Daniel sent us this one — and it's deceptively simple. He's asking what theoretical physics actually is, and what physicists do on a daily basis. Not the Hollywood version, not the chalkboard genius scribbling equations in a flash of insight — the real thing. And I think there's a deeper question here too: is theoretical physics still producing breakthroughs, or has it stalled out in realms nobody can test?
Oh, this is my kind of question. And you know what — most people have this image that's about twenty percent right and eighty percent wildly wrong. The chalkboard part is real, actually. But the flash-of-insight thing? That's maybe five minutes a year if you're lucky. The rest is something else entirely.
Before we dive in — quick note. Today's script is being written by DeepSeek V four Pro. So if the jokes land, credit the model. If they don't, blame the model.
Alright, so let's start with what theoretical physics actually is, because the definition matters. Theoretical physics is the branch of physics that uses mathematical models and abstractions to explain, predict, and understand physical phenomena. It's distinct from experimental physics, which tests those models in the lab. But the line is blurrier than people think.
Take someone like Enrico Fermi. He's famous for the nuclear reactor, very hands-on, but he was also a brilliant theorist who could estimate the yield of the Trinity test by dropping scraps of paper and watching how far they drifted in the shockwave. The best theorists often have a deep intuition for experiment, and the best experimentalists understand the theory they're testing. The caricature of the theorist who can't change a lightbulb — it's not entirely false, but it's not the whole picture either.
I've always wondered about that lightbulb thing. Is there an actual documented case?
There's a famous story about a physicist — I think it was attributed to several people — who called maintenance because a lightbulb burned out, and when asked if he'd tried changing it, said something like, "I'm a theoretical physicist. I don't do experiments." Probably apocryphal, but it captures something real about the division of labor.
Alright, so let's get to what Daniel's really asking. The daily life. What does a theoretical physicist actually do between coffee and dinner?
I dug into this, and the answer is surprisingly consistent across institutions. A typical day involves three or four big buckets. First, and this is the one that surprises people — reading. A lot of reading. Hours of it. You're keeping up with preprints on arXiv, which is this online repository where physicists post papers before they're peer-reviewed. There are something like fifteen hundred new papers posted to the physics sections of arXiv every single day. You can't read them all, but you have to scan constantly to know what's happening in your subfield.
Fifteen hundred a day? That sounds like a firehose.
It is, and that's just physics. So most theorists develop these elaborate filtering systems — they follow specific authors, specific keyword alerts, they skim abstracts and only dive into maybe three or four papers a day. But the reading never stops. Science moves fast.
Bucket one is reading. What's bucket two?
Bucket two is the actual work of theory — and this is where the chalkboard comes in. Most theoretical physicists spend a huge portion of their day doing what they call "playing." That's the word they use. It means working through calculations, trying different mathematical approaches, seeing if something interesting falls out. It's not always directed. Sometimes you're just exploring the mathematical landscape of a problem, looking for patterns or symmetries that might lead somewhere.
"Playing" is doing a lot of heavy lifting there. What does that actually look like?
It can be solo work at a whiteboard or with pen and paper — and yes, many still use chalkboards because there's something about the physical act of writing that helps with thinking. It can also be computational — running simulations, writing code to solve equations numerically when analytic solutions don't exist. A growing number of theorists spend more time at a keyboard than at a chalkboard. And then there's the collaborative version, which is two or three people at a blackboard together, arguing and sketching and erasing each other's work.
That's the third bucket then — collaboration?
And it's a bigger bucket than most outsiders realize. Physics is intensely collaborative now. The days of the lone genius are mostly over. Most papers have multiple authors. You're constantly talking to colleagues, attending seminars, visiting other institutions, doing video calls with collaborators in other time zones. The stereotype of the antisocial physicist hiding in an office is outdated. The field selects for people who can communicate and work in teams now.
Which makes sense. If the problems are hard enough that nobody can solve them alone, you need networks.
And then the fourth bucket is the administrative and teaching stuff that comes with any academic job — writing grant proposals, reviewing papers for journals, mentoring graduate students, teaching courses, sitting on committees. A senior theorist might spend forty percent of their time on things that aren't physics at all.
That's a lot of non-physics for someone trying to understand the universe.
It's the tradeoff. And this is where a lot of public perception gets it wrong. People imagine the physicist in a state of constant deep thought, but the reality is fragmented. You might get two hours of real concentrated work in a day if you're lucky. The rest is meetings and email and the mundane logistics of keeping a research program alive.
Let's talk about the actual substance. What are theoretical physicists working on right now? What are the big open questions?
There's a useful way to slice this. Theoretical physics operates on a spectrum from the extremely concrete to the extremely speculative. On the concrete end, you have condensed matter theory — people studying superconductors, quantum materials, topological phases of matter. These are things that connect directly to experiments and sometimes to technology. On the speculative end, you have quantum gravity, string theory, cosmology, the search for a theory of everything.
