where are you going?", this exchange has still nothing to do with scales in physics. At most, the exchange describes a world line in an Arab region of the spacetime. But it has nothing to do with the renormalization group. Perhaps both situations involves diagrams with oriented lines – but that's too small an amount of common ancestry.
Again, one can't avoid thinking: this awkward beginning was probably a not-so-hidden message to the Muslim British taxpayers. Sorry, I have a problem with that. And I think that so do the Muslim Britons who actually care about physics. And no British Muslim will buy a book about string theory because it contains an Arabic greeting so this kind of bootlicking is ineffective, anyway. The bulk of the chapter dedicates many pages to describing the size of many objects. I think that what makes it boring is that Conlon doesn't seem to communicate any deeper and nontrivial – or, almost equivalently, a priori controversial – ideas (something that books like Wilczek's book on beauty are full of). It seems to me that the book is addressed to some moderately intelligent people with superficial ideas about physics and it encourages them to think that they're not really missing anything important. The logic of the renormalization group, "integrating out", or its relationships with reductionism etc. aren't really discussed.
The following, third chapter wants to cover the pillars of 20th century and pre-stringy physics. It starts by talking about special relativity. Conlon argues that the words "In the beginning..." in the Bible (as well as the whole subject of history etc.) contradict relativity. Sorry, there isn't any contradiction like that. Even in relativity, one may sort events chronologically. Different observers may do so differently but it's still possible. And in the history of events on the Earth, the spatial distances are so short relatively to the times times \(c\) (and the reasonable velocities to consider are so much smaller than the speed of light) that all the observers' choices of the coordinates end up being basically equivalent, anyway. So the reference to the Bible has nothing to do with special relativity, just like the Arabic greeting that had started the previous chapter has nothing to do with scales. Perhaps it was a message to the Christian taxpayers. Or the violent atheist taxpayers – because the comment about the Bible was a negative one.
Now, a page is dedicated to special relativity and less than a page to foundations of general relativity. It's really too little and nothing is really explained there. Moreover, general relativity is framed as a "replacement" of special relativity. That's not correct. Einstein would describe it as a generalization, not replacement, of special relativity (look at the name), relevant for situations in which gravity matters. In the modern setup, we view general relativity as the unique framework that results from the combination of special relativity and spin-two massless fields (which are needed to incorporate gravity). In this sense, general relativity is an application of special relativity – and in some sense a subset of special relativity.
Quantum mechanics is given several pages and Conlon says that it absolutely works which is good news. But aside from a few sentences about the quantum entanglement, the pages are mostly spent with repeating that quantum mechanics is needed for chemistry. There are several more sections about the pre-stringy pillars of physics – some cosmology, something about symmetries.
The fourth chapter wants to argue that something beyond the Standard Model and GR is needed. So it's the chapter mentioning the non-renormalizability of gravity etc. Some important points are made, including the point that quantum mechanics must hold universally (Conlon surely is pro-QM). But I can't see what kind of readers (with what background) will understand the explanations at this level. The explanations vaguely depend on some quasi-expert's jargon but they don't say enough for you to reconstruct any actual arguments. I've done lots of this semi-expert writing and it seems absolutely obvious to me that you need to extend the semi-technical explanations at least by an order of magnitude relatively to Conlon's short summaries to actually convey some helpful, verifiable, usable, nontrivial ideas.
If I try to characterize the people who are waiting for this genre that is linguistically heavy but lacking the actual arguments, I think it's right to say that they're "intellectuals who are ready to parrot sentences, even complicated sentences with the jargon similar to the experts' jargon, to defend their intellectual credentials (i.e. impress other people with intelligently sounding sentences)" but who don't really understand anything properly. And I think it's not right to increase the number of such people.
