Monthly Archives: December 2025

For Newtonmas, One Seventeenth of a New Collider

Leave a reply

Individual physicists don’t ask for a lot for Newtonmas. Big collaborations ask for more.

This year, CERN got its Newtonmas gift early: a one billion dollar pledge from a group of philanthropists and foundations, to be spent on their proposed new particle collider.

That may sound like a lot of money (and of course it is), but it’s only a fraction of the 15 billion euros that the collider is estimated to cost. That makes this less a case of private donors saving the project, and more of a nudge, showing governments they can get results for a bit cheaper than they expected.

I do wonder if the donation has also made CERN more bold about their plans, since it was announced shortly after a report from the update process for the European Strategy for Particle Physics, in which the European Strategy Group recommended a backup plan for the collider that is just the same collider with 15% budget cuts. Naturally people started making fun of this immediately.

Credit to @theory_dad on X

There were more serious objections from groups that had proposed more specific backup plans earlier in the process, who are frustrated that their ideas were rejected in favor of a 15% tweak that was not even discussed and seems not to really have been evaluated.

I don’t have any special information about what’s going on behind the scenes, or where this is headed. But I’m amused, and having fun with the parallels this season. I remember writing lists as a kid, trying to take advantage of the once-a-year opportunity to get what seemed almost like a genie’s wish. Whatever my incantations, the unreasonable requests were never fulfilled. Still, I had enough new toys to fill my time, and whet my appetite for the next year.

We’ll see what CERN’s Newtonmas gift brings.

Academia Tracks Priority, Not Provenance

3 Replies

A recent Correspondence piece in Nature Machine Intelligence points at an issue with using LLMs to write journal articles. LLMs are trained on enormous amounts of scholarly output, but the result is quite opaque: it is usually impossible to tell which sources influence a specific LLM-written text. That means that when a scholar uses an LLM, they may get a result that depends on another scholar’s work, without realizing it or documenting it. The ideas’ provenance gets lost, and the piece argues this is damaging, depriving scholars of credit and setting back progress.

It’s a good point. Provenance matters. If we want to prioritize funding for scholars whose ideas have the most impact, we need a way to track where ideas arise.

However, current publishing norms make essentially no effort to do this. Academic citations are not used to track provenance, and they are not typically thought of as tracking provenance. Academic citations track priority.

Priority is a central value in scholarship, with a long history. We give special respect to the first person to come up with an idea, make an observation, or do a calculation, and more specifically, the first person to formally publish it. We do this even if the person’s influence was limited, and even if the idea was rediscovered independently later on. In an academic context, being first matters.

In a paper, one is thus expected to cite the sources that have priority, that came up with an idea first. Someone who fails to do so will get citation request emails, and reviewers may request revisions to the paper to add in those missing citations.

One may also cite papers that were helpful, even if they didn’t come first. Tracking provenance in this way can be nice, a way to give direct credit to those who helped and point people to useful resources. But it isn’t mandatory in the same way. If you leave out a secondary source and your paper doesn’t use anything original to that source (like new notation), you’re much less likely to get citation request emails, or revision requests from reviewers. Provenance is just much lower priority.

In practice, academics track provenance in much less formal ways. Before citations, a paper will typically have an Acknowledgements section, where the authors thank those who made the paper possible. This includes formal thanks to funding agencies, but also informal thanks for “helpful discussions” that don’t meet the threshold of authorship.

If we cared about tracking provenance, those acknowledgements would be crucial information, an account of whose ideas directly influenced the ideas in the paper. But they’re not treated that way. No-one lists the number of times they’ve been thanked for helpful discussions on their CV, or in a grant application, no-one considers these discussions for hiring or promotion. You can’t look them up on an academic profile or easily graph them in a metascience paper. Unlike citations, unlike priority, there is essentially no attempt to measure these tracks of provenance in any organized way.

Instead, provenance is often the realm of historians or history-minded scholars, writing long after the fact. For academics, the fact that Yang and Mills published their theory first is enough, we call it Yang-Mills theory. For those studying the history, the story is murkier: it looks like Pauli came up with the idea first, and did most of the key calculations, but didn’t publish when it looked to him like the theory couldn’t describe the real world. What’s more, there is evidence suggesting that Yang knew about Pauli’s result, that he had read a letter from him on the topic, that the idea’s provenance goes back to Pauli. But Yang published, Pauli didn’t. And in the way academia has worked over the last 75 years, that claim of priority is what actually mattered.

