Theories of Everything with Curt Jaimungal - How Tiny Ancient Black Holes Could Revolutionize Physics

Episode Date: May 12, 2025

In this episode of Theories of Everything, MIT physicist and historian David Kaiser explores primordial black holes which are hypothetical entities that may have formed before stars or atoms. David discusses their potential role in explaining dark matter, their connections to cosmic inflation, and how they might reshape our understanding of the early universe. This conversation connects the realms of quantum theory, cosmology, and the history of physics. As a listener of TOE you can get a special 20% off discount to The Economist and all it has to offer! Visit https://www.economist.com/toe Join My New Substack (Personal Writings): https://curtjaimungal.substack.com Listen on Spotify: https://tinyurl.com/SpotifyTOE Become a YouTube Member (Early Access Videos): https://www.youtube.com/channel/UCdWIQh9DGG6uhJk8eyIFl1w/join Links Mentioned: •⁠ ⁠David Kaiser's published papers: https://arxiv.org/a/kaiser_d_1.html •⁠ ⁠Bell, J. S. “On the Einstein‐Podolsky‐Rosen paradox” (1964): https://cds.cern.ch/record/111654/files/vol1p195-200_001.pdf •⁠ ⁠“Ultra-High-Energy Neutrinos from Primordial Black Holes” (2025): https://arxiv.org/pdf/2503.19227 •⁠ ⁠“Cosmic Bell Test Using Random Measurement Settings from High-Redshift Quasars” (2018): https://arxiv.org/pdf/1808.05966 •⁠ ⁠“Close Encounters of the Primordial Kind” (2023): https://arxiv.org/pdf/2312.17217 •⁠ ⁠“Primordial Black Holes from Multifield Inflation with Non-minimal Couplings” (2022): https://arxiv.org/pdf/2205.04471 •⁠ ⁠“Testing Bell’s Inequality with Cosmic Photons” (2013): https://arxiv.org/pdf/1310.3288 •⁠ ⁠“Planck Constraints & Gravitational-Wave Forecasts for PBH Dark Matter Seeded by Multifield Inflation” (2023): https://arxiv.org/pdf/2303.02168 •⁠ ⁠“Light Scalar Fields Foster Production of Primordial Black Holes” (2025): https://arxiv.org/pdf/2504.13251 •⁠ ⁠“Debye Screening of Non-Abelian Plasmas in Curved Spacetimes” (2023): https://arxiv.org/pdf/2309.15385 •⁠ ⁠“Primordial Black Holes with QCD Color Charge” (2023): https://arxiv.org/pdf/2310.16877 •⁠ ⁠A Brief History of Time – Stephen Hawking: https://www.amazon.com/dp/0553380168 •⁠ ⁠In Search of Schrödinger’s Cat – John Gribbin: https://www.amazon.com/dp/0553342533 •⁠ ⁠How the Hippies Saved Physics – David Kaiser: https://www.amazon.com/dp/039334231X •⁠ ⁠Drawing Theories Apart – David Kaiser: https://www.amazon.com/dp/B002Y5W2X2 Timestamps: 00:00 – What Are Primordial Black Holes? 01:41 – Could They Be Dark Matter? 05:21 – Kaiser’s Academic Journey 10:56 – Studying Physics and Its History 11:57 – Cosmic Inflation Basics 15:31 – Direct Collapse vs. Stellar Collapse 25:14 – Bell’s Theorem Explained 38:32 – Quasars and the Cosmic Bell Test 43:04 – High-Precision Astronomy 47:38 – Learning Curves & Interdisciplinary Research 48:17 – Scalar Fields and Inflation Models 55:05 – Black Hole Formation from Inflation 58:41 – Black Hole Mass as a Cosmic Clock 1:02:50 – Quark-Gluon Plasma & Color Charge 1:08:46 – Critical Collapse and Mass Spread 1:11:34 – Charged Primordial Black Holes 1:13:54 – Big Bang Nucleosynthesis Implications 1:20:14 – Detecting Black Holes Locally 1:23:51 – Tracking Planetary Wobbles 1:26:04 – Hawking Radiation & Positron Signatures 1:30:06 – Why Track Mars, Not Earth? #science Learn more about your ad choices. Visit megaphone.fm/adchoices

I want to take a moment to thank today's sponsor, Huel, specifically their black edition ready to drink. So if you're like me, you juggle interviews or researching or work or editing, whatever else life throws at you, then you've probably had days where you just forget to eat or you eat something quickly and then you regret it a couple hours later. That's where Huel has been extremely useful to myself. It's basically fuel. It's a full nutritionally complete meal in a single bottle. 35 grams of protein, 27 essential vitamins and minerals, and it's low in sugar. I found it especially helpful on recording days so I don't have to think about prepping for food or stepping away to cook. I can just grab something in between conversations and keep going. It's convenient, it's consistent, it doesn't throw off my rhythm.

You may know me, I go with the chocolate flavor. It's simple and it doesn't taste artificial. That's extremely important to me. I was skeptical at first but it's good enough that I keep coming back to it, especially after the gym. Hey, by the way, if it's good enough for Idris Elba, it's good enough for me. New customers visit huel.com slash theoriesofeverything today and use my code theoriesofeverything to get 15% off your first order plus a free gift.

That's huel, H-U-e-l dot com slash theories of everything, all one word. Thanks again to Huell for supporting the show. Become a Huell again by visiting the link in the description. What's better than a well marbled ribeye sizzling on the barbecue? A well marbled ribeye sizzling on the barbecue that was carefully selected by an Instacart shopper and delivered to your door? A well marbled rib eye you ordered without even leaving the kiddie pool? Whatever groceries your summer calls for, Instacart has you covered.

Download the Instacart app and enjoy $0 delivery fees on your first three orders. Service fees, exclusions, and terms apply. Instacart, groceries that over-deliver. Today we have something different for the audience of Theories of Everything, and I'm super excited to speak about it. I'm going to get into exactly why today's episode is different, but I'll ask this preliminary question and perhaps in your answer it'll be clear which direction we're going. But what are primordial black holes and why should anyone care? Good. Okay. So primordial black holes are as yet hypothetical.

We don't know they exist, but they're really intriguing idea. And they were put forward by a few different researchers more than half a century ago. So the idea has a long history by now. The idea in brief, and I'm sure we can unpack it together soon, is that these are black holes that would have formed not through the ordinary route,

by having a star that exhausts its nuclear fuel, gravity wins, it collapses and crushes down, and forms what we now call an astrophysical, or stellar collapse black hole. We now know those are real real and they litter the universe. They're very common in fact, these stellar collapse or astrophysical black holes. These primordial black holes are hypothesized

to follow a different route that they would actually short circuit all of stellar evolution and it would form by the direct collapse of some original early universe or primordial lumpiness, some inhomogeneity in the distribution of matter and energy, which is different from saying you had a star and a whole life cycle collapse. So these things could form not only independent of stars, but long, long before there existed

stars, in fact, before there existed stable atoms. So these really have a very, very different history if they exist in our cosmos. And so we can unpack that and talk about it some more. But one among many reasons why they're now of interest to a growing number of researchers across sort of fundamental physics and astronomy and cosmology is because these might be a candidate, for example, for dark matter. If they have certain masses and properties, we can talk about that. If they form with larger masses,

then they might be candidates that could explain these supermassive black holes that we now know lurk within pretty much every galaxy that's been seen. So we have lots of questions about the cosmos and primordial black holes seem to offer a pretty cool way to maybe start to answer some of those really long-standing mysteries. Broadly speaking, there are these two ways of learning large swaths of material and connecting them together. So one is to learn everything,

like everything that's in a term theory of everything or this channel's name. However, it's difficult to do so in a manner that's more than, say, three or four layers deep on any given subject, just due to time constraints. Now, the next method is to paradoxically specialize in one tiny domain and then do that extremely well. The sounds counter to the whole spirit of the wide scope of everything.

But to learn that one specific topic, you have to then approach it from multiple angles. I don't know if you've played this game Katamari or if you've heard of it. I don't know it, no. Okay, so in Katamari, you're this tiny little figure that then pushes one special object. It's small. There's a stickiness to it. So you start to get some small bits of paper attached to it, maybe some toothpicks, and then eventually a glass bottle, and then eventually cows and buildings. So when I talked to you, initially we were going to speak about the history of physics

and we'd still touch on that later on in this conversation. But speaking off air, it was clear that you used primordial black holes as this sort of subject that touches every other area related to fundamental physics. And I don't think you intended it to be as such. So I want this episode to not only be an exploration of primordial black holes and not only every other area of fundamental physics or as much as we can touch, but also this generic process of having a topic that allows one to become both a jack of all trades and a master of

some. Yes. I think that's a really great way of putting it, Kurt. And I agree. And that's, again, that's an unintentional journey I now find myself on. I didn't plan that when I began working in a more focused way on Permanent Black Holes three, four years ago with some amazing colleagues and students, collaborators. But I think you're right. Now in hindsight, looking at the path I myself have really enjoyed wandering and research in the last few years, following the primordial black holes, following the idea of primordial black holes really

has led me to not every area of physics, of course, but to a bunch that were familiar to me. I could start from a familiar home base. That's how I began first thinking about these, when you talk about that. But then really to other areas of physics that I knew a little about and some that I knew very, very little about.

And now I've had the great luck to get, to spend some time and learn more about those other areas as well. Always, as you say, connected to how might these relate to primordial black holes as my central question. And then you, I don't know if the spokes radiate inward or outward, but around that, that becomes a node around which I,

and again, many very wonderful colleagues, can try to connect lots of dots among areas of physics, among subfields, among topics that often are treated kind of as if they're on separate lanes. So that's been a great joy ride. It's been really actually very fun for me to do, I think, very much like what you're saying.

Tell me more about this familiar starting point of yours. Yeah, sure. Why don't you just walk us through your journey in physics, or even your just academic journey in general. Maybe it didn't start with physics. Sure. No, I'd be glad to. So we can turn the clock way back. As a high school student,

like maybe many people today who enjoyed this channel, I was really, really hooked on what you might call popular science. At that point, it was mostly books, not amazing multimedia stuff, cheap paperbacks. I was growing up in the era when Stephen Hawking's book, A Brief History of Time, first came out. So I was still in high school when that book appeared

in the late 1980s, for example. But even before then, I was reading just a slew of, I think, really good, really high quality books written for non-specialists for broad readerships, written often by practicing scientists, some by very talented science writers, and some people who were really kind of combining the two. And it was just thrilling.

