Tag Archives: academia

Value in Formal Theory Land

What makes a physics theory valuable?

You may think that a theory’s job is to describe reality, to be true. If that’s the goal, we have a whole toolbox of ways to assess its value. We can check if it makes predictions and if those predictions are confirmed. We can assess whether the theory can cheat to avoid the consequences of its predictions (falsifiability) and whether its complexity is justified by the evidence (Occam’s razor, and statistical methods that follow from it).

But not every theory in physics can be assessed this way.

Some theories aren’t even trying to be true. Others may hope to have evidence some day, but are clearly not there yet, either because the tests are too hard or the theory hasn’t been fleshed out enough.

Some people specialize in theories like these. We sometimes say they’re doing “formal theory”, working with the form of theories rather than whether they describe the world.

Physics isn’t mathematics. Work in formal theory is still supposed to help describe the real world. But that help might take a long time to arrive. Until then, how can formal theorists know which theories are valuable?

One option is surprise. After years tinkering with theories, a formal theorist will have some idea of which sorts of theories are possible and which aren’t. Some of this is intuition and experience, but sometimes it comes in the form of an actual “no-go theorem”, a proof that a specific kind of theory cannot be consistent.

Intuition and experience can be wrong, though. Even no-go theorems are fallible, both because they have assumptions which can be evaded and because people often assume they go further than they do. So some of the most valuable theories are valuable because they are surprising: because they do something that many experienced theorists think is impossible.

Another option is usefulness. Here I’m not talking about technology: these are theories that may or may not describe the real world and can’t be tested in feasible experiments, they’re not being used for technology! But they can certainly be used by other theorists. They can show better ways to make predictions from other theories, or better ways to check other theories for contradictions. They can be a basis that other theories are built on.

I remember, back before my PhD, hearing about the consistent histories interpretation of quantum mechanics. I hadn’t heard much about it, but I did hear that it allowed calculations that other interpretations didn’t. At the time, I thought this was an obvious improvement: surely, if you can’t choose based on observations, you should at least choose an interpretation that is useful. In practice, it doesn’t quite live up to the hype. The things it allows you to calculate are things other interpretations would say don’t make sense to ask, questions like “what was the history of the universe” instead of observations you can test like “what will I see next?” But still, being able to ask new questions has proven useful to some, and kept a community interested.

Often, formal theories are judged on vaguer criteria. There’s a notion of explanatory power, of making disparate effects more intuitively part of the same whole. There’s elegance, or beauty, which is the theorist’s Occam’s razor, favoring ideas that do more with less. And there’s pure coolness, where a bunch of nerds are going to lean towards ideas that let them play with wormholes and multiverses.

But surprise, and usefulness, feel more solid to me. If you can find someone who says “I didn’t think this was possible”, then you’ve almost certainly done something valuable. And if you can’t do that, “I’d like to use this” is an excellent recommendation too.

Amplitudes 2025 This Week

1 Reply

Summer is conference season for academics, and this week held my old sub-field’s big yearly conference, called Amplitudes. This year, it was in Seoul at Seoul National University, the first time the conference has been in Asia.

(I wasn’t there, I don’t go to these anymore. But I’ve been skimming slides in my free time, to give you folks the updates you crave. Be forewarned that conference posts like these get technical fast, I’ll be back to my usual accessible self next week.)

There isn’t a huge amplitudes community in Korea, but it’s bigger than it was back when I got started in the field. Of the organizers, Kanghoon Lee of the Asia Pacific Center for Theoretical Physics and Sangmin Lee of Seoul National University have what I think of as “core amplitudes interests”, like recursion relations and the double-copy. The other Korean organizers are from adjacent areas, work that overlaps with amplitudes but doesn’t show up at the conference each year. There was also a sizeable group of organizers from Taiwan, where there has been a significant amplitudes presence for some time now. I do wonder if Korea was chosen as a compromise between a conference hosted in Taiwan or in mainland China, where there is also quite a substantial amplitudes community.

One thing that impresses me every year is how big, and how sophisticated, the gravitational-wave community in amplitudes has grown. Federico Buccioni’s talk began with a plot that illustrates this well (though that wasn’t his goal):

At the conference Amplitudes, dedicated to the topic of scattering amplitudes, there were almost as many talks with the phrase “black hole” in the title as there were with “scattering” or “amplitudes”! This is for a topic that did not even exist in the subfield when I got my PhD eleven years ago.