Thankfully, things get better from the fifth chapter that begins with string theory proper. The first event is Veneziano's work in 1968. Conlon describes about 10 non-physics events in the year 1968. It's not clear to me why there are so many events like that. But in such a long list, I think it is crazy not to mention Prague Spring in Czechoslovakia (and the student riots in Paris should be featured more prominently, too). It ended on August 21st, 1968 when 750,000 Warsaw Pact troops occupied my country. To say the least, it was the largest military operation since the war which, I believe, is more important than the cancellation of last steam engines on British railways etc.
The world sheet duality that was deduced from Veneziano's formula hides some cool mathematics but the book unfortunately avoids equations so none of this content is communicated. The beauty and power of all these things may only be understood along with the mathematical relationships etc. which is why I am afraid that this "more detailed" but purely verbal story about the discoveries doesn't bring much to a thinking reader. It's like a book praising a beautiful painting – which doesn't actually show you the painting.
There are various stories, e.g. about Claude Lovelace who realized that bosonic string theory requires 26 dimensions. Lovelace has never completed his PhD but he was hired as a professor at Rutgers, anyway – I was meeting him for many years when I was a PhD student (he died in 2012). A quote in Conlon's book suggests that Lovelace's promotion was insufficient relatively to his contributions. I wouldn't agree with that. At Rutgers, I also knew Joel Shapiro, another early father of string theory. He's a fun guy – he taught group theory to us. A very good course. At a colloquium I gave later, he suggested that the term "first string revolution" should indeed be used for the 1968-1973 era, as Conlon indicates. Whatever is the right name (I called it the zeroth string revolution), it wasn't a superstring revolution because there was no supersymmetry yet!
What seems problematic to me is that the exact chronology of the historical events became the heart of Conlon's prose. But string theory isn't a collection of random historical events, like the Second World War. It's primarily a unifying theory of everything that we don't understand perfectly but the current incomplete understanding is much more accurate and makes much more sense than what people knew in the late 1960s, or in the following 4 decades. A book that is really about physics just can't put all historical events on the same level. The history was just about the "Columbus' journey to the New World" but it's the New World itself, and not details of the journey, that should be the point of a book "Why the New World".
Various discoveries and dualities etc. are mentioned in one or two sentences per discovery. I think it's just too little information about each of them. It may be OK for people who read dozens of redundant books a year and who don't feel the urge to think about every idea that is being pumped into them. For certain reasons, I think that it's counterproductive when people learn about too many facts about string theory (or another theory) without really understanding their relationship and inevitability. If they learn many things, they must feel that string theorists are just inventing random garbage. It feels like science could live equally well without those things (if the events were replaced by totally different events with a different outcome) – just like the mankind could have survived without the Second World War. But it isn't the case. They're deducing something that can't be otherwise – a point you may only verify if you actually know the technology. Well, at some moment, you may start to trust claims about exact dualities etc. by certain authors. But you must see the strong evidence for or a derivation of at least one such an amazing result (or several) to see that it's not just a pile of fairy-tales.
While e.g. Richard Dawid exaggerates (to say the least) the changes in thinking during the string theory era, he does correctly capture the importance of uniqueness (only game in town) and unexpected explanatory interconnections for the string theorists' focus on string theory. Conlon, while a string theorist, seems to completely overlook if not explicitly reject these facts and principles. But they're essential for the understanding where theoretical physicists will look for new insights in the future and how they use the accumulated knowledge to find the new one. So the vision or motivation for the future is therefore basically absent in Conlon's book, too.
Another chapter is about AdS/CFT and the landscape. AdS/CFT was revealed in 1997 – and just like for 1968, Conlon lists many events in 1997. Tony Blair won some elections in a landslide. Holy cow. Every year, there are hundreds of elections in the world and someone wins them. Even the elections in the U.K. are rather frequent. Moreover, I don't understand the logic by which a book like this one should be preferably read by the Britons only. The scientific curiosity is transnational. The book may be "dedicated" to U.K. taxpayers but if it's about science, then it must be equally interesting for the Canadian and other taxpayers, right?