It just felt like an intellectual adventure in my teenage years, my high school years. And some of them, I remember some very dearly by the author John Gribbin. And so he had a whole series called In Search of Blank, In Search of the Big Bang, In Search of Schrodinger's Cat, and he had many that eventually filled the shelf. Those two that I mentioned, the Big Bang and Schrodinger's Cat ones really grabbed me, and they came out even before Stephen Hawking's, you know, much better known book.

And the first of those, the one on the Big Bang, was really a tour without very much mathematics, probably none at all, but a tour of the big ideas. They came together to what we would now recognize as the Big Bang model. And he closes the book with some early hints about cosmic inflation and a kind of revised understanding of what we might now call the Big Bang. The book came out just a few years after the first proposals by the real experts on cosmic inflation had been

published, as I later came to realize. these books were published in the mid 1980s. And the kind of foundational papers on cosmic inflation were published in 1981, 82, 83. This was hot stuff. By the time we'd already made it into these popular books. Likewise for Gribben's book on Schrodinger's cat, of course, as the title suggests, it was a really, I thought,

engaging, inviting introduction to quantum theory. Some of the juicy, juicy, you know, nuggets that many of us still stay up late at night thinking about. Things like what we might call the measurement problem, the role of supervision, of course, quantum entanglement, and so on. Bell's theorem. And again, I just was hooked. Okay. So, I get to my undergraduate studies thoroughly convinced in my soul, I want to do physics that turned out to remain to be true. But really curious also about these human stories. Who were these people who stayed up late at night wondering about these things and often having very extended arguments and debates?

It was actually a really remarkable mentor, my first real advisor in college physics, a person named Joseph Harris, who's an expert in classical general relativity. That was his great passion and what he'd studied for a long, long time. But by the time I entered college, Joe had cultivated really broad interests,

kind of on the side, and he'd be reading postmodern Italian poetry, or be reading the notebooks of the German novelist Thomas Mann. He was just this remarkable broad-minded person within and well beyond physics. And it was Joe who said to me, if you have all these interests,

there's this thing called the history of science, you should go check that out. I said, I never heard of it. What did I know? I was 18 years old. So it was really Joe, the classical relativist who really helped open my eyes to a second field that I very rapidly fell in love with and get to pursue to this day, the history of science.

Joe connected me or pushed me to go meet two actual historians, historians of science, at the campus I had just begun my studies. They very generously took me under the wing. As an undergraduate, I did a double major in physics and in the history of science. On the physics side, I delved more and more into early universe cosmology, learning about the still relatively new ideas about cosmic inflation, origins of large-scale structure,

all these kind of very cool ideas that we could use, that we could try to address using the tools of things like quantum field theory, especially quantum field theory in curved space-time, which has its own kind of beautiful formalism to it. I got a little taste of that even as an undergraduate and was able to do my undergraduate thesis on cosmic inflation and so on. And then as my undergraduate years were passing along, I met a few people by that point who had done

this strange sounding thing where they had gone to graduate school and done a PhD in a scientific field and a PhD in the history of that scientific field. One of my undergraduate mentors on the history side was Naomi Oreskes, still a very dear friend. And Naomi had done, by that point only recently completed, a PhD in geology and a PhD in the history of earth sciences. And her history advisor had been Peter Gallison,

another very dear friend of mine. Peter had done a PhD in particle theory, essentially beyond standard model particle theory, and a PhD in the theory, essentially beyond standard model particle theory, and a PhD in the history of modern physics. And so I figured, well, two points define a line, there's at least two instances, there are more than only those two I've come to learn. But with their example in mind, I wound up applying to graduate school to do both theoretical physics and the history of science with their support. And I was lucky and able to do that. So for my PhD, I did a PhD in theoretical physics

and a PhD in parallel in the history of science. And on the physics side, I continued to explore more and more this early universe cosmology, these ideas about things like cosmic inflation. And luckily for me, my main thesis advisor became Alan Guth, who had helped of course to invent this whole body of work.

And he again remains a very dear friend and now we run a research group together. It's kind of a dream come true. So anyway, so from undergraduate days through my PhD, I've been really immersed in cosmic inflation. We can of course talk more about that. And so I didn't work on primordial black holes right from then. Other people were even in the 90s working on them. They were not such

a central topic then. I had other interests that I pursued. My dissertation was on how would inflation have come to an end? So this era we can now call post inflation reheating, which is really sort of setting up the conditions for the standard Big Bang model. Lots of fun, juicy stuff to study. And I was really having, frankly, a lot of fun with that. And then many years later, I kind of came around to an idea that I say some people had been pursuing for quite some time, that during this phase of cosmic inflation, where we know, when we got very good at calculating the expected spectrum of primordial perturbations, of essentially density perturbations, these are arising in our account now from quantum fluctuations of the fundamental field or multiple

fields that were responsible for driving that phase of inflations, phase of accelerated, very rapid expansion of space for a brief moment of time, but very rapid growth in size. And that already is the kind of framework within most of us think about the origin of large scale structure generally in our universe. Why are there clusters of galaxies and then huge voids? There's a remarkable inhomogeneity in the universe across length scales on the order of say tens to hundreds of megaparsecs and below. And if of course granted across still longer length scales, the universe looks remarkably smooth. How do we account for this smooth and is giving way to structure? What

is seeded that structure? And that was a pressing problem from the 70s and 80s and well into the 90s and beyond. And cosmic inflation provides a, I think, really elegant framework to try to begin to answer that question remarkably by saying these things ultimately come from quantum fluctuations of the sort that we otherwise study in other classes and other laboratories whose wavelength was stretched to astronomical scale during this very rapid but brief period of stretching of accelerated expansion called inflation. So we already had to get very good and very careful at calculating the spectrum of primordial perturbations during inflation to compare with high

precision measurements of the cosmic microwave background and now many more measurements that we care about. And so as many people have been wondering for a long time, could that same basic process during inflation have led to amplification of a still sharper higher peak on much, much shorter length scales of quantum fluctuations that could actually induce gravitational collapse? So the ones that I was used to studying that we want to think about for the cosmic microwave

background, those are necessarily of a very small amplitude and very, very long wavelength after they get stretched during inflation. And the idea was could that same phase of the early universe, if there were some distinct dynamics later during inflation but before the end of inflation, long after the perturbations we care about for the cosmic microwave background had already done their thing and been stretched far outside the Hubble radius, could there be other dynamics during inflation that could lead to a much sharper peak of these primordial overdensities, curvature perturbations, that could then cross outside the Hubble radius, a little while later come back inside the Hubble radius,

and induce collapse directly to a black hole. That's a direct collapse that I mentioned a little while ago, that would short circuit the need for stellar evolution. Could black holes form because they were very strong, likely narrow peak but high amplitude fluctuations of essentially the quantum mechanical nature that got amplified late during inflation for reasons again, we'd be happy to dig into if you'd like. And then those could collapse to form a population of black holes. And then those would have sort of different, some sets of different properties compared to stellar collapse like holes

that astronomers had gotten very good about thinking about in the interim. So there are two directions here I want to take it. Good. I'm not sure where to go. Okay, for one, I want to rewind and ask about your colleagues who went into the history of geology, for instance, and then also studied geology. So you did physics, so history of X, but also studying X. Is the history of X in service to studying X

or is it just something to appreciate in and of itself? That's a great question, Kurt. So the person I mentioned who did the history of geology and was trained and was an active geologist for many years is Naomi Oreskes. She was one of my very important undergraduate advisors. But your question is more general. I think you're right. For a long time, in my own thinking, I thought they were both wonderful bodies of knowledge about which I was deeply curious and wanted to learn more. I didn't think that either was necessarily in the service of

the other except in some limited way. For example, my historical interests then is now are fairly recent physical sciences, sort of 20th century and even often last half century or so, pretty recent. When as you know, a lot of work in modern theoretical physics got pretty complicated, pretty technical. I wanted to make sure that as you know, a lot of work in modern theoretical physics got pretty complicated, pretty technical. So I wanted to make sure that as an historian, I could follow what the people were doing in the 1940s, 50s and 60s.

And that meant making sure I had my own chops sharp, that I could really follow not just the published and polished research articles in the journals, but the more messy notes, the correspondence, the summer school lectures, the kind of incomplete thoughts that sometimes are captured on paper as well that I found just really fascinating. I want to make sure I could do justice to what they thought they were doing and why. So that meant I had to make sure my own physics training was adequate to make sense of what they were doing not

so long ago without, while being on guard, about falling into a kind of anachronism or presentism while we now know this about the behavior of quarks or Phil and the black, which they didn't know then. So you want to make sure you don't actually start reading things into the past just because they seem more obvious or self-evident now. It's a little bit of a balancing act, and one that frankly I enjoy. I wanna be careful not to misrepresent

what people thought they were doing in the 1940s, 50s, 60s, or before or after. But I also wanna make sure I can, so to speak, read the language. I wanna make sure I'm conversant with what was likely on their minds, why they pursued this calculation. Oh, look, they made an error on page three, but I get it.

Here's why that came up. You know, so it's in that sense, the physics was in the service of my history of science in a limited, in a kind of, let's say capacity sort of way. But I didn't think that one was otherwise deeply informing the other in either direction. And maybe later on we'll get to talk about a pretty fun counter example, which I didn't expect. But the preview for that is that I wrote a book on aspects of the history

of quantum entanglement in Bell's theorem. As an historian, I was fascinated by the topic. Who cared about that topic when and where? Why was it pursued in some places, not others? Just as an historian, I wanted to know more about the history of people grappling with foundations of quantum theory, including Bell's theorem and entanglement. So I wrote that as a historical, you know, kind of exercise. And I was, I had a lot of fun doing it, dug in with everything I had to find the right sources. Just really great fun. And then after that, began talking about that topic with some of my young physics colleagues. And we realized, they originally realized, and then I was with some of my young physics colleagues, and we realized, they originally realized, and then I was lucky to join them in the next steps,

that given what we know now as astrophysicists and cosmologists about the large-scale structure of the universe, we can actually go back and imagine doing new types of tests of Bell's inequality in novel ways to address loopholes that have been identified, I'd learned about as an historian,

that had been identified in I'd learned about as an historian, that had been identified in the literature 50 years earlier. So that was one where the historical work actually helped catalyze a whole new multi-year research program that I, again, just had an amazing amount of fun pursuing. This is what became the cosmic bell experiments using quasars and all that. So we can talk about that. But that's an example where it went in the direction I didn't expect.