With that said, gravitational wave astronomy wasn’t quite as dominant at the conference as Buccioni’s bar chart suggests. There were a few talks each day on the topic: I counted seven in total, excluding any short talks on the subject in the gong show. Spinning black holes were a significant focus, central to Jung-Wook Kim’s, Andres Luna’s and Mao Zeng’s talks (the latter two showing some interesting links between the amplitudes story and classic ideas in classical mechanics) and relevant in several others, with Riccardo Gonzo, Miguel Correia, Ira Rothstein, and Enrico Herrmann’s talks showing not just a wide range of approaches, but an increasing depth of research in this area.

Herrmann’s talk in particular dealt with detector event shapes, a framework that lets physicists think more directly about what a specific particle detector or observer can see. He applied the idea not just to gravitational waves but to quantum gravity and collider physics as well. The latter is historically where this idea has been applied the most thoroughly, as highlighted in Hua Xing Zhu’s talk, where he used them to pick out particular phenomena of interest in QCD.

QCD is, of course, always of interest in the amplitudes field. Buccioni’s talk dealt with the theory’s behavior at high-energies, with a nice example of the “maximal transcendentality principle” where some quantities in QCD are identical to quantities in N=4 super Yang-Mills in the “most transcendental” pieces (loosely, those with the highest powers of pi). Andrea Guerreri’s talk also dealt with high-energy behavior in QCD, trying to address an experimental puzzle where QCD results appeared to violate a fundamental bound all sensible theories were expected to obey. By using S-matrix bootstrap techniques, they clarify the nature of the bound, finding that QCD still obeys it once correctly understood, and conjecture a weird theory that should be possible to frame right on the edge of the bound. The S-matrix bootstrap was also used by Alexandre Homrich, who talked about getting the framework to work for multi-particle scattering.

Heribertus Bayu Hartanto is another recent addition to Korea’s amplitudes community. He talked about a concrete calculation, two-loop five-particle scattering including top quarks, a tricky case that includes elliptic curves.

When amplitudes lead to integrals involving elliptic curves, many standard methods fail. Jake Bourjaily’s talk raised a question he has brought up again and again: what does it mean to do an integral for a new type of function? One possible answer is that it depends on what kind of numerics you can do, and since more general numerical methods can be cumbersome one often needs to understand the new type of function in more detail. In light of that, Stephen Jones’ talk was interesting in taking a common problem often cited with generic approaches (that they have trouble with the complex numbers introduced by Minkowski space) and finding a more natural way in a particular generic approach (sector decomposition) to take them into account. Giulio Salvatori talked about a much less conventional numerical method, linked to the latest trend in Nima-ology, surfaceology. One of the big selling points of the surface integral framework promoted by people like Salvatori and Nima Arkani-Hamed is that it’s supposed to give a clear integral to do for each scattering amplitude, one which should be amenable to a numerical treatment recently developed by Michael Borinsky. Salvatori can currently apply the method only to a toy model (up to ten loops!), but he has some ideas for how to generalize it, which will require handling divergences and numerators.

Other approaches to the “problem of integration” included Anna-Laura Sattelberger’s talk that presented a method to find differential equations for the kind of integrals that show up in amplitudes using the mathematical software Macaulay2, including presenting a package. Matthias Wilhelm talked about the work I did with him, using machine learning to find better methods for solving integrals with integration-by-parts, an area where two other groups have now also published. Pierpaolo Mastrolia talked about integration-by-parts’ up-and-coming contender, intersection theory, a method which appears to be delving into more mathematical tools in an effort to catch up with its competitor.

Sometimes, one is more specifically interested in the singularities of integrals than their numerics more generally. Felix Tellander talked about a geometric method to pin these down which largely went over my head, but he did have a very nice short description of the approach: “Describe the singularities of the integrand. Find a map representing integration. Map the singularities of the integrand onto the singularities of the integral.”

While QCD and gravity are the applications of choice, amplitudes methods germinate in N=4 super Yang-Mills. Ruth Britto’s talk opened the conference with an overview of progress along those lines before going into her own recent work with one-loop integrals and interesting implications of ideas from cluster algebras. Cluster algebras made appearances in several other talks, including Anastasia Volovich’s talk which discussed how ideas from that corner called flag cluster algebras may give insights into QCD amplitudes, though some symbol letters still seem to be hard to track down. Matteo Parisi covered another idea, cluster promotion maps, which he thinks may help pin down algebraic symbol letters.

The link between cluster algebras and symbol letters is an ongoing mystery where the field is seeing progress. Another symbol letter mystery is antipodal duality, where flipping an amplitude like a palindrome somehow gives another valid amplitude. Lance Dixon has made progress in understanding where this duality comes from, finding a toy model where it can be understood and proved.