But there are more serious bugs with the content, I think. We're told that after AdS/CFT, almost no one would view string theory primarily as a unifying fundamental theory of Nature. Sorry but that's rubbish. Virtually all top string theorists do. The fact that there are lots of articles that use AdS/CFT methods outside "fundamental physics" doesn't imply that the links of string theory with fundamental physics have been weakened. You may find millions of T-shirts with \(E=mc^2\) which doesn't mean that it's the most important insight made by Einstein.
Similarly, it's wrong to say that the AdS/CFT made string theory "less special". The AdS/CFT correspondence has found a new powerful equivalence between string theory and quantum field theory – but the two sides operate in different spacetimes or world volumes. This holographic duality has made string theory more "inseparable" from the established physics and it became less conceivable that string theory could be "cut" away from physics again – it's because dynamics of string theory inevitably emerges if you study the important established theories, quantum field theories, carefully enough (especially in certain limits).
But if you use a consistent description, it's true (just like it was true before AdS/CFT was found) that in any spacetime where you can see the effects of quantum gravity, you may also see something like strings or M2-branes and the extra dimensions (for the total to be 10 or 11) and other things that come with them. AdS/CFT doesn't allow you to circumvent this fact in any way. It only gives you a new description of this physics of strings or membranes in terms of a theory on a different space, the boundary of the AdS space – a new QFT-based tool to directly prove that there are strings, branes, and various other stringy effects in the bulk. This theory on the boundary happens to be a quantum field theory. But the importance of QFTs in string theory wasn't new, either. Perturbative string theory was always described in terms of a QFT, namely the two-dimensional conformal field theory. The essential point is that this 2D CFT lives on a different space, the world sheet, than the actual spacetime where we observe the gravity. Aside from the world sheet CFT, AdS/CFT has also told us to use the boundary CFT – another QFT-style way to describe stringy physics. But the physics of quantum gravity in the same spacetime as the spacetime of quantum gravity is as stringy as it was before. AdS/CFT has allowed us to explicitly construct many phenomenologically unrealistic sets of equations for quantum gravity (by constructing some boundary CFTs) but it hasn't made the problem of combining particular non-gravitational matter contents with quantum gravity less constraining. An ordinary generic QFT used as a boundary CFT produces a "heavily curved gravitating AdS spacetime" and those may have become "easy" in some way. But the actual, low-curvature theories of quantum gravity are as rare as before.
At least, I found Conlon's discussion of the landscape OK. The large number of solutions is neither new not a problem. The anthropic principle is non-vacuous but it may easily degenerate into explanations that may look sufficient to someone but that are demonstrably not the right ones.
In another chapter, Conlon starts to talk about the "problem of strong coupling". I am afraid that the basic idea that "something is easy at weak coupling, hard at strong coupling" etc. is very easy, much like the usage of the buzzword "nonperturbative". But people who don't really understand and who misinterpret what "easy" and "hard" and "nonperturbative" mean will do so after reading these pages by Conlon, too. Conlon continues with the discussion of the high number of citations of AdS/CFT and reasons why it's exact and correct. An exact agreement about a complicated polynomial-in-zeta-function formula for the dimension of the Konishi operator for many colors makes a lesson clear.
Many pages talk about the application of AdS/CFT correspondence to heavy ion physics; the next section similarly talks about AdS/condensed matter physics. There are many true facts and factoids there. I disagree with Conlon's conclusion in the heavy ion section that adding corrections on top of simplified models is the universal "modus operandi" of science. He uses this thesis to explain that the "exact AdS theory" of the heavy ion physics has to be supplemented with corrections for it to work. That's true and it's normal in much of physics but 1) it is always preferred in physics when adjustments don't have to be added, and 2) it is not how string theory in the strict sense works. String theory does not allow one to add any continuous corrections to its physics, ever. Everything is completely determined by discrete data (identifying the vacuum solution) and the fact that the adjustments are possible in AdS/heavy ion physics shows that those methods are just string-inspired, not examples of full-fledged string theory.