Where a really kind of in-depth book-length history study, you know, 350 pages, a thousand footnotes, like all the good juicy history stuff I worked so hard on, that actually helped lead to new questions for when I put my physicist cap back on. Typically until that time, I'd kept them as, you know, let's say parallel pursuits and tried not to let one kind of bleed into the other too much. Why do you have to try to not let one bleed into the other? You mentioned an

example of quarks that in the 1970s or so on we know quarks do this and that or 1980s or what have you. Yeah. And then you're reading some material from the 40s and you said it's easy to read into it quarks, something like that. Can you give me an example? Yeah, that's right. It might not be quarks per se, but let's take the topic of renormalization. My first book as an historian coming out of my history dissertation was on the history of

Feynman diagram techniques in quantum electrodynamics during the early stages of what becomes together as renormalization. Well, as you know, people have thought about renormalization in quantum field theories lots of different ways over time. That is not a stable target. And so by the time we get into things like the very different view from, let's say, a more modern perspective with effective field theories, where we deal with non-renormalizable interactions all the time and don't break a sweat over it,

that somehow the status and the role of renormalization is really quite different to a working theorist today than in the 1940s or really into the 1960s and early 70s. That's one example. Then the actual techniques of performing renormalization have changed. So in the early days they weren't doing They were very rarely doing, you know dimensional regularization. They're doing other kinds of techniques So I just and I want to make sure like why did why what was their toolkit? Why did they think that was a productive way forward?

What do they get stuck on and not say and not always be second guess like oh But but wasn't this answer obvious because it wasn't obvious and this took generations, right? That's what I mean. So I don't want to kind of misrepresent the path that seemed obvious to them at the time because it takes much more time and many more pairs of eyes and hands for where we are today to have come about. for where we are today to have come about. Okay, this is interesting. Had Dirac lived after the Wilsonian revolution, do you think he still would have said renormalization was sweeping infinities under the rug?

It's a great question. So of course, I don't know, obviously, it's a counterfactual, but it's interesting. I don't know if the EFT framework would have struck Dirac as sufficiently beautiful. It might have. There are some things that I find aesthetically amazing about RG and RG flow and all this new, more modern way of seeing it. But Dirac, he had a, and he wasn't apologetic about it, he had a very explicit kind of aesthetic sensibility with his very austere mathematics. And I don't know if this would have met his approval or not.

It's an interesting question, I don't know. Now, I don't mean whether he would have found it beautiful or not. I mean, would he still have found that you can get finite answers doing something with infinity? What would his view of that be after Wilson? That's what I mean. And it's a good question, but I'm not sure

if he would have found the EFT or RG framework sufficiently beautiful because for Dirac, beauty really was often a criterion for truth or for likely truth. And as he was, he often was led by the sense of the kind of powers of the mathematical formalism and the more bare bones, the better. He was famously a person of very few words himself.

He barely spoke, these kind of stories that still resound. His amazing textbook on quantum mechanics still in print, first published in 1930, it's a pretty good lifetime for a book. It's very sparse. I mean, he doesn't want to get lost in a lot of verbiage. And not just words, he wants the mathematics to be kind of as crisp and clean as possible. And whether he would give the gold star to the most modern techniques today,

it's hard for me to judge. I don't know. I don't know. Okay, here's a fun counterfactual. Yeah. So Dirac famously said to Feynman, I have an equation du. Now, what if Feynman had retorted a couple years later, I have an integral, I have a diagram, do you? What do you think Dirac would have said?

That's a good question. So I don't know that Dirac would have been so enamored of the Feynman diagrams. I have a feeling Dirac's reaction might have been more like Julian Schwinger's. Here, I'm guessing, right? But Schwinger might for this might be a good proxy for Dirac in the reactions. Great. And as you might know, Schwinger was certainly in the early years, no fan of the Feynman diagrams.

He once sniffed very haughtily. The Feynman diagrams brought computation to the masses and that was not meant to be a compliment. So, oh, if anyone can do this from drawing little cartoons, you know. So Schwinger, I think in a very Dirac-like way, was really enamored by this pristine kind of algebraic, you know, kind of austerity. So, I can imagine Dirac's reaction to Feynman's approach to have been maybe more like Schwinger's and therefore maybe not so thrilled at first. Yeah. Okay. What do quasars tell us about Bell's theorem?

What is Bell's theorem and what are quasars as well? Good, good, good. So let's start with what is Bell's theorem. Bell's theorem is just a landmark, landmark of modern science. And I say not just modern physics. I think it's broader even than that for its intellectual sweep.

It's merely six pages. It's a very elegant and brief journal article. It rewards rereading to this day. Bell, I think, was an exceptionally careful and disciplined writer. It's very clear. He wants to make his assumptions as clear as possible. It's clear. The article that we're talking about is called On the Einstein-Podolsky-Rosen Paradox. He's clearly referencing the so-called EPR paper, which had come out almost 30 years prior to that point. And Bell's article was published late in the autumn of 1964. So we just passed, what? I guess it's 60th anniversary, I guess. So it's been with us for a while. Mm-hmm.

In this very brief paper, John Bell was concerned not so much about quantum theory per se, but about possible alternatives to quantum theory in the language that was then known at the time as hidden variables theories. So inspired by work by people like Albert Einstein and other architects of quantum theory like Erwin Schrödinger who came to be very skeptical and dissatisfied with what we might recognize today as kind of ordinary or orthodox quantum mechanics. Bell wondered, was there any way to put into a quantum-like framework, a way of ascribing to quantum objects,

definite properties prior to and independent of our measurement of them. If so, could those properties also nonetheless obey what we might call locality or local causality? Is there a way to make quantum mechanics look more compatible with relativity where nothing travels fast in the speed of light, that's the locality, and in which there are really are definite properties to little bits of matter that we can attribute whether we

perform a measurement or not. And those are guiding principles that that people like Albert Einstein thought should be part of any acceptable physical theory and Bell thought those were awfully reasonable principles. So Bell is wondering what would it take to develop a theory of the micro world in which those elements held true where you could attribute properties to particles ahead of time before measurement and in which local events yield only local outcomes, local causes yield only local effects, let's put it that way. And he winds up formulating this very, again, very succinct, very elegant framework and finds that in any putative theory of the micro-world in which those two postulates hold, then there's an upper limit to how strongly correlated

the outcomes of measurements can be on any pair of particles if they had been prepared together but since traveled apart. So he has in mind the EPR paper, Einstein-Podolsky-Rosen. He's thinking about what we would now call pairs of entangled particles and he's wondering if a theory of nature is going to have these very reasonable sounding attributes, objects have their own properties, nothing travels fast in light, then what are the implications empirically for things like performing measurements on pairs of particles that have traveled in opposite directions far apart? And what he derives is an upper bound, that's why it's an inequality,

on a measure of how correlated the outcomes can be, even in principle, on measurements and questions we might ask of each of those particles. If the theory describing them obeys sort of Einstein's preferred postulates, he finds there is an upper bound, and then he goes on very quickly to show a now standard calculation in ordinary quantum theory. The quantum theory predicts stronger correlations. That if you prepare particles in a particular quantum state, let's say, you know, a quantum singlet state for two particles, a classic entangled state, shoot those particles in opposite directions, perform measurements in different bases,

different choices of what to measure on each side, that for clever choices of the quantum state and clever choices of the measurements to be performed, the outcomes of those measurements can be more strongly correlated. They'll line up much more often, dramatically more often than any Einstein-like theory could ever allow. So that quantum mechanics does not obey

this conjoined pair of postulates of basically what became known as local hidden variables. So that's pretty amazing. And he also says, now he was a theoretical physicist, but he realizes in principle something that could be measured in a laboratory. So this becomes known as Bell's inequality orquality or Bell's Theorem such that there's an upper limit to the degree of correlation, behavior of particles if they're obeying these local causal relationships. And then years go by, several years go by before pretty much anyone pays any attention.

One of the first to pay attention was a then very young experimental physicist, John Clauser, who got very excited about this. He saw the implications right away. He was a PhD student at Columbia University at the time, in the late 60s, and was very actively discouraged by his PhD advisor to pay any attention to this. He was disparaged as mere philosophy. Why?

I think the general question or the topic of the foundations of quantum theory generally, let alone very specific topics like Bell's theorem, were really out of favor, out of fashion throughout the physics community, especially in the US, but not only in the US at the time. I've written a bit about why I think that was the case. I think it has as much to do with intellectual trends as with institutional changes in the way physicists were being trained coming out of the Second World War. And again, I'm happy to talk about that. But for a confluence of reasons, Clauser was of that generation, of the few generations, that were actively discouraged and sometimes really in very strong terms from pursuing any of these questions at the foundations of quantum theory including

Bell's inequality. So, Clauser then finishes PhD on a different topic, got his first postdoctoral appointment and then kind of was curious to go back to this question that had now been lodged in his mind for three or four years. Could one really do an honest to goodness laboratory test of this Bell's inequality? After all, he was by that point a really very well trained experimental physicist. So he wrote directly to John Bell in 1969 and said, has anyone done this experiment since then? You know, I was told not to, has anyone done it? And Bell wrote back with great excitement saying it was among the first questions he's gotten from any physicist in the world about this work.

Wow. Four years later. And Bell confirmed no one had done the experiment, few people showed any interest at all. It would be amazing to do it. And as famously as Bell concludes his private letter to Clauser, if you find results that are different from quantum mechanics, that would shake the world. That's the phrase that Bell had used. The stakes seemed high. So Clauser was fired up and he teamed up with a small number of like-minded colleagues to pursue this.