Others pushed the boundaries of methods specific to N=4 super Yang-Mills, looking for novel structures. Song He’s talk pushes an older approach by Bourjaily and collaborators up to twelve loops, finding new patterns and connections to other theories and observables. Qinglin Yang bootstraps Wilson loops with a Lagrangian insertion, adding a side to the polygon used in previous efforts and finding that, much like when you add particles to amplitudes in a bootstrap, the method gets stricter and more powerful. Jaroslav Trnka talked about work he has been doing with “negative geometries”, an odd method descended from the amplituhedron that looks at amplitudes from a totally different perspective, probing a bit of their non-perturbative data. He’s finding more parts of that setup that can be accessed and re-summed, finding interestingly that multiple-zeta-values show up in quantities where we know they ultimately cancel out. Livia Ferro also talked about a descendant of the amplituhedron, this time for cosmology, getting differential equations for cosmological observables in a particular theory from a combinatorial approach.

Outside of everybody’s favorite theories, some speakers talked about more general approaches to understanding the differences between theories. Andreas Helset covered work on the geometry of the space of quantum fields in a theory, applying the method to a general framework for characterizing deviations from the standard model called the SMEFT. Jasper Roosmale Nepveu also talked about a general space of theories, thinking about how positivity (a trait linked to fundamental constraints like causality and unitarity) gets tangled up with loop effects, and the implications this has for renormalization.

Soft theorems, universal behavior of amplitudes when a particle has low energy, continue to be a trendy topic, with Silvia Nagy showing how the story continues to higher orders and Sangmin Choi investigating loop effects. Callum Jones talks about one of the more powerful results from the soft limit, Weinberg’s theorem showing the uniqueness of gravity. Weinberg’s proof was set up in Minkowski space, but we may ultimately live in curved, de Sitter space. Jones showed how the ideas Weinberg explored generalize in de Sitter, using some tools from the soft-theorem-inspired field of dS/CFT. Julio Parra-Martinez, meanwhile, tied soft theorems to another trendy topic, higher symmetries, a more general notion of the usual types of symmetries that physicists have explored in the past. Lucia Cordova reported work that was not particularly connected to soft theorems but was connected to these higher symmetries, showing how they interact with crossing symmetry and the S-matrix bootstrap.

Finally, a surprisingly large number of talks linked to Kevin Costello and Natalie Paquette’s work with self-dual gauge theories, where they found exact solutions from a fairly mathy angle. Paquette gave an update on her work on the topic, while Alfredo Guevara talked about applications to black holes, comparing the power of expanding around a self-dual gauge theory to that of working with supersymmetry. Atul Sharma looked at scattering in self-dual backgrounds in work that merges older twistor space ideas with the new approach, while Roland Bittelson talked about calculating around an instanton background.

Also, I had another piece up this week at FirstPrinciples, based on an interview with the (outgoing) president of the Sloan Foundation. I won’t have a “bonus info” post for this one, as most of what I learned went into the piece. But if you don’t know what the Sloan Foundation does, take a look! I hadn’t known they funded Jupyter notebooks and Hidden Figures, or that they introduced Kahneman and Tversky.

Publishing Isn’t Free, but SciPost Makes It Cheaper

5 Replies

I’ve mentioned SciPost a few times on this blog. They’re an open journal in every sense you could think of: diamond open-access scientific publishing on an open-source platform, run with open finances. They even publish their referee reports. They’re aiming to cover not just a few subjects, but a broad swath of academia, publishing scientists’ work in the most inexpensive and principled way possible and challenging the dominance of for-profit journals.

And they’re struggling.

SciPost doesn’t charge university libraries for access, they let anyone read their articles for free. And they don’t charge authors Article Processing Charges (or APCs), they let anyone publish for free. All they do is keep track of which institutions those authors are affiliated with, calculate what fraction of their total costs comes from them, and post it in a nice searchable list on their website.

And amazingly, for the last nine years, they’ve been making that work.

SciPost encourages institutions to pay their share, mostly by encouraging authors to bug their bosses until they do. SciPost will also quite happily accept more than an institution’s share, and a few generous institutions do just that, which is what has kept them afloat so far. But since nothing compels anyone to pay, most organizations simply don’t.

From an economist’s perspective, this is that most basic of problems, the free-rider problem. People want scientific publication to be free, but it isn’t. Someone has to pay, and if you don’t force someone to do it, then the few who pay will be exploited by the many who don’t.