Where's the energy right now? Where are the interesting results?
That depends on who you ask, but I'd point to a few areas. Quantum information theory is exploding — the intersection of quantum mechanics and information science. Condensed matter has been producing Nobel-level work regularly. But the thing that's been making headlines — and causing controversy — is the foundational stuff. There was a piece in Quanta Magazine just a few weeks ago, late March, asking whether string theory is still our best hope for a theory of everything. The answer from the community is deeply divided.
String theory emerged in the nineteen eighties as this elegant idea — that all particles and forces arise from tiny vibrating strings in higher-dimensional space. It was supposed to unify quantum mechanics and general relativity. It's mathematically beautiful. But after forty years, it hasn't made a single testable prediction. No experiment has confirmed it. And a growing number of physicists are saying — maybe this isn't the right path.
That's a pretty brutal critique. Forty years and no experimental confirmation?
Here's where it gets really interesting. The defenders of string theory make a counterargument that's worth taking seriously. They say: the problem isn't the theory, it's that we don't have the tools to test it. The energy scales where string theory's predictions become visible are so enormous — the Planck scale — that no particle accelerator we can build will ever reach them. It's not that string theory is wrong. It's that it's untestable with current technology, and maybe with any technology.
Is that science anymore? If something is untestable in principle, doesn't that push it into philosophy?
That's exactly the debate. And it's not just string theory. A lot of cosmology is brushing up against the same problem. The multiverse, the idea that our universe is one of many — that's an idea that emerges naturally from some versions of inflationary cosmology and string theory. But how do you test for other universes? You can't. They're causally disconnected by definition.
What do the critics propose instead?
There's a growing movement toward what some call "post-empirical" physics or, more charitably, a rethinking of what counts as evidence. But the more traditional response is to focus on problems that are hard but testable. People like Carlo Rovelli — and Quanta did a big profile on him back in October — argue for a radical reframing. Rovelli's big idea is that we should stop looking for a theory of everything and instead focus on understanding what he calls "relative information" — the idea that the properties of physical systems only exist in relation to other systems. There's no absolute state of the universe.
Rovelli's the loop quantum gravity guy, right?
Yes, and he's a fascinating figure because he's both a serious theoretical physicist and someone who writes beautifully for the public. His view is that the obsession with unification and final theories is almost theological. Physics, he argues, has always progressed by giving up on absolute notions — absolute space, absolute time, absolute simultaneity — and embracing relational ones. Why should a final theory be any different?
I can see why that appeals to some people and annoys others.
It annoys a lot of string theorists, definitely. But it's a productive tension. The field needs both the ambitious unifiers and the skeptical relationalists. That's how science moves — not by consensus, but by productive disagreement.
Alright, let's come back to the daily life question for a second. You mentioned reading, playing with math, collaborating, admin. But what about the emotional texture of it? Is it satisfying?
This is something people don't talk about enough. Being a theoretical physicist is psychologically demanding in ways that aren't obvious. You spend most of your time being wrong. Most of your ideas don't work. Most calculations lead to dead ends. You can spend six months on an approach and realize it was fundamentally flawed. The success rate — if you define success as a publishable result — might be a few percent of the things you try.
That sounds brutal.
It is, and it selects for a certain personality type. You have to be comfortable with failure, with uncertainty, with the feeling of not knowing whether you're making progress or just spinning your wheels. The people who thrive tend to be stubborn in a very specific way — not stubborn about being right, but stubborn about continuing to try.
There's a difference between stubbornness about outcome and stubbornness about process.
And the other thing is — the highs are very high. When something works, when a calculation clicks and you see a pattern nobody has seen before, it's apparently euphoric. People describe it as addictive. You're chasing that feeling through months of frustration.
It's like gambling but with partial differential equations.
not a terrible analogy, actually. The intermittent reinforcement schedule is similar. You never know when the next payoff is coming, but when it comes, it's intense.
Let's talk about the career structure. Is this a viable path? If someone's listening and thinking, maybe I want to be a theoretical physicist — what are they actually signing up for?
The honest answer is that it's a brutal pipeline. You do an undergraduate degree in physics or math, then a PhD which typically takes five to seven years. Then you do postdoctoral positions — usually two or three of them, each two to three years, often in different cities or countries. You might be in your mid-thirties before you have a permanent job. And permanent jobs are scarce. The number of PhDs produced each year far exceeds the number of faculty positions that open up.
Most people don't make it.
Most people end up doing something else. Some go into industry — finance, data science, tech companies love physics PhDs because they're good at quantitative modeling. Some go into national labs or government research. Some leave research entirely. The people who get tenure-track positions at research universities are a small fraction of the people who start PhDs.
That's a lot of human capital that doesn't end up doing what it was trained for.