Again, happy to go into more of the history, but that's the kind of Bell's inequality part. Okay. Now, let's fast forward a little bit. It turns out that Bell's theorem is a mathematical theorem, which means it depends on starting assumptions, right? And so, do those assumptions hold in the real world? If you're going to do an experimental test, you have to show that your experiment is consistent with the starting assumptions and not just that you found some results any old way. And so what Bell himself and Clauser and others like Abner Shimon in a whole

list of people in the mid 70s began to realize and identify is there are all these kind of what came to be called loopholes that have to be addressed in any given experiment if you're really going to conclude that the strong correlations that you presumably are going to measure are because of a violation of Bell's inequality. That's to say there are all kinds of subtle, sometimes weird sounding scenarios in which a perfectly Einstein-like theory, perfectly consistent with local hidden variables, could yield these very strong correlations. One obvious one that Clauser and Bell themselves wrote to each other about right in the early years of this would be if you somehow if information could be kind of flowing throughout the experiment, if information could be leaking from one side of the test to the other, if one particle is measured here, let's say particle A is measured here, and then some enough

time goes by so that a single light signal could have traveled from here to there just at the speed of light nothing fancy. And then later you measure properties of second particle well maybe there's room for coordination of the outcomes because it was sharing information. If I got I measured this I asked this question here and got this answer make sure you're line yours lines up. Yeah. Now and so that's what became known as the locality loophole. It's very hard to address experimentally. In the early years, it was really technically a great challenge. But it was, if you don't take that into account,

then you're not proving a violation of Bell's inequality if you nonetheless find strong correlations if you don't have the right space-time arrangement of each relevant event in your experiment. They go, they play these games over and over again throughout the 70s. Another one that they came up with that Bell himself had overlooked, and it was pointed out by people like John Clouser, Abner Shimonie, Michael Horn, about a dozen years after Bell first published his theorem, published in 1976, in a little out of the way place, a little newsletter, was something that comes to be called today the freedom of choice loophole.

It has other names. That's what it's often called. This is not about the flow of information during a given experiment, did detector A message detector B? But instead it's about shared common causes. Could there have been any subtle influence or event that you otherwise hadn't taken into account that could have nudged or previewed the series of questions

to be asked at each detector in advance without changing what it can do, even if you know what questions will be asked when, and then could have communicated that in advance to the source of entangled particles before the particles are emitted. In that case, it's like getting a copy of a pop quiz the night before, right? If you know exactly the order of what questions will be asked when, then you and your

twin at home can say, okay, this one's going to hear, let's make sure our answers line up. Yes. Now they're leaving on the school bus stop. So it's no longer mysterious. It's consistent with Einstein's principles, have strong correlations. If there was some flow of information, not during the conduct of an experiment, but from some shared common cause before. So that's what got my colleagues and me really excited. That's what we wound up thinking about after I'd written this book on the history of entanglement in Bell's theorem.

This was with Andy Friedman and Jason Calicchio originally. They were very good friends in graduate school. Andy had just come to MIT to start a post-doc with me working on other aspects of cosmology. We thought we'd work on dark energy and stuff. Andy read my history book and got intrigued by it. We all began talking about, could we address this really stubborn,

the last of the most stubborn loopholes in Bell tests, using what we now know as astrophysicists and cosmologists about the large-scale structure of the cosmos since the Big Bang. This point in space-time could not have shared a single light beam with that point in space-time, that kind of question, which is something that's kind of bread and butter for cosmologists today. Wasn't so common or certainly not as well-known or

constrained in the 1960s to 70s. So we wrote up a proposal, a whole article coming out in PhysRev letters saying, if we use very distant astronomical objects like high redshift quasars on opposite sides of the sky, and we trigger in real time on some measurement of that astrophysical light, let's say quasar A over here, its light was emitted most of the history of universe ago.

It's so far away from us now. That light's been, the light we measure now in a telescope had been traveling for 8 billion, 12 billion years out of a 14 billion year universe, that kind of thing. You measure it in a tiny fraction of a second here on Earth and you perform something like the color of that light. Is it more red or more blue than average for that quasar? So you do some prep ahead of time.

Here's the typical spectrum for that quasar. Is the light you measure right now more red or more blue than that average one? Do the same exercise with a different, very carefully chosen quasar on the opposite side of the sky, whose light is coming toward, can be measured at detector B. These are now separated on the face of the Earth.

Likewise, ask is that light in that tiny microsecond, more red or more blue. You perform these real-time updates after a pair of entangled particles are prepared in your Earthbound laboratory and sent on the merry way. So the choice of what measurement to perform was not knowable even in principle at the time the entangled particles were emitted. So sometimes people get a little confused.

I think we're measuring entanglement from the sky. I wish. That's also a cool question that I'm interested in. That's a separate thing. Here what we're doing is using as thoroughly unentangled, as uncorrelated random bit streams as possible. So it's just a way of you saying, look, we need distant observers or distant people to choose the measurement.

Let's use the quasars. That's right. In a way that information about that choice could not have been previewed or whispered in the ear of any other part of the experiment ahead of time. Right. Because that one bit of astrophysical light was traveling for eight billion years, for 12 billion years. We even had to make the alignment very careful so the causal wave trains of one could not

have reached any other part of the experiment in time. So really, frankly, very lovely relativistic astrometry, basically to say that what's the information that could possibly have been gleaned from that quasar now about this quasar photon and that's a way of shielding anything about the choice of measurement to be performed at detector B from either the source of entangled particles or detector A and vice versa. So it really was to say we want the choice of measurements to perform, not to have any possible kind of cross coordination or any statistical correlation with each other, but especially not with the source of entangled particles.

No one could have gotten the quiz ahead of time. There's no way that people got the quiz. The questions for the quiz weren't even written until after the particles left their laboratory to start their journey. And just a moment, the choice of a quasar other than some other extremely distant object, the choice of the quasar is why? Good, because we knew our optical astronomy friends have gotten very good at performing very rapid cadence, precise measurements of light from quasars. We proposed this in a theory paper.

Luckily, both Andy and Jason had very strong backgrounds in observational astronomy, which I do not. But we wrote this first as a proposal saying, if you had a telescope with this size mirror, you'd count this many photons per second from an appropriately distant source, it will be feasible to do. Then we were extremely lucky to get to pitch this idea to Anton Seilinger,

just a renowned wizard in the field, expert in quantum optics and a lifelong expert in testing topics like Bell's inequality and putting quantum entanglement to work. So Anton, we pitched this to Anton, he got very excited and enthusiastic, which was just, I mean, a dream come true. And so then we were able to secure some, frankly, modest funding from the National Science Foundation. Anton secured funding from the Austrian Academy of Sciences. He's based in Vienna. So we put the collaboration together. We did a pilot test in Vienna with bright Milky Way stars and hobby scale telescopes.

We literally took a copy of Sky and Telescope magazine, when Anton was visiting MIT, turn to the back page, said, Anton, buy us two of these, please. Eight to 10 inch simple hobby telescopes would be fine for the pilot test. We do bright stars, they shoot out a gdillion photons a second,

which will prove that we can do the electronics at timing right. So we did that in Vienna, produced already a remarkably improved experimental test of Bell's inequality, because the most recent time when there could have been any coordination among local factors to account for these strong correlations that we measure among the entangled

photons by something other than ordinary quantum mechanics, we pushed that back to be something like 600 years. And until our experiment, the longest that had been pushed back was like a millisecond before a given experiment. So we went from 10 to the minus three seconds to 600 years with our pilot test, with our cheapo pilot test, which is a great thrill. Many orders of magnitude. On the strength of that, Anton in particular was able to

persuade the telescope operators on the island of La Palma, the Roque de los Muchachos Observatory on the island of La Palma, the Roque de los Muchachos Observatory on the island of La Palma, the Canary Islands. That has some of the largest optical telescopes on the planet. And in particular, it has two of the medium-sized ones, two four-meter optical telescopes that were able to commandeer for several nights, all night, even though these are in such high demand for the astronomers. So because the pilot tested gone well and we went on the astronomers off season, we got time on these telescopes. And again, the idea there was to use four meter telescopes, you know, with roughly 13 feet across. You can collect light from very distant,

very dim objects like these high rich of quasars, the light that had been traveling for most of the history of the cosmos. So that's really what it came down to. The Antons group was able to do extremely rapid measurements of the relative color of each quasar photon. That's something as quantum optics people, they could filter on color extremely rapidly, knowing what the optimum would be.

So a little far off, higher or lower frequency, they could measure that. In rapidly, knowing what the optimum would be, so a little far off, higher or lower frequency, they could measure that in a tiny fraction of a second. They could then actuate with something called a Pockel cell. Given the outcome of that astronomical measurement, they could then, because they're wizards at this, could literally rotate and change the basis within which an earth bound entangled photon would be measured.

They could change the polarization basis, something called the Paco cell, that could change every half of a microsecond. So then the challenge is to have the baseline be long enough that the travel time for the entangled photons is several microseconds, which means you have to be on the order of kilometer. Because light travels so fast, you need to be able to make an astronomical measurement, physically change the instrumentation,

change the measurement basis in which you'll tickle, you'll measure that incoming entangled photon and do it all after those entangled photons had already been emitted. So they had no foreknowledge of the particular measurement to which they'd soon be subjected. Man, what's most interesting to me is that you, along with some other people, and a few thousand dollars, were able to improve upon a previous result by several orders of magnitude. It was a joyride. I mean, so, and we couldn't have done it without the team. I mean, so part of what I enjoyed so much about this was, again, like we were saying in the beginning, I got to learn all kinds of things I didn't know much about before.

I have no training in laser optics. I still know not very much, but I knew more than I ever did thanks to working very closely for close to five years with these amazing friends and colleagues, the ones for whom that's their expertise. Even on the theory side, I had studied Bell's theorem as an undergraduate.

I got totally excited about that early on. I probably wanted to write that history book as a later scholar. But I'd learned the textbook version. I knew how to do the simple calculation show that quantum mechanics predicts violations. But to really, really get into the guts of Bell's inequality and the loopholes and all the thought that people put in on the theory side, again, that's

a very advanced developed body of knowledge that's sort of newly relevant in ways that I had no inkling of when I was an undergraduate for things like quantum encryption and quantum information technologies more generally. So it turns out we often now use bell tests to confirm the security of a quantum encrypted channel. Well, okay, what if your bell test is susceptible to one of these loopholes? Either because nature behaves differently than we thought or because you actually have to worry about

a person who's actually trying to hack your system and fool you. Identifying these loopholes for bell tests took on an importance that I had no inkling of ahead of time for many areas of physics I find really exciting and beautiful that gave me a chance to learn more than just a little enough to write a couple of good papers on it at least,

in partnership with friends and colleagues for whom that was more their daily stuff. So I don't wanna say like I'm a gadfly and I had some of those tendencies, but it was an opportunity to go learn pretty hard stuff, certainly hard for me, that was well beyond what I've been trained in through all my years as an undergraduate PhD student

and even as a young faculty member. And that, the joy of a new learning curve is pretty amazing. This stuff is cool and hard and I think I can do it, but let me try again." That feeling of making sure I'm not just getting stuck, doing what now feels familiar. In hindsight, I've been able to do that a couple of times over in my career,

and I find that just really, really important for my own, not just my own curiosity, but I feel like I think I know more about the world. I think I have tools with which to try to explore questions I wouldn't have even posed before and that feels really very exciting. Okay. So let's get to this new learning curve. Good.