There’s more worth saying, though.

First, it’s worth pointing out that SciPost isn’t paying the same cost everyone else pays to publish. SciPost has a stripped-down system, without any physical journals or much in-house copyediting, based entirely on their own open-source software. As a result, they pay about 500 euros per article. Compare this to the fees negotiated by particle physics’ SCOAP3 agreement, which average to closer to 1000 euros, and realize that those fees are on the low end: for-profit journals tend to make their APCs higher in order to, well, make a profit.

(By the way, while it’s tempting to think of for-profit journals as greedy, I think it’s better to think of them as not cost-effective. Profit is an expense, like the interest on a loan: a payment to investors in exchange for capital used to set up the business. The thing is, online journals don’t seem to need that kind of capital, especially when they’re based on code written by academics in their spare time. So they can operate more cheaply as nonprofits.)

So when an author publishes in SciPost instead of a journal with APCs, they’re saving someone money, typically their institution or their grant. This would happen even if their institution paid their share of SciPost’s costs. (But then they would pay something rather than nothing, hence free-rider problem.)

If an author instead would have published in a closed-access journal, the kind where you have to pay to read the articles and university libraries pay through the nose to get access? Then you don’t save any money at all, your library still has to pay for the journal. You only save money if everybody at the institution stops using the journal. This one is instead a collective action problem.

Collective action problems are hard, and don’t often have obvious solutions. Free-rider problems do suggest an obvious solution: why not just charge?

In SciPost’s case, there are philosophical commitments involved. Their desire to attribute costs transparently and equally means dividing a journal’s cost among all its authors’ institutions, a cost only fully determined at the end of the year, which doesn’t make for an easy invoice.

More to the point, though, charging to publish is directly against what the Open Access movement is about.

That takes some unpacking, because of course, someone does have to pay. It probably seems weird to argue that institutions shouldn’t have to pay charges to publish papers…instead, they should pay to publish papers.

SciPost itself doesn’t go into detail about this, but despite how weird it sounds when put like I just did, there is a difference. Charging a fee to publish means that anyone who publishes needs to pay a fee. If you’re working in a developing country on a shoestring budget, too bad, you have to pay the fee. If you’re an amateur mathematician who works in a truck stop and just puzzled through something amazing, too bad, you have to pay the fee.

Instead of charging a fee, SciPost asks for support. I have to think that part of the reason is that they want some free riders. There are some people who would absolutely not be able to participate in science without free riding, and we want their input nonetheless. That means to support them, others need to give more. It means organizations need to think about SciPost not as just another fee, but as a way they can support the scientific process as a whole.

That’s how other things work, like the arXiv. They get support from big universities and organizations and philanthropists, not from literally everyone. It seems a bit weird to do that for a single scientific journal among many, though, which I suspect is part of why institutions are reluctant to do it. But for a journal that can save money like SciPost, maybe it’s worth it.

AI Can’t Do Science…And Neither Can Other Humans

6 Replies

Seen on Twitter:

AI cannot in principle do science but it can be used as a tool to help process data. AI is not autonomous, & does not come up with its own problems to solve. That’s so obvious if you don’t understand you need to do some philosophy.
— Prof. Lee Cronin (@leecronin) March 9, 2025

I don’t know the context here, so I can’t speak to what Prof. Cronin meant. But it got me thinking.

Suppose you, like Prof. Cronin, were to insist that AI “cannot in principle” do science, because AI “is not autonomous” and “does not come up with its own problems to solve”. What might you mean?

You might just be saying that AI is bad at coming up with new problems to solve. That’s probably fair, at least at the moment. People have experimented with creating simple “AI researchers” that “study” computer programs, coming up with hypotheses about the programs’ performance and testing them. But it’s a long road from that to reproducing the much higher standards human scientists have to satisfy.

You probably don’t mean that, though. If you did, you wouldn’t have said “in principle”. You mean something stronger.

More likely, you might mean that AI cannot come up with its own problems, because AI is a tool. People come up with problems, and use AI to help solve them. In this perspective, not only is AI “not autonomous”, it cannot be autonomous.

On a practical level, this is clearly false. Yes, machine learning models, the core technology in current AI, are set up to answer questions. A user asks something, and receives the model’s prediction of the answer. That’s a tool, but for the more flexible models like GPT it’s trivial to turn it into something autonomous. Just add another program: a loop that asks the model what to do, does it, tells the model the result, and asks what to do next. Like taping a knife to a Roomba, you’ve made a very simple modification to make your technology much more dangerous.