Should we try to track provenance? Maybe. Maybe the emerging ubiquitousness of LLMs should be a wakeup call, a demand to improve our tracking of ideas, both in artificial and human neural networks. Maybe we need to demand interpretability from our research tools, to insist that we can track every conclusion back to its evidence for every method we employ, to set a civilizational technological priority on the accurate valuation of information.

What we shouldn’t do, though, is pretend that we just need to go back to what we were doing before.

Energy Is That Which Is Conserved

2 Replies

In school, kids learn about different types of energy. They learn about solar energy and wind energy, nuclear energy and chemical energy, electrical energy and mechanical energy, and potential energy and kinetic energy. They learn that energy is conserved, that it can never be created or destroyed, but only change form. They learn that energy makes things happen, that you can use energy to do work, that energy is different from matter.

Some, between good teaching and good students, manage to impose order on the jumble of concepts and terms. Others end up envisioning the whole story a bit like Pokemon, with different types of some shared “stuff”.

Energy isn’t “stuff”, though. So what is it? What relates all these different types of things?

Energy is something which is conserved.

The mathematician Emmy Noether showed that, when the laws of physics are symmetrical, they come with a conserved quantity. For example, because the laws of the physics are the same from place to place, momentum is conserved. Similarly, because the laws of physics are the same from one time to another, Noether’s theorem states that there must be some quantity related to time, some number we can calculate, that is conserved, even as other things change. We call that number energy.

If energy is that simple, why are there all those types?

Energy is a number we can calculate. It’s a number we can calculate for different things. If you have a detailed description of how something in physics works, you can use that description to calculate that thing’s energy. In school, you memorize formulas like \frac{1}{2}m v^2 and m g h. These are all formulas that, with a bit more knowledge, you could calculate. They are the things that, for a something that meets the conditions, are conserved. They are things that, according to Noether’s theorem, stay the same.

Because of this, you shouldn’t think of energy as a substance, or a fuel. Energy is something we can do: we physicists, and we students of physics. We can take a physical system, and see what about it ought to be conserved. Energy is an action, a calculation, a conceptual tool that can be used to make predictions.

Most things are, in the end.

Ideally, Exams Are for the Students

3 Replies

I should preface this by saying I don’t actually know that much about education. I taught a bit in my previous life as a professor, yes, but I probably spent more time being taught how to teach than actually teaching.

Recently, the Atlantic had a piece about testing accommodations for university students, like extra time on exams, or getting to do an exam in a special distraction-free environment. The piece quotes university employees who are having more and more trouble satisfying these accommodations, and includes the statistic that 20 percent of undergraduate students at Brown and Harvard are registered as disabled.

The piece has kicked off a firestorm on social media, mostly focused on that statistic (which conveniently appears just before the piece’s paywall). People are shocked, and cynical. They feel like more and more students are cheating the system, getting accommodations that they don’t actually deserve.

I feel like there is a missing mood in these discussions, that the social media furor is approaching this from the wrong perspective. People are forgetting what exams actually ought to be for.

Exams are for the students.

Exams are measurement tools. An exam for a class says whether a student has learned the material, or whether they haven’t, and need to retake the class or do more work to get there. An entrance exam, or a standardized exam like the SAT, predicts a student’s future success: whether they will be able to benefit from the material at a university, or whether they don’t yet have the background for that particular program of study.

These are all pieces of information that are most important to the students themselves, that help them structure their decisions. If you want to learn the material, should you take the course again? Which universities are you prepared for, and which not?

We have accommodations, and concepts like disability, because we believe that there are kinds of students for whom the exams don’t give this information accurately. We think that a student with more time, or who can take the exam in a distraction-free environment, would have a more accurate idea of whether they need to retake the material, or whether they’re ready for a course of study, than a student who has to take the exam under ordinary conditions. And we think we can identify the students who this matters for, and the students for whom this doesn’t matter nearly as much.

These aren’t claims about our values, or about what students deserve. They’re empirical claims, about how test results correlate with outcomes the students want. The conversation, then, needs to be built on top of those empirical claims. Are we better at predicting the success of students that receive accommodations, or worse? Can we measure that at all, or are we just guessing? And are we communicating the consequences accurately to students, that exam results tell them something useful and statistically robust that should help them plan their lives?

Values come in later, of course. We don’t have infinite resources, as the Atlantic piece emphasizes. We can’t measure everyone with as much precision as we would like. At some level, generalization takes over and accuracy is lost. There is absolutely a debate to be had about which measurements we can afford to make, and which we can’t.

But in order to have that argument at all, we first need to agree on what we’re measuring. And I feel like most of the people talking about this piece haven’t gotten there yet.