New as in past decade or so. Yeah. With primordial black holes. Right. So please tell me how primordial black holes connect to other areas of physics. Well, many different ways. Especially the unexpected ways.

Yeah, good. Let's start with the more expected ways. And so that's how I got into them in the first place, or was more expected for me at least. As we talked about before, most of my physics training had been on early universe cosmology, topics like cosmic inflation. I was already well-practiced at calculating the perturbation spectra. What's the degree of primordial lumpiness,

to speak a little loosely, that we would expect from various models of inflation, compare with observations of the cosmic micrograph. That was my bread and butter. I love it. I still love it. I find it amazing that we can do that. I began thinking my entree into Preminal Black Holes for me, it was familiar to many people by then, but what brought me into it was think about models of

inflation where they might have something else that happens near the end of inflation, beyond just a vanilla, what we often call slow roll toward the end, where you could sort of build a model, hopefully a well-motivated model, with ingredients that we think should be there anyway from fundamental high energy physics. And would those provide the kind of different dynamics toward the end of inflation that

would lead to this very large dramatic amplification of the fluctuations that could lead to black holes? So for me, that meant thinking about models of inflation that move well beyond the kind of single field toy models that are very familiar and very helpful, but ultimately really I think a cartoon and they don't fit super well, I think, with the better articulated ideas from whether they're coming down from string theory or any kind of UV complete ideas about, let's say, Planck scale physics, super gravity inspired or otherwise. So some of those ingredients include more than one scalar field, right? Even in the standard model, our beloved and exquisitely well tested standard model, there

are four scalar fields in the standard model. There's the Higgs field and the three Goldstone modes. At lower energies, high for us, but you know, like at the LHC and around that, we tend to go into unitary gauge. We talk not about Goldstone modes, but about sort of the massive vector bosons like the W's and Z's. We know ultimately those are really coming from, that is to say the masses are coming from these sort of eaten Goldstone modes. So what had been massless vector bosons become massive and they have three positions. That usual story is I think amazing. But the point is we often get away with dealing with one scalar field in the standard model, the Higgs field,

and we treat the Goldstones as polarization states of massive vectors, the W's and Z's. Fine, that works great. It's perfectly self-consistent. But at very high energies, unitary gauge is not renormalizable. And if we want to talk about energy scales that are below the Planck scale, but much closer to that than to, you know, kind of GEV or TEV scales, we're not doing LHC physics, then in the renormalizable gauges, the Goldstone mode stay in the spectrum. Just the standard model is

a multi-scalar field theory when described self-consistently at high energies. That's already cool. Then again, as you and I'm sure many guests on your show have emphasized, every known beyond standard model theory building introduces more and more scalar fields. Maybe it's an axi-verse, maybe they're modular, who knows what they are.

Sure. But there's no shortage of scalar fields once you go even beyond the standard model. And in the standard model, as I say, it's already a multi-scalar field framework. So one of the things that I find really important or helpful in thinking about inflation is to build models that have more than one scalar field. Since at very high energies and very early times, that seems relevant. That seems like a relevant ingredient in the spectrum. Okay, that's part one.

Another ingredient that I think is more often overlooked, but I think it's actually really important, are what are called non-minimal couplings between the scalar fields and the space-time curvature. That's to say, at the level of the action of direct coupling between the scalar fields and the spacetime curvature. That's to say, at the level of the action, a direct coupling between the scalar field and the Ritchie scalar that describes our spacetime curvature. These have been thought about for a long, long time in even classical GR.

They're required, they're induced by quantum loops, even if you don't put them in by hand. They show up from all kinds of compactification schemes. You are starting from some sort of UV physics that's beyond standard model. Another fairly generic ingredient would be these so-called non-minimal couplings. Well, if you want to be agnostic and go back to EFT, Effective Field Theory Review, that we've talked briefly about before, these are dimension four operators. How do you not include them in an EFT, right? So if you just start from writing down an EFT with all the self-consistent dimension

four operators, then you write those down, right? And then there are words you can put around whether they're motivated by this or that physics. So that means I think it's really important to be building realistic or at least more better motivated models of inflation with at least those two key ingredients. Multiple scalar fields, each with a nominal coupling to gravity, okay? That or better motivated models of inflation with at least those two key ingredients, multiple scalar fields, each with a nominal coupling to gravity. Okay, that suddenly is a playground

that's different from the kind of simple cartoon-like single field models of inflation that hopefully could help us connect better to kind of, to higher energy, potentially kind of UV physics. Okay, once we start doing that, one of the first papers I wrote on primitive black holes with a whole slew of wonderful colleagues and students,

was that automatically, without putting anything else in but those ingredients, you automatically start getting these kind of directions in the effect of potential. And your potential now is a multi-dimensional object. You can have phi 1, phi 2, and maybe more than those. There'll be directions for the evolution of that system that will

yield exactly the dynamics, exactly the dynamics that people had found in these single field constructions that will lead to a spike and promotable black holes. We didn't put those features in by hand. We need to look for them, thanks to these pioneering works on single field constructions of the effective potential for inflation. But they just fall out when you start from ingredients that I consider better motivated anyway. We even went to the work, my colleague Evan McDonough did most of this part, of showing you really have a self-consistent UV embedding. This really flows from a certain super gravity construction.

This is a, you know, a, not the, but a well-motivated model of the very early universe. And for free, we find these regions in which you should get this unusual behavior before the end of inflation, where you would expect a large amplification of the quantum fluctuations, inflation ends, those cross back ends of the Hubble radius. There's such localized over densities, bang, you never made a population of primal black holes. That's one example where, again, I hadn't worried very much about scalar potentials and some of the machinery of supergravity. I got to learn enough to be able to participate in this paper and then help with other parts because I've been thinking a lot about nominal couplings and multiple. Okay, so that's step one, one way that primordial black holes

has led me to think about other areas of fundamental physics, in this case closest to home base for me, but even that a bit of a stretch, right? Sorry, before you get to part two. Sure. Okay, you mentioned that with one scalar field there's some prediction or there's some model, and then you said you were able to get to it from multiple scalar fields, and you felt like these multiple scalar fields were more well-motivated. But it sounds like you're introducing more ingredients. Why would you want to recover something that someone can

explain with one when you're explaining with five? How is that better? So wonderful question. Good. Partly because to get this to work, to actually make primordial back holes with single field models, you require an awful lot of fine tuning, which is not what cosmologists like typically, and in particular what that meant in practical terms. For each of these very clever single-field kind of worked examples, proofs of principle, you get as extreme an amplification

of the perturbations as you need to cross a threshold to induce gravitational collapse. So you really need a very large amplification of these fluctuations. To get that, you had to have at least one dimensionless constant, one parameter in your Lagrangian, tuned to some really uncomfortable degree, like six or seven decimal places. So that doesn't seem like that's just going to happen on its own. So part of what got us excited is we were finding with these multi-field constructions you get these dynamics while reducing considerably the amount of fine-tuning of any given parameter. So one motivation was we think it's more natural anyway to think the universe was filled with multiple scalar fields,

nominal couplings, and then we started finding that we actually use fewer parameters than predictions. I'll come back to that in a second. We're not overfitting. We need to fit eight numbers to percent level accuracy with five free parameters. And we don't want to tune any parameter to such an extreme degree. So suddenly, you know, we're in a different regime than a single field construction, which can do it in principle, but what looks, again, we might say unexpected or unnatural or fine-tuned. And that's not knocking those papers. Because all those cool constructions

existed, we knew kind of what to look for in our expanded toolbox. That leads actually to the max point. Another thing, my first and so far only Markov chain Monte Carlo simulation, which is bread and butter across so much physics, It's an amazing tool, right? My first one that I got to really do with, again, with the help of amazing set of co-authors, was to subject these multi-field models with a couple of free parameters, throw them into an MCMC, let lots and lots of these so-called walkers in the computer,

solve the equations over and over again, compare the predictions with a very precise body of observational data, and then find what a region of parameter space where this works. Do you just get lucky once, or is there a kind of trade-off between parameter generacies? Which is another way of asking how likely or unlikely do we think this is, and getting toward a kind of Bayesian, it's not quite formally Bayesian, but in terms of kind of explanatory power, you don't have to get lucky with all your parameters lining up once you in fact see trade-offs and kind of blobs in these corner plots where you would match all the high precision measurements of the cosmic microwave background

and make black holes and the masses would be right for dark matter down, down, down, down the list with a handful of ingredients ingredients fewer than you're trying to match, no one of it has to be pushed so extremely into a corner. For me again, growth opportunity to put it mildly, incredibly powerful techniques like MCMC, and that's still within my own wheelhouse of early universe cosmology, multiple field, nominal couplings, dynamics, spectrum of perturbations, stuff that otherwise I knew about, I had written about.

The next one then was to ask, well, okay, these black holes would form at a very particular moment in cosmic history. If they're going to be much or all of the dark matter today, then we know for reasons again I'm happy to talk about, there's a window within which their masses have to land. If they're too big, they're already ruled out for being all of dark matter, maybe they're a percent, but not all of dark matter. If they're too small, they would essentially Hawking evaporator rating, they can't be around today, but we need dark matter around today.

So there's a window about six orders of magnitude in mass within which, at least as of what we know today from the various constraints, all of dark matter in principle could be accounted for based on a population of these primordial black holes may be formed from these inflationary perturbations and their masses have to fit within a box. And the box is about 10 billion times smaller than a solar mass and below.

So you go from 10 to the minus 10 solar masses down another 600 magnitude from there to fit within that box. Okay, turns out this direct collapse I mentioned, the way that you form black holes not because stars form and they die, but directly from the collapse of primordial perturbations. The mass of the resulting black hole, the mass of these PBHs if they formed, is really a clock. It tells you how large was the Hubble sphere at the time those black holes formed. These black holes formed by swallowing most, but not all,

of the mass enclosed within the Hubble radius at the time that they form, and that had been identified, again, many, many years earlier. And we know how the mass within the Hubble radius evolves over time, because we know how the Hubble radius evolves over time. That's just saying how, at what rate was spacetime stretching after inflation.