You might object, though, that this simple modification of GPT is not really autonomous. After all, a human created it. That human had some goal, some problem they wanted to solve, and the AI is just solving the problem for them.

That may be a fair description of current AI, but insisting it’s true in principle has some awkward implications. If you make a “physics AI”, just tell it to do “good physics”, and it starts coming up with hypotheses you’d never thought of, is it really fair to say it’s just solving your problem?

What if the AI, instead, was a child? Picture a physicist encouraging a child to follow in their footsteps, filling their life with physics ideas and rhapsodizing about the hard problems of the field at the dinner table. Suppose the child becomes a physicist in turn, and finds success later in life. Were they really autonomous? Were they really a scientist?

What if the child, instead, was a scientific field, and the parent was the general public? The public votes for representatives, the representatives vote to hire agencies, and the agencies promise scientists they’ll give them money if they like the problems they come up with. Who is autonomous here?

(And what happens if someone takes a hammer to that process? I’m…still not talking about this! No-politics-rule still in effect, sorry! I do have a post planned, but it will have to wait until I can deal with the fallout.)

At this point, you’d probably stop insisting. You’d drop that “in principle”, and stick with the claim I started with, that current AI can’t be a scientist.

But you have another option.

You can accept the whole chain of awkward implications, bite all the proverbial bullets. Yes, you insist, AI is not autonomous. Neither is the physicist’s child in your story, and neither are the world’s scientists paid by government grants. Each is a tool, used by the one, true autonomous scientist: you.

You are stuck in your skull, a blob of curious matter trained on decades of experience in the world and pre-trained with a couple billion years of evolution. For whatever reason, you want to know more, so you come up with problems to solve. You’re probably pretty vague about those problems. You might want to see more pretty pictures of space, or wrap your head around the nature of time. So you turn the world into your tool. You vote and pay taxes, so your government funds science. You subscribe to magazines and newspapers, so you hear about it. You press out against the world, and along with the pressure that already exists it adds up, and causes change. Biological intelligences and artificial intelligences scurry at your command. From their perspective, they are proposing their own problems, much more detailed and complex than the problems you want to solve. But from yours, they’re your limbs beyond limbs, sight beyond sight, asking the fundamental questions you want answered.

Physics Gets Easier, Then Harder

Ways Freelance Journalism Is Different From Academic Writing

2 Replies

A while back, I was surprised when I saw the writer of a well-researched webcomic assume that academics are paid for their articles. I ended up writing a post explaining how academic publishing actually works.

Now that I’m out of academia, I’m noticing some confusion on the other side. I’m doing freelance journalism, and the academics I talk to tend to have some common misunderstandings. So academics, this post is for you: a FAQ of questions I’ve been asked about freelance journalism. Freelance journalism is more varied than academia, and I’ve only been doing it a little while, so all of my answers will be limited to my experience.

Q: What happens first? Do they ask you to write something? Do you write an article and send it to them?

Academics are used to writing an article, then sending it to a journal, which sends it out to reviewers to decide whether to accept it. In freelance journalism in my experience, you almost never write an article before it’s accepted. (I can think of one exception I’ve run into, and that was for an opinion piece.)

Sometimes, an editor reaches out to a freelancer and asks them to take on an assignment to write a particular sort of article. This happens more freelancers that have been working with particular editors for a long time. I’m new to this, so the majority of the time I have to “pitch”. That means I email an editor describing the kind of piece I want to write. I give a short description of the topic and why it’s interesting. If the editor is interested, they’ll ask some follow-up questions, then tell me what they want me to focus on, how long the piece should be, and how much they’ll pay me. (The last two are related, many places pay by the word.) After that, I can write a draft.

Q: Wait, you’re paid by the word? Then why not make your articles super long, like Victor Hugo?

I’m paid per word assigned, not per word in the finished piece. The piece doesn’t have to strictly stick to the word limit, but it should be roughly the right size, and I work with the editor to try to get it there. In practice, places seem to have a few standard size ranges and internal terminology for what they are (“blog”, “essay”, “short news”, “feature”). These aren’t always the same as the categories readers see online. Some places have a web page listing these categories for prospective freelancers, but many don’t, so you have to either infer them from the lengths of articles online or learn them over time from the editors.

Q: Why didn’t you mention this important person or idea?