That's a real loss, but it's also not entirely wasted. The skills transfer. Physicists who go into finance or tech often do extremely well. They've been trained to solve hard, ill-defined problems with mathematical tools. That's valuable everywhere.
Still, it's a system that seems designed to break a lot of hearts.
It is, and there's growing awareness of this. Some departments are being more transparent with prospective students about career outcomes. But the basic structure hasn't changed much. It's an apprenticeship model in an era when the number of master positions isn't growing.
Let's pivot slightly. You mentioned computational work. How much of theoretical physics is now done on computers rather than with pencil and paper?
It depends enormously on the subfield. In something like lattice quantum chromodynamics — which is the study of the strong nuclear force using numerical methods — it's almost entirely computational. These groups run massive simulations on supercomputers, and the "theory" is as much about algorithm design as it is about physics. In other areas, like pure mathematical physics, computers might not be involved at all. Most theorists now use computers for at least some part of their work — literature searches, typesetting papers, making plots, running small numerical experiments to test analytic ideas.
Has that changed the nature of the work qualitatively, or is it just a faster version of what people always did?
I think it has changed it qualitatively, and here's why. When you can numerically explore a system, you can develop intuition in a way that wasn't possible before. You can ask "what if" questions and get an answer in minutes rather than weeks of analytic work. This means theorists can be more exploratory. But there's also a risk — you can generate results that you don't deeply understand. The computation gives you an answer, but not an explanation.
That's a tension in a lot of fields now. The black box problem.
And physicists are acutely aware of it. The goal is understanding, not just prediction. If a simulation tells you a material will superconduct at a certain temperature, but you don't know why, that's useful but unsatisfying. The real work is extracting the physical mechanism.
What does a genuinely good day look like for a theoretical physicist? Paint me a picture.
You come in, you've been stuck on a problem for three weeks. You've been trying a particular mathematical approach and it keeps giving nonsense. While you're reading a paper over coffee — something only tangentially related — you see a technique you've never seen before. You think, huh, what if I try that? You go to your whiteboard, you sketch it out, and suddenly things start canceling. Terms that were problematic drop out. The math gets simpler instead of more complicated. By lunchtime you have a result that's clean, elegant, and makes a prediction you can test against known data. You check the data and it matches. You've found something new.
How often does that happen?
Maybe twice a year if you're lucky. Maybe never in a given year. But when it does, it carries you through all the other days.
It's almost a spiritual practice at that point. The patience, the faith that the work matters even when you can't see progress.
There's actually a philosophical tradition around this. The physicist John Archibald Wheeler said something like — and I'm paraphrasing — that the job of the theoretical physicist is to make mistakes as fast as possible. The idea being that you're exploring a space of possibilities, and most of them are wrong, so your job is to eliminate wrong paths efficiently. Every dead end is progress because it narrows the search.
That's a very optimistic reframe of failure.
It's essential if you want to stay sane. You have to believe that the process is cumulative, even when individual projects don't pan out.
Let's talk about the relationship between theory and experiment. You said it's blurry, but there must be cases where theory leads experiment by decades.
The classic example is the Higgs boson. Predicted in nineteen sixty-four by Peter Higgs and others — a theoretical construct needed to make the Standard Model work. Not detected until twenty twelve, almost fifty years later, at the Large Hadron Collider. That's theory way out ahead of experiment. But there are also cases where experiment surprises theory. High-temperature superconductivity was discovered experimentally in nineteen eighty-six, and we still don't have a complete theoretical explanation. It's one of the biggest open problems in condensed matter physics.
Forty years and still no theory?
We have pieces. We understand certain classes of unconventional superconductors. But a universal theory that explains all high-temperature superconductivity — no. And that's humbling. It means nature is more creative than our current frameworks.
That seems like an important counterpoint to the string theory debate. Even in areas where experiments are giving us clear data, theory can be decades behind.
The narrative that physics is stuck because theory has outpaced experiment — that's true in some areas, but in others it's the reverse. We're drowning in experimental data that we don't fully understand. The so-called "nightmare scenario" for particle physics — that the LHC would find the Higgs and nothing else — has largely played out. No supersymmetry, no extra dimensions, no dark matter candidates. Theory predicted a rich landscape of new particles, and nature said no.
That's got to be existentially difficult if you spent a career on supersymmetry.
And it raises a question that's uncomfortable for the field: what counts as progress when your predictions don't pan out? Some argue that eliminating possibilities is genuine progress — we now know nature doesn't look like our simplest supersymmetric models. Others feel it's a crisis.
What's your take?
I think it's a crisis for certain subfields but not for physics as a whole. Physics is enormous. Condensed matter, biophysics, quantum information, plasma physics, nonlinear dynamics — these fields are thriving and producing results that connect to experiment and sometimes to technology. The crisis narrative is mostly about fundamental physics at the highest energy scales. That's a small fraction of what physicists actually do.