And we have very good checks on that. So suddenly the mass of the resulting black holes, if they're going to account for most or all of dark matter today, tells us not only what mass they have but when they must have formed. Because they had to be an appropriate fraction of a Hubble mass, that's a moving target, bang! Now we know the clock when they had to happen. That's pretty cool. Well, it turns out to be all of dark matter, these black holes had to form really, really early after the end of inflation, but long before Big Bang Nucleosynthesis, which starts at around one second, long before the electroweak phase transition,

which happens around 10 to the minus 12 seconds, long, long before the QCD confinement transition, which happens around 10 to the minus five seconds. You know, we have these benchmarks in cosmic history that span the first second. It's amazing. We can slice and dice the first second with such, you know, precision. And these black holes formed way before each of those kind of milestones. So that means that the universe must have been filled with a very hot plasma of unconfined quarks and gluons. They're not yet bound into color neutral states.

They've not yet undergone the QCD confinement transition. Standard model SU3, beautiful QCD, at very high temperatures, it's an unconfined theory and this plays into things like so-called asymptotic freedom and so on. They're weakly coupled to each other. They're not bound in color neutral states. The universe is color neutral. If you coarse grain over the entire whole Hubble sphere or even need to go even less distant than that, there's a balance among the

color charges in any region of space, but they're not bound into kind of hydronic states the way they would be after around 10 to the minus fifth of a second. These black holes are forming by scooping up shovelfuls, tiny shovelfuls, of unconfined quarks and gluons of QCD-charged, color-charged matter. Yeah. Wait, I'm not understanding. Are you saying that there are color perturbations that from afar it looks white, but then when you look closely, there is some little intensities of red or green or what have you? You nailed it.

That's exactly right, Kurt. And I should say this thing, I didn't know that in any detail before the primordial black holes led me to this topic. Experts in QCD have known this for a long time. I hadn't worked in QCD. Here's another example where there's a body of expertise, including some of my very dear colleagues right at MIT.

I could literally just walk down the hallway and say, hey, wait a minute, what should I be reading? What's the review article? Can you answer these questions? A lot of help locally, to be sure. But there's quite a lot I was led to ask about what's called the quark-gluon plasma because I was following the black holes. They had to form really right then in time.

The universe was filled with a certain kind of stuff, unconfined quarks and gluons. Go. Right now, that's my son with an amazing partner, a colleague, a PhD student in fact, Elba Alonso-Monsalve with whom I was working on this part. So the idea is, as it's been known for many years, just as you say, at very high temperatures, the quark-gluon plasma is net color neutral, but has color charge fluctuations

on a typical length scale. Imagine anything, it's probably not gonna be perfectly pristine along arbitrarily short distances. And that's true for the distribution of charges in a plasma. Very similar to even a classical electromagnetic plasma, as it turns out. It needn't have been that way.

But partly because we're in this kind of asymptotic freedom regime, it behaves, there are non-abelian corrections but it's a lot like an abelian or like a U1 E&M plasma, which had been worked out by the wizards of this area over the course of the 1980s and 90s and refined since then. So it was there for me and Elba and others to begin to dig into. So the community of QCD experts, very high temperature QCD, this includes both theoretical work and now very fancy lattice simulations, QCD lattice simulations.

They were able to show within a static Minkowski space, that's all they had to worry about, that there are these color charge fluctuations in the plasma and it's set by something called the Debye screening length, which depends on the temperature of the plasma. Dubai? Debye, like Peter Debye, the Dutch physicist from the 20s and 30s.

So Debye had been working this out for electromagnetic, classical electromagnetic plasmas in the laboratory as a theorist. And then this became a very well-known body of work to plasma physicists and so on. And then the QCD folks realized it's remarkably parallel to study this, in other words, a very exotic system of very high temperature quarks and gluons. So any charge you might measure on any given charged particle is screened by the screening medium. The degree of screening is set by

this characteristic length scale, the Debye screening length. That depends on the gauge charge, the dimensionless charge, and the temperature of the plasma essentially, and the gluons, the number of quark flavors. But basically, it's a number times the temperature. The length goes as inverse temperature.

So that had been worked out in the Calc. The first thing Elba and I then had to do was say, well, we don't want to apply it to a flat space time or a static one. We want to apply it to all the glories of a bending, warping curve background because this is an early universe, it's expanding rapidly, and we want to study it near a black hole. So you can have really significant space-time curvature. It should be not very much like Minkowski.

So the first thing that Elba and I had to do then was learn as much as we could about the Minkowski space treatments of effective field theories, of very high temperature quarks and gluons, amazing beautiful stuff that had been worked out by many people. And then some very nice pedagogical review articles, plus my friends down the hall, we could do it. We could dig in and do it with work, but we could do it. And then apply that to this scenario I've been talking about.

What happens if a bunch of black holes start forming amidst that kind of medium, if that's the fluid that undergoes gravitational collapse. Okay, sorry, just as a point here of clarification. So when primordial black holes are forming due to perturbations, these are matter perturbations. They're not just, so in Einstein field equations you have something related to the metric and curvature on one side and the matter on the other. So you could conceivably have perturbations of just the metric which produce black holes, but you could also have matter which sources the metric. So which one is

it? Both. Because it's a great point, Kurt. What we have to do is work with gauge and variant quantities. And so what you just described is every one of the metric perturbations we write down, if we're not careful, is a gauge-dependent quantity. When we write down things like delta rho over rho, that's gauge-dependent as well. So on both the geometry side and the matter source side, if we're not working with gauge invariant quantities, we're very likely to fool ourselves with gauge artifacts.

So again, very smart people decades ago, often in the context of inflationary cosmology, worked out a whole series of gauge invariant combinations, linear combinations, of a kind of metric perturbation in a certain parameterization and a measure of a kind of delta row. And we work with these gauge invariant curvature perturbations as an example. There are Bardeen potentials, many of these that have stood the test of time. So what we're doing is essentially the answer is a yes to your question. You linearize Einstein's field equations,

work to linear order in these perturbations, and then only work in these gauge and variant combinations, track the evolution of those. So we're confident we're not fooling ourselves with a gauge or artifact, so to speak, from either side. So in that sense, these really are combinations of perturbations in the fluid and perturbations of the metric and you work with a linear combination.

So that's what we do. And so then we could, again, work in the language of things like Debye screening length and look at the distribution of color charge in this roiling hot plasma, which is not uniform on short length scales. Now, it turns out these black holes form from a really amazing process called critical collapse, which again was worked out 30 years ago, found by accident. The black holes form in a way that's a lot like a phase transition from StatMak.

It's like you have a order parameter and a universal scaling exponent. This is just another example where who would have thought that's going to show up here in stunning black holes. I think it's just gorgeous. This I think maybe Dirac would have liked. I don't know, but it seems so pristine and so beautiful and is very well tested now numerically and analytically. And so the idea is that you make actually a whole distribution of masses. The perturbations that cross back in, they form a mass distribution that has a very distinct peak.

You make most of your black holes with this characteristic mass, and that tells you your clock. That's why you know it had to happen now and not later to make the characteristic mass fit within your box. But then you make a power law tail, a small mass tail. So you make fewer and fewer black holes of smaller and smaller masses with a rate at which that falls off that again is controlled by properties of the fluid, by universal scaling exponent.

What that means in practice for this is that whenever these perturbations make a whole bunch of black holes that are, you know, of size M1, you make a few of them that could be exponentially smaller. That means they're forming from the collapse of correspondingly smaller regions of space. And what Elba and I realized is that you could have some black holes at the tail of that distribution that formed by swallowing up one charge correlated region of charge-correlated region of the plasma,

where practically everything that falls in has charged red-anti-green, it's mostly the gluons, or, you know, blue-anti-red, or whatever it's going to be. The gluons have their little charge vectors, you know, lining up in our SU3 space within a region set by the length of the Debye screening length. Most black holes you make swallow so many exponentially large number of these regions, they're color neutral as well. Even though they're so tiny on human scales, they're much larger than

the Debye screening length at the time of formation. So on this model, dark matter would be neutral, it would be electric neutral, it would be color neutral, dark matter would be inert and boring and acting gravitationally, which is what we want dark matter to do. But in the course of making most of the black holes there, as Elba and I trace it through very carefully, you'll make a smaller subpopulation that are smaller in mass, formed from the

collapse of correspondingly smaller regions or volumes of space, within which the color charged particles have their charges more or less aligned. So you make a subpopulation can be extremely highly charged under SU3. And that's amazing. So it's a novel state of matter. This is, you know, having like 10 to the 13 charge units sitting within a black hole on top of each other.

That's just amazing. So what do you do with that? So that leads to other questions about fundamental black hole physics, about how do you discharge such a highly charged QCD object, all kinds of questions and get loaded into the queue from following our nose to like black holes form early, the universe is filled with QCD plasma, wait a second, we better learn about QCD plasma. You see, that's another example, a long example of how much fun this is to track through,

follow the black holes, build on some stuff that's known very well, modify it for the situation, and then that leads to still new questions that even our friends in QCD had never had to broach before. Have you found any implications or violations of the no hair theorem or cosmic censorship or even BPS bounds? These are exactly the kinds of questions that now we're very eagerly

pouring into. On the observational side, one thing we're really interested in is could this lead to, could this whole scenario be constrained or ruled out, or could you find evidence for it? Because these very, very tiny mass black holes with tons of charge, they would eventually hawking evaporate. So I don't think those would be around anywhere near today. In fact, they probably wouldn't even last much more than a second. They form so early, a second would be a very long time to them.