Because pieces pay more by the word, it’s easier as a freelancer to sell shorter pieces than longer ones. For science news, favoring shorter pieces also makes some pedagogical sense. People usually take away only a few key messages from a piece, if you try to pack in too much you run a serious risk of losing people. After I’ve submitted a draft, I work with the editor to polish it, and usually that means cutting off side-stories and “by-the-ways” to make the key points as vivid as possible.

Q: Do you do those cool illustrations?

Academia has a big focus on individual merit. The expectation is that when you write something, you do almost all of the work yourself, to the extent that more programming-heavy fields like physics and math do their own typesetting.

Industry, including journalism, is more comfortable delegating. Places will generally have someone on-staff to handle illustrations. I suggest diagrams that could be helpful to the piece and do a sketch of what they could look like, but it’s someone else’s job to turn that into nice readable graphic design.

Q: Why is the title like that? Why doesn’t that sound like you?

Editors in journalistic outlets are much more involved than in academic journals. Editors won’t just suggest edits, they’ll change wording directly and even input full sentences of their own. The title and subtitle of a piece in particular can change a lot (in part because they impact SEO), and in some places these can be changed by the editor quite late in the process. I’ve had a few pieces whose title changed after I’d signed off on them, or even after they first appeared.

Q: Are your pieces peer-reviewed?

The news doesn’t have peer review, no. Some places, like Quanta Magazine, do fact-checking. Quanta pays independent fact-checkers for longer pieces, while for shorter pieces it’s the writer’s job to verify key facts, confirming dates and the accuracy of quotes.

Q: Can you show me the piece before it’s published, so I can check it?

That’s almost never an option. Journalists tend to have strict rules about showing a piece before it’s published, related to more political areas where they want to preserve the ability to surprise wrongdoers and the independence to find their own opinions. Science news seems like it shouldn’t require this kind of thing as much, it’s not like we normally write hit pieces. But we’re not publicists either.

In a few cases, I’ve had people who were worried about something being conveyed incorrectly, or misleadingly. For those, I offer to do more in the fact-checking stage. I can sometimes show you quotes or paraphrase how I’m describing something, to check whether I’m getting something wrong. But under no circumstances can I show you the full text.

Q: What can I do to make it more likely I’ll get quoted?

Pieces are short, and written for a general, if educated, audience. Long quotes are harder to use because they eat into word count, and quotes with technical terms are harder to use because we try to limit the number of terms we ask the reader to remember. Quotes that mention a lot of concepts can be harder to find a place for, too: concepts are introduced gradually over the piece, so a quote that mentions almost everything that comes up will only make sense to the reader at the very end.

In a science news piece, quotes can serve a couple different roles. They can give authority, an expert’s judgement confirming that something is important or real. They can convey excitement, letting the reader see a scientist’s emotions. And sometimes, they can give an explanation. This last only happens when the explanation is very efficient and clear. If the journalist can give a better explanation, they’re likely to use that instead.

So if you want to be quoted, keep that in mind. Try to say things that are short and don’t use a lot of technical jargon or bring in too many concepts at once. Convey judgement, which things are important and why, and convey passion, what drives you and excited you about a topic. I am allowed to edit quotes down, so I can take a piece of a longer quote that’s cleaner or cut a long list of examples from an otherwise compelling statement. I can correct grammar and get rid of filler words and obvious mistakes. But I can’t put words in your mouth, I have to work with what you actually said, and if you don’t say anything I can use then you won’t get quoted.

Government Science Funding Isn’t a Precision Tool

Freelancing in [Country That Includes Greenland]

4 Replies

(Why mention Greenland? It’s a movie reference.)

I figured I’d give an update on my personal life.

A year ago, I resigned from my position in France and moved back to Denmark. I had planned to spend a few months as a visiting researcher in my old haunts at the Niels Bohr Institute, courtesy of the spare funding of a generous friend. There turned out to be more funding than expected, and what was planned as just a few months was extended to almost a year.

I spent that year learning something new. It was still an amplitudes project, trying to make particle physics predictions more efficient. But this time I used Python. I looked into reinforcement learning and PyTorch, played with using a locally hosted Large Language Model to generate random code, and ended up getting good results from a classic genetic programming approach. Along the way I set up a SQL database, configured Docker containers, and puzzled out interactions with CUDA. I’ve got a paper in the works, I’ll post about it when it’s out.

All the while, on the side, I’ve been seeking out stories. I’ve not just been a writer, but a journalist, tracking down leads and interviewing experts. I had three pieces in Quanta Magazine and one in Ars Technica.