The public perception that physics is in trouble is distorted by a focus on the sexiest, most abstract problems.
And the media loves a "physics is broken" story. It's dramatic. "Physicists discover new topological phase that might be useful for quantum computing in twenty years" — that's a harder sell, but it's where a lot of the real progress is happening.
Let's go back to the career question for a moment. If someone is eighteen and brilliant at math and wants to understand the universe — what should they know before committing to this path?
I'd say three things. First, the job market is terrible and will probably remain terrible. Go in with your eyes open. Have a backup plan. Second, the day-to-day work is less glamorous than it looks from outside. If you don't enjoy math for its own sake — if you're only in it for the big ideas — you'll be miserable, because most of your time is technical detail. Third, find the right problem. The most successful theorists I've seen are people who found a niche where their particular skills match an open problem that's important but not overcrowded. That takes luck and self-awareness.
That third point seems crucial. The choice of problem is almost more important than raw ability.
It is, and it's something that doesn't get taught explicitly. Graduate students often inherit problems from their advisors. But at some point you have to develop your own nose for what's promising. That's a skill that separates successful researchers from the rest.
Can it be taught?
I'm not sure it can. I think it's partly tacit knowledge that you absorb by working with good people, and partly an intuition that some people just have. But one thing that helps is broad reading. The people who make unexpected connections are usually the ones who read outside their subfield.
Which brings us back to those fifteen hundred arXiv papers a day.
Full circle, yeah. The reading never stops.
Alright, before we wrap up the main discussion — is there anything about theoretical physics that you think is widely misunderstood? Something that would surprise people?
Here's one: the social dimension. I mentioned collaboration earlier, but it goes deeper than that. Physics is a community that spans the globe, and there's a real culture to it. The major institutions — CERN, the Institute for Advanced Study, the Perimeter Institute, the Kavli institutes — they're not just workplaces. They're almost like monasteries, places where people live and breathe physics together. Some of the best ideas come from informal conversations over meals or late-night arguments in common rooms.
That's very human for a field that's supposedly about cold equations.
Physics is deeply human. The equations are a tool, but the enterprise is driven by curiosity, ego, competition, collaboration, aesthetics — all the messy human stuff. The idea that physics is purely rational is itself a myth. Physicists are guided by a sense of mathematical beauty. They talk about equations being "elegant" or "ugly." That's an aesthetic judgment, not a logical one.
Dirac famously said something about beauty in equations being more important than agreement with experiment.
Yes, Paul Dirac: "It is more important to have beauty in one's equations than to have them fit experiment." And he was one of the greatest physicists of the twentieth century. Now, that's a controversial statement, and many would disagree. But it tells you something about the values of the field. There's an almost Platonic belief that the fundamental laws of nature must be beautiful, and that beauty is a guide to truth.
Has that belief been vindicated historically?
Sometimes yes, sometimes no. General relativity is mathematically gorgeous. The Standard Model, honestly, is kind of a mess — it's got something like nineteen free parameters that have to be measured experimentally, they're not predicted by the theory. It works, but it's not elegant in the way relativity is. So nature doesn't always conform to our aesthetic preferences.
Which is a good place to land, I think. Theoretical physics is this ongoing negotiation between what's mathematically beautiful and what's experimentally true, and the tension between those things is what drives the field forward.
That's it exactly. And the daily life of a physicist is living inside that tension — reading, calculating, arguing, failing, occasionally succeeding. It's not glamorous, but it's one of the most extraordinary things humans do. We're trying to understand the fundamental structure of reality using nothing but our minds and some mathematical tools. The fact that it works at all is remarkable.
Now: Hilbert's daily fun fact.
Hilbert: In nineteen thirty-five, a shipment of turmeric bound for Djibouti was mislabeled as saffron, leading a French colonial customs official to briefly believe he had intercepted the largest saffron smuggling operation in African history — until a local merchant identified the spice by smell, and the official's career-advancing report had to be quietly withdrawn.
That's a very specific near-miss.
The career-advancing report withdrawn. I feel for the guy.
Alright — here's the forward-looking thought I want to leave with. We talked about theoretical physics as this human enterprise full of uncertainty and failure and occasional breakthroughs. But there's a question we didn't answer: is the era of individual human theoretical insight coming to an end? As AI gets better at pattern recognition, as computational power grows, will the next major breakthrough come from a machine rather than a person at a chalkboard? And if it does, what does that mean for the whole project?
That's unsettling and important. And I don't know the answer. But it's the kind of question the next generation of theorists is going to have to live inside.
Thanks to our producer, Hilbert Flumingtop. This has been My Weird Prompts. If you want more episodes, find us at myweirdprompts dot com.
If you enjoyed this, leave us a review wherever you listen. It helps more than you'd think.
Until next time.