But if you have enough of them hanging around within one region of space, as late as one second, then they'll be emitting very high energy exotic hydronic states into what's supposed to be a thermal equilibrium of protons and neutrons that start to undergo big bang nucleosynthesis. So if you start messing around with the relative balance of protons and neutrons, you start tweaking the proportion of this isotope of lithium compared to that isotope of boron. You start potentially messing around with Big Bang Nucleosynthesis, which is all about

relative abundances of very specific isotopes. Others had looked into this idea for particle-based dark matter. If you have other dark matter that could be decaying and shooting out energetic stuff at the onset of Big Bang Nuc can look for observational effects or constrain such scenarios. Again, we can build on a body of knowledge others have worked out, modify it. And here it wouldn't be new fundamental particles decaying like a kind of particle dark matter scheme,

as interesting as that might be. It will be perfectly standard model particles, only standard model ingredients in a perfectly well-described classical gravitational state, a black hole, but that becomes our new ingredient. That could then be undergoing late-stage Hawking evaporation on the order of one second after the Big Bang. Could that be either constrained, because it would have messed up too much BBN? Could it maybe help alleviate some of the tensions within BBN these days? Or what? So that's a more observational, it's a stretch, but something that we're curious to chase down. It also leads us to ask, much like you were asking as well, some really juicy fundamental

questions about black hole physics. You know, black hole physics, I just find, I love it, I love it. And it's filled with some extremely beautiful theorems, almost all of which concern a single black hole in vacuum in an asymptotically flat spacetime. As it turns out, there has never existed a single black hole in vacuum in asymptotically flat spacetime. So this entire, and I say that not to disparage

any of these amazing Nobel Prize winning results. But again, the kind of body of knowledge in black, in some areas of black hole physics, these pristine things like cosmic censorship, No-Hair Theorem and others, they have typically been worked out in scenarios that don't really mesh very well with our universe. Because black holes aren't alone

and space time isn't asymptotically flat. To be fair, in particle physics you often say that the particle is prepared in the infinite past and then it's in the infinite future when you detect it. That's right. And because that's a clear shortcoming, people worked out additional formulas like the so-called in-in or Schwinger-Kaldis form exactly to avoid that, right? Because sometimes that really, that assumption really fails. I'm not sure we've caught up yet

in the community on black hole physics to the equivalent of that, just to your point, Kurt, actually. What's the complementary formalism that might need to be developed to really answer these beautiful questions about hair, about no hair, or about cosmic censorship for scenarios in which a black hole is immersed in an active medium and no part of this necessarily becomes asymptotically flat space-time.

So that's something that it's really hard. Elba and I are working on that now. For example, we don't have any clear results, but that's the question. The question which we are led to by following black holes and the medium and there's only is that So again, it's a chance for me to learn Some really cool physics and see well, that's not quite what we need to answer these questions The questions are driven from following the black holes Let's dig in what can we do here?

Could we could we may try to contribute something here does very beautiful body knowledge? I have such great admiration for but these because these particular questions seem not really to have been broached or at least not really answered yet. Another great example. Okay, so let's call this theory inflationary PBH theory just to give it a moniker. Now some people may be wondering why is it that you're looking at the data today and saying okay what would dark matter, sorry, what would the properties of the primordial black holes have to be in order for them to be dark matter candidates? And then going to

your theory here, why wouldn't it be that your theory here could give you indications about this direction in order for you to invalidate it? And the response may be something like, well, this isn't a single theory, it's a whole space of theories. And so what we can do is we can say, okay, what constraints would there be on this inflationary PBH side if we were to think that primordial black holes are what comprise dark matter? And then, okay, so now you've carved out a little niche here. That's right. Then you could say, okay, given that these are the constraints on that theory, are there signatures that we would expect from such a

theory, from such a constrained theory that we can then look for? Maybe it's not primordial black holes, maybe it's something else, but it would be an indication that we're in this parameter space. Good. I think that's exactly right. And so just as you say, the inflationary models, what I've described are families of models and regions of parameter space, not a model and a set of parameters. So exactly you say, the nature of the distribution of black holes that result depends on the inflationary dynamics beforehand.

So we've shown kind of existence proofs that with, again, with ingredients I consider realistic, with much less fine tuning than before, then the single field one, we can produce populations of black holes with kind of gross characteristics that are in line with what we want. The peak mass fits within this box, let's say, and that have the tails and so on.

But it's not uniquely picking out. It's saying it's showing there are production mechanisms that are congruent with other ideas from very high-energy theory that don't require wishing come on a star to make sure everything worked out in somewhat unnatural way. But that's different saying this is the theory and the single prediction. It does depend on which member within that family, what region of parameter space go. Also, that still doesn't tell us, do black holes, do primordial black holes exist?

And if they do, are they 10 to the minus 8% of the dark matter abundance today or 100% or somewhere between. And so that really does call then require looking to the contemporary universe or the recent universe anyway, and doing what my friends and I call kind of direct detection of black hole dark matter. It's not quite direct, but local detection. We've had 50 years of very heroic and very expensive efforts to

detect particle-based dark matter with so far exactly zero compelling results. Again, I say that not to fault the people who work on this day in and day out. Sorry, wait, particle-based dark matter? Yeah. What if dark matter- Oh, wait, sorry, I thought you said particle-based black holes. No, no, right, no. So, no, what if dark matter is what most physicists,

I think, would still expect? Some new particle or maybe a whole dark sector, a family of particles. Great, we haven't found any of them in any so-called direct detection experiments. And again, that's not for lack of trying. And the sensitivities of the experiments have gotten outrageously better.

It's an amazing effort that so far has yielded exactly zero dark matter particle candidates. Okay. So what's the parallel to try to figure out whether dark matter consists all or mostly of these primordial black holes? So that set, again, some wonderful students and collaborators and I thinking about late universe local detection. It involves things like gravitational perturbations, gravitational waves, and also ejecta. These things really would be undergoing Hawking evaporation.

What can we look for for that? Now I get to play with experts in cosmic ray experimentation, energy cosmic ray detection, experts from LIGO and beyond on gravitational wave detection, and as well as you know my own more local gang were able to show that there would be a kind of predictable, countable number of these black holes that would fly through the solar system once every kind of 3 to 10 years, which is a pretty nice human scale cadence. They won't happen every month, but you don't have to wait 3,000 years to look for an example.

And when they do, if they have the mass of an asteroid, these so-called asteroid mass black holes, but the size of like a hydrogen atom, they're not going to hit anything. The odds of that are astronomically tiny. But you have a flyby and even a purely Newtonian impulse. You can do it more carefully with general relativity, but even Newtonian end body gets you most of the way there, remarkably. To say a black hole whizzes through at 200 kilometers a second at say five astronomical units.

That's a large enough sphere where you expect it to be a couple of these hanging around and going on these kind of joy rides. And suddenly we have a remarkably well-instrumented inner solar system to look for very tiny but indeed measurable perturbations to the motions of mundane objects we track all the time, like the planet Mars. I don't know about tracking planet Mars, there's another example. Follow the black holes and now I get to learn about some other cool stuff like ranging within the inner solar system. So I knew from actually some colleagues that astronomers had been doing laser ranging of the moon since the Apollo 11 mission, since 1969. So one of the first things the Apollo 11 astronauts

did was put up special reflectors on the moon, retro reflectors, and astronomers can shoot lasers to those reflectors and very carefully measure the return. So we know the Earth-Moon distance with an accuracy of about one millimeter. Hmm. That's a quarter of a million miles away, and we know the distance to the accuracy of one millimeter. That's astonishing. Because of 20 years of Mars, you know, orbiters, rovers, landers, and some very long baseline interferometry, a range of techniques, astronomers know the Earth-Mars distance

to an accuracy of about, on the order of 10 centimeters. Okay, that's much further than the moon. And if the error is on the order of 10s, 10 or 10s of centimeters, that's astonishing. So Mars is being tracked and the Earth-Mars distance is being, you know, kind of constantly calibrated even as both are moving. And that's fed into some extremely sophisticated solar system dynamics models, so-called ephemerides models, run by a groups around the world, where they're running sort of end-body simulations of 1.5 million objects

that they track in the solar system, not just like the planets and the sun. Lots and lots of moving parts and constantly benchmarking with the latest high-precision data from things like tracking of Mars. So suddenly, if a tiny little hypothetical black hole, that's part of the local dark matter density cuts through, sort of transects the solar system, it is likely to produce perturbations on

the motion of Mars that will exceed its otherwise very small error budget of where it's supposed to have been, you know, measurably and not arbitrarily long after the flyby. So that sort of thing where could we get better at detecting basically gravitational perturbations to well-tracked objects within the solar system as a beacon, as an indication that something was a perturber that flew by. Now, that effect depends really only on the mass of the perturber. They're very far away. You can do N-body, not worry about, you know, kind of tidal forces. And so what if it was just a mundane space rock of the same mass?

So, okay, well, what's the expected background for that? It turns out there are online databases maintained by groups like NASA, the Jet Propulsion Laboratory, which attract almost half a million near-Earth encounter objects. In fact, they attract anything that got within three astronomical units of any planet in the solar system for the last 100 years. Wow, that's a great database. They can do things like what's the inferred velocity and other orbital characteristics.

You can realize that the black hole path should be really disjoint from that entire distribution. So it doesn't prove it was a black hole if you see Mars wobble with a certain time signature, but it would be highly unlikely to have been any of the known and well tracked objects for the last hundred years. That sets a baseline expectation. It's probably not, is not likely to merely have been a space rock, mundane space rock. And then we can get better at reconstructing the path of the protuberance based on the time series of the perturbations of the object we track. Right.

You infer the current location of that object. And again, astronomers have gotten very good at finding very small space rocks in the solar system, which have much smaller mass than these black holes would, but are not the spatial size of black hole. They're rocks, so they could be tens of meters to kilometers across, and they'd be made of rock. So they'd have an albedo that'd be typically trackable even optically.

Again, it's not proof that we found that the lack of such a visual component means it was a black hole, but we have ways of saying there's an unusual wobble and no clear visual component. That's at least increasing the odds that there's a black hole. Then we can go a few steps further. If it's really a black hole, then some fraction of those would be undergoing Hawking emission today. The black holes at the smaller end of this allowed range of masses for which they could still

be all or most of dark matter should mostly be quiescent. There would be highly inefficient Hawking emitters. And in fact, that's where that bound comes from. If they were efficient Hawking emitters, if they were later in their lifespan and a smaller mass, then we'd be a washing cosmic rays that we don't measure. So you can actually strain the fraction that would be kind of late Hawking emitter black holes. Nonetheless, there are extended mass functions. If most of the black holes are essentially quiescent in terms of Hawking radiation today, there's a distribution. Some of those would necessarily be smaller masses today and those would be a little further along in their kind of evaporation life cycle.

So, what are the odds that you have a gravitational perturbation and, you know, a certain, say, positron excess that would be consistent with Hawking emission from your perturber, things like that. And that would certainly not have come from, you know, a mundane passing space rock. So, that lets me play with some amazing colleagues with things like, what would it really take to detect excess positrons? What really would be the time series signature for that? So we're not just fooling ourselves. Is that visible from existing experiments?