Based on that, I know I can make money doing science journalism. What I don’t know yet is whether I can make a living doing it. This year, I’ll figure that out. With the project at the Niels Bohr Institute over, I’ll have more time to seek out leads and pitch to more outlets. I’ll see whether I can turn a skill into a career.

So if you’re a scientist with a story to tell, if you’ve discovered something or accomplished something or just know something that the public doesn’t, and that you want to share: do reach out. There’s a lot that can be of interest, passion that can be shared.

At the same time, I don’t know yet whether I can make a living as a freelancer. Many people try and don’t succeed. So I’m keeping my CV polished and my eyes open. I have more experience now with Data Science tools, and I’ve got a few side projects cooking that should give me a bit more. I have a few directions in mind, but ultimately, I’m flexible. I like being part of a team, and with enthusiastic and competent colleagues I can get excited about pretty much anything. So if you’re hiring in Copenhagen, if you’re open to someone with ten years of STEM experience who’s just starting to see what industry has to offer, then let’s chat. Even if we’re not a good fit, I bet you’ve got a good story to tell.

Newtonmas and the Gift of a Physics Background

3 Replies

This week, people all over the world celebrated the birth of someone whose universally attractive ideas spread around the globe. I’m talking, of course about Isaac Newton.

For Newtonmas this year, I’ve been pondering another aspect of Newton’s life. There’s a story you might have heard that physicists can do basically anything, with many people going from a career in physics to a job in a variety of other industries. It’s something I’ve been trying to make happen for myself. In a sense, this story goes back to the very beginning, when Newton quit his academic job to work at the Royal Mint.

On the surface, there are a lot of parallels. At the Mint, a big part of Newton’s job was to combat counterfeiting and “clipping”, where people would carve small bits of silver off of coins. This is absolutely a type of job ex-physicists do today, at least in broad strokes. Working as Data Scientists for financial institutions, people look for patterns in transactions that give evidence of fraud.

Digging deeper, though, the analogy falls apart a bit. Newton didn’t apply any cunning statistical techniques to hunt down counterfeiters. Instead, the stories that get told about his work there are basically detective stories. He hung out in bars to catch counterfeiter gossip and interviewed counterfeiters in prison, not exactly the kind of thing you’d hire a physicist to do these days. The rest of the role was administrative: setting up new mint locations and getting people to work overtime to replace the country’s currency. Newton’s role at the mint was less like an ex-physicist going into Data Science and more like Steven Chu as Secretary of Energy: someone with a prestigious academic career appointed to a prestigious government role.

If you’re looking for a patron saint of physicists who went to industry, Newton’s contemporary Robert Hooke may be a better bet. Unlike many other scientists of the era, Hooke wasn’t independently wealthy, and for a while he was kept quite busy working for the Royal Society. But a bit later he had another, larger source of income: working as a surveyor and architect, where he designed several of London’s iconic buildings. While Newton’s work at the Mint drew on his experience as a person of power and influence, working as an architect drew much more on skills directly linked to Hooke’s work as a scientist: understanding the interplay of forces in quantitative detail.

While Newton and Hooke’s time was an era of polymaths, in some sense the breadth of skills imparted by a physics education has grown. Physicists learn statistics (which barely existed in Newton’s time) programming (which did not exist at all) and a wider range of mathematical and physical models. Having a physics background isn’t the ideal way to go into industry (that would be having an industry background). But for those of us making the jump, it’s still a Newtonmas gift to be grateful for.

Which String Theorists Are You Complaining About?

4 Replies

Do string theorists have an unfair advantage? Do they have an easier time getting hired, for example?

In one of the perennial arguments about this on Twitter, Martin Bauer posted a bar chart of faculty hires in the US by sub-field. The chart was compiled by Erich Poppitz from data in the US particle physics rumor mill, a website where people post information about who gets hired where for the US’s quite small number of permanent theoretical particle physics positions at research universities and national labs. The data covers 1994 to 2017, and shows one year, 1999, when there were more string theorists hired than all other topics put together. The years around then also had many string theorists hired, but the proportion starts falling around the mid 2000’s…around when Lee Smolin wrote a book, The Trouble With Physics, arguing that string theorists had strong-armed their way into academic dominance. After that, the percentage of string theorists falls, oscillating between a tenth and a quarter of total hires.

Judging from that, you get the feeling that string theory’s critics are treating a temporary hiring fad as if it was a permanent fact. The late 1990’s were a time of high-profile developments in string theory that excited a lot of people. Later, other hiring fads dominated, often driven by experiments: I remember when the US decided to prioritize neutrino experiments and neutrino theorists had a much easier time getting hired, and there seem to be similar pushes now with gravitational waves, quantum computing, and AI.