The next chapter talks about the interactions between physics and mathematics. It starts with the pride of physicists. Physics is the deepest science and physicists are Sheldon. No one else can match them – perhaps with the exception of mathematicians. Some insights and facts about mathematics are picked (perhaps a bit randomly) but the main point to be discussed is the flow of ideas in between mathematics and physics.
Monstrous moonshine and mirror symmetry are discussed as the two big examples of string theory's importance in mathematics (excellent topics except that one can't see the beauty without the mathematical "details") while the next section argues against "cults" and chooses Feynman and Witten as the two "cults" that should be avoided. (I think that Witten's cult is basically non-existent, at least outside 3-4 buildings in the world, and I would say "unfortunately".) Progress since the 1980s wouldn't have taken place if everyone were like Feynman; or everyone were like Witten, Conlon says. I actually disagree with both statements. Diversity is way too overrated here. If you had 1,000 Feynmen in physics in the 1980s, I am pretty sure that they would have found the things in the first superstring revolution, too, aside from many other discoveries. One can approach all these things in Feynman's intuitive way. And Conlon overstates how "just intuitive" Feynman papers were. He could have made discoveries with easier formulae because he was normally making fundamental discoveries. But he was the first guy who systematically calculate the Feynman diagrams – from the path integrals to the Feynman parameterization, \(bc\) ghosts that Feynman de facto invented, and beyond. This is in no way "just heuristic/intuitive science".
The disadvantange of Feynman in the real world was that there was only one. Things are even clearer with Witten. I don't agree with Conlon that Witten is only good at things that are "at the intersection of mathematics and physics". Witten has done lots of phenomenology, too, including things like cosmic strings, SUSY breaking patterns, detailed calculations on \(G_2\) holonomy manifolds. 100 Wittens and no one else since 1968 would have been enough to find basically everything we know today. People are different and may have different strengths but that doesn't mean that most of these idiosyncrasies are irreplaceable. It can take more effort for someone to find something – than it takes to someone else – but science ultimately works for everyone who is sufficiently intelligent and hard-working. To say otherwise means to believe that science depends on some magic unreproducible skills.
Chapter 10 is meant to focus on Conlon's characteristic research topics – stabilization of the moduli in compactifications and axions. You may imagine that one needs to know quite a lot to follow what e.g. his papers could have contributed. I think it's basically impossible to convey the information in a semipopular book but he tries. The following Chapter 11 is about quantum gravity in string theory – Strominger and Vafa etc. It doesn't get to recent, post-2009 advances, as far as I can see.
Another chapter argues that all styles of doing physics – revolutionaries and hard workers etc. etc. – are important. It may sound OK but in reality, it's not really possible to classify most physicists by their styles into these boxes at all. Whether someone makes a revolution is ultimately not about his styles and emotions, anyway. And as I said, good physicists may "emulate" what others are doing, despite their having different methods and styles.
The following chapter does a pretty good job in replying some common criticisms of string theory. Then there is another chapter where it's discussed e.g. why loop quantum gravity has remained unsuccessful. I don't think that Conlon describes the status of that proposed theory accurately.
There's a lot of facts and ideas to be found in this book and I obviously agree with a large portion of it. But because of the combination of the "difficult language" and "shortage of actual explanations with the beef", the target audience isn't clear to me, the text seems to be driven by financial and career-wise considerations at too many places (and many of us find these sociological etc. things to be too distracting), and it doesn't go into the sufficient depth for the reader to actually understand that string theory isn't a conglomerate of randomly invented ideas that people are adding arbitrarily (even though Conlon knows very well and explicitly writes that string theory cannot be described in this way). It is not really a book that explains something hard enough (for the layman or the non-expert scientist) and I think that Conlon isn't really an "explainer" in this sense. And I even think that the book reinforces some misconceptions spread by some critics of string theory (e.g. about the impact of AdS/CFT on the status of string theory as a TOE).
You may want to buy the book anyway, to see that it's perhaps not as bad as this text makes it sound.