Thinking about the situation in this way, though, ignores what many of the critics have in mind. That’s because the “string” column on that bar chart is not necessarily what people think of when they think of string theory.

If you look at the categories on Poppitz’s bar chart, you’ll notice something odd. “String” its itself a category. Another category, “lattice”, refers to lattice QCD, a method to find the dynamics of quarks numerically. The third category, though, is a combination of three things “ph/th/cosm”.

“Cosm” here refers to cosmology, another sub-field. “Ph” and “th” though aren’t really sub-fields. Instead, they’re arXiv categories, sections of the website arXiv.org where physicists post papers before they submit them to journals. The “ph” category is used for phenomenology, the type of theoretical physics where people try to propose models of the real world and make testable predictions. The “th” category is for “formal theory”, papers where theoretical physicists study the kinds of theories they use in more generality and develop new calculation methods, with insights that over time filter into “ph” work.

“String”, on the other hand, is not an arXiv category. When string theorists write papers, they’ll put them into “th” or “ph” or another relevant category (for example “gr-qc”, for general relativity and quantum cosmology). This means that when Poppitz distinguishes “ph/th/cosm” from “string”, he’s being subjective, using his own judgement to decide who counts as a string theorist.

So who counts as a string theorist? The simplest thing to do would be to check if their work uses strings. Failing that, they could use other tools of string theory and its close relatives, like Calabi-Yau manifolds, M-branes, and holography.

That might be what Poppitz was doing, but if he was, he was probably missing a lot of the people critics of string theory complain about. He even misses many people who describe themselves as string theorists. In an old post of mine I go through the talks at Strings, string theory’s big yearly conference, giving them finer-grained categories. The majority don’t use anything uniquely stringy.

Instead, I think critics of string theory have two kinds of things in mind.

First, most of the people who made their reputations on string theory are still in academia, and still widely respected. Some of them still work on string theory topics, but many now work on other things. Because they’re still widely respected, their interests have a substantial influence on the field. When one of them starts looking at connections between theories of two-dimensional materials, you get a whole afternoon of talks at Strings about theories of two-dimensional materials. Working on those topics probably makes it a bit easier to get a job, but also, many of the people working on them are students of these highly respected people, who just because of that have an easier time getting a job. If you’re a critic of string theory who thinks the founders of the field led physics astray, then you probably think they’re still leading physics astray even if they aren’t currently working on string theory.

Second, for many other people in physics, string theorists are their colleagues and friends. They’ll make fun of trends that seem overhyped and under-thought, like research on the black hole information paradox or the swampland, or hopes that a slightly tweaked version of supersymmetry will show up soon at the LHC. But they’ll happily use ideas developed in string theory when they prove handy, using supersymmetric theories to test new calculation techniques, string theory’s extra dimensions to inspire and ground new ideas for dark matter, or the math of strings themselves as interesting shortcuts to particle physics calculations. String theory is available as reference to these people in a way that other quantum gravity proposals aren’t. That’s partly due to familiarity and shared language (I remember a talk at Perimeter where string theorists wanted to learn from practitioners from another area and the discussion got bogged down by how they were using the word “dimension”), but partly due to skepticism of the various alternate approaches. Most people have some idea in their heads of deep problems with various proposals: screwing up relativity, making nonsense out of quantum mechanics, or over-interpreting on limited evidence. The most commonly believed criticisms are usually wrong, with objections long-known to practitioners of the alternate approaches, and so those people tend to think they’re being treated unfairly. But the wrong criticisms are often simplified versions of correct criticisms, passed down by the few people who dig deeply into these topics, criticisms that the alternative approaches don’t have good answers to.

The end result is that while string theory itself isn’t dominant, a sort of “string friendliness” is. Most of the jobs aren’t going to string theorists in the literal sense. But the academic world string theorists created keeps turning. People still respect string theorists and the research directions they find interesting, and people are still happy to collaborate and discuss with string theorists. For research communities people are more skeptical of, it must feel very isolating, like the world is still being run by their opponents. But this isn’t the kind of hegemony that can be solved by a revolution. Thinking that string theory is a failed research program, and people focused on it should have a harder time getting hired, is one thing. Thinking that everyone who respects at least one former string theorist should have a harder time getting hired is a very different goal. And if what you’re complaining about is “string friendliness”, not actual string theorists, then that’s what you’re asking for.

4 gravitons

Stories about physics from someone who's been there