Should we be able to propose building new ones? So suddenly it's about like CubeSats and very sensitive clocks and laser ranging. I didn't do that in grad school. That's like amazing. With experts who know what they're doing, so I'm not hopefully not just fooling myself, but a chance to build new collaborations, learn with new partners and ask these questions because we're led by trying to follow

this hypothesis of primordial black holes, which maybe they exist, maybe they're much or all of dark matter. How would we know? How would they show up in our cosmic neighborhood? Not just could we imagine how they might have been produced very early in cosmic history. Okay, so the reasoning is that if the black hole was of the mass of an asteroid, but the size of a hydrogen atom, then it would bounce like a billiard ball would bounce off of

Mars and just perturb its where it would have been. What about Earth? Mars and just perturbates where it would have been. What about Earth? Well, we don't even consider direct impacts. The odds of the cross-section is so incredibly tiny. So the odds of it hitting Mars or the Earth or the Moon are essentially zero in the whole age of the universe. All we need is for this thing to have zipped by in otherwise empty space,

Okay. two astronomical, three astronomical units away from Mars, it's a large mass traveling fast. There's an impulse, even a Newtonian impulse, let alone a fancy relativistic one. There's a gravitational interaction at a distance, at an impact parameter that could be genuinely macroscopic, I mean, astronomical units. That alone is enough to make Mars wobble tens of centimeters sort of off course. It's a self-correcting, so the perturbations would damp over a long time period, but not so quickly that they wouldn't be visible from this very sensitive tracking. So we're not thinking about impacts per se, we're thinking about flybys, where the black hole just tootles on in its way at actually a fairly

rapid clip. But because it's now not gravitationally bound, it's not coplanar, it's not in the ecliptic, it can cause very specific types of perturbations to visual objects we track very closely. I see. But what about the Earth? Why not if it passes by the Earth? Why are you focusing on Mars? The main reason is because that's a great question, Kurt.

We first thought about the Earth-Moon or we have, and other people have written a paper about this, a lovely paper, about both about Moon and also about the constellation of GPS satellites and related other systems. So, you know, for GPS to work, the people need to know the instantaneous location of those satellites to sort of centimeter or tens of centimeter accuracy. That's so that we know where we are on Earth from when we get those signals. So you have 30 plus GPS satellites that are well tracked. Here's the reason why that I think is more complicated. We thought we had a cleaner signature.

Because if it's that close to either GPS satellites or even to the moon, then you really have to worry about tidal effects, about local deformations, there's not end bodies, not point masses at a distance. Whereas if a black hole passes far away from Mars, those are two point particle-like interactions, and Earth-Mars system is much more reliably a two-point system. Essentially, there are highly subdominant tidal effects

between Earth and Mars because they're so far away. So tidal effects fall off more rapidly than one over r squared. Whereas for Earth-Moon, the Moon makes our tides. We have to worry about tidal effects all the time. So to back out a clean signature of a wobble and then reconstruct the path of perturber, the Earth-Mars system is in that sense more clean. There are fewer confounding effects, gravitational or otherwise, that we that we'd have to worry about. Wasn't there a recent high-energy strave neutrino? There was and and that was another great fun example. So there there are a couple of these. There's one very high-energy one. The record holder so far was announced I

think only in February of this year, recently. It had been detected roughly two years earlier and the collaboration wrote up the paper recently. But there have been other ones of lower than that but still really high energies found by things like the IceCube collaboration, which is in the South Pole. So IceCube has been operating for approximately 15 years. They've detected many, many neutrinos from outer space. For a small number of them so far, they've been able to identify a point-like astrophysical source, a so-called blazar. There's something that went bang in the sky,

and all kinds of stuff came out, high energy electromagnetic radiation, neutrinos, and so on. And they could identify the path and timing. So far of that set, there are about six high energy neutrinos that IceCube has found that are so much higher energy than those, about a million GeV. So it's called a PEV,

10 to the sixth GeV and above, not one of which has so far been identified with any known astrophysical source, point-like or otherwise. Then the neutrino I think you have in mind, Kirk, is about 100 times more energetic than that. So it was about 220 PEV, so 10 to the 8 GEV. That was detected by a different collaboration,

a KM3Net collaboration, which operates an enormous neutrino detector within the Mediterranean Ocean. So large, large, large cross-section. So again, with a terrific PhD student, Alexandra Clipful, she and I realized that again, if this hypothesis has any legs at all, the dark matter consists all or in part of primordial black holes. Again, critically, the black holes form

with some non-trivial mass distribution. Most come out with one mass, but there is a small subpopulation of smaller ones. Those smaller ones, some of them would survive to this day and be actively Hawking emitting. Our understanding, at least as of now, of Hawking emission, there really is a runaway process. That the black hole takes a long, long,

long, long time, emits hardly anything at all, gently loses mass and then falls off a cliff. So that in the last fraction of a second of the black hole's lifetime, it'll be emitting all kinds of extraordinarily high energy particles, all the standard model degrees of freedom, and if there exists any beyond standard model degrees

of freedom, at energies that in principle could approach the Planck scale. And then you can calculate the flux, how many particles per energy come out from these exploding black holes. And in fact, you get very few particles, if any, at the Planck scale because the black hole is so short-lived by that point. It's such a short lifetime, it's just the countable rates are few. But you'll get a countable number of particles coming out with

energy on the order of 100 PEV. If a black hole is going to that last death rattle of Hawking evaporation, at some distance like 300 astronomical units away from us, doesn't have to be right next to us. So it can be a large volume which has some likelihood to have happened. So what Alexander and I show is that this is actually perfectly likely to have happened on the order of one time in the last 15 years since these detectors have been in operation. If we consider a volume of space as the order of 300 or so astronomical units away, that's larger than the solar system. And if this is a kind of straggler from the small

mass tail of otherwise ordinary dark matter that consists of black holes, meaning this would be a black hole that formed in the universe down that mass tail, it formed with a smaller mass than typical, it was further along its evaporation lifetime now, and you have a not unreasonable likelihood putting in a realistic form for the mass distribution of formation, carefully evolving that forward with careful numerics, and then saying, here's the size of my box, 300 AU, what are the odds? And the odds are pretty good. Then you have about one of these very, very high energy

events every 10 to 15 years, which is at least consistent. That's not proving that that's the origin of this neutrino. But it's showing a really, I think, lovely congruence. The pieces really fit. That's not proof that's the source. But there are not other, to my mind, very well understood sources that are competing with that explanation. There are lots of papers coming out on this.

It's an incredibly intriguing event, this extremely high energy neutrino. To my understanding, there is not as yet any, let's say more straightforward astrophysical explanation that's been put forward. Maybe there will ultimately be one, we don't know. But again, it was at least a self-consistent to say, if we really think black holes are out there, and they're really all or most of dark matter, you can have some straggler rare events.

Let's be open to those as well. One in every 15 years that it hits Earth or that it'll hit the detectors on Earth? Good. We split the difference. That would hit a region on Earth that's larger than any given detector, but one such that wouldn't hit both detectors. One might say, well, why didn't IceCube also see it or see others' ones from that explosion? So I think the surface area we took, you'd have to look up in the paper.

It's a little bit bigger than just KM3 net, but a tiny fraction of the surface of the Earth. So we didn't give ourselves the entire Earth as our target. We thought that wouldn't be realistic. So we wanted a cross section where it's reasonable to hit that, like hit the meta training, but not the South Pole. I mean, that kind of thing. So we've touched on a variety of topics, just with primordial black holes.

There's the standard model. There's what's beyond the standard model. There's the Big Bang and cosmology. There's inter-solar system physics, which I didn't even know, I didn't think about prior to this conversation. There's various experimental apparatus and then experimental thinking. So to tie this all up, what are you looking forward to? What's next for you? And I would also like to get to advice for students who similarly want to tackle everything.

Let me start with the last one first. Don't tackle everything at once. I think, boy, that's a recipe for frustration. I don't know if it was sufficiently clear as I was narrating and rambling along. I've had such an amount of fun tackling these projects not exactly one at a time, but one flows to the other.

That meant that I've been very lucky, extremely lucky to be able to take a time, but one kind of flows to the other. So that meant that I've been very lucky, extremely lucky to be able to take the time, learn what I can on my own, critically learn with groups of colleagues and students, project by project, because each of these kind of requires and deserves just a lot of close focus. So it's not like we're going to just do all at once, that's just a recipe for heartbreak

and frustration. So partly it's be open to some really fun questions, but also recognize that each of these is gonna deserve and require the really sitting still, really sitting with these and going through all the kind of emotional cycles of this is brilliant, it's terrible, it's brilliant, it's terrible, I did it, I lost it, you know, all that kind of ride day to day, week by week, month by month. It's not easy. For easy might not be frankly so fun.

But the happier flip side of that is, you know, we live in a really amazing universe that's complicated with lots of moving parts. And lots of people know a lot about aspects of that, including how we can learn more, like an instrumentation and experimentation, as well as theoretical techniques. And so none of us does this alone. None of us should try to do this alone. Not only is that lonely, it's just like, you know, I just, my own horizons have been so much broadly expanded by the opportunity to work with experts and students who are becoming experts on a range of things. And I have to know enough to make sure I'm not fooling myself or my colleagues.

I have to get up, do the work, steep learning curve, joy with learning new things, but I'm not doing it kind of on my own each time by any measure. Nor do I think would that be fun or intellectually satisfying. So with these antagonist experiments, the fact that Anton Zeilinger thought this would be interesting and fun, that's what made these possible. That's what made these feasible. And it can play that game over and over again, each of the projects that we've talked about here. So for advice for students, I'd say, don't shy from really fun questions and then get the help that you need, I'd say team up with people who don't know those answers either, but probably have other tools that are already quite familiar to them that might be new to you and you'll know things they don't know.

And that sounds very kind of sweet sweet and we shall join hands, but I just have really experienced that over and over again, that there's a way of putting things together. The joy is better with people in the car. Yeah, that's right. Because I didn't know where I was going and they knew how to fix the spare or whatever the analogy would be. That's right. So how do you find people to come into your car? Yes.

Drive by a gas station and then say, hey, hop in. I mean, one way is to get a gig at a university. That's hard to beat that. I mean, so I'm immersed in a community of colleagues and scholars from undergraduates through PhD students, post-docs, and fellow faculty across the Institute, across MIT, and of course, beyond. So partly, there's something I think really magical,

I really mean that, about the academic research community. It's very precious and it took a long time to build it into its current state. But one way is we have people coming who want to ask similar questions all the time, and that helps to get people in the car, so to speak. Another way I think is to, I don't know how to say it, if you can, kind of have fun with it, because that hopefully sets a tone to saying, you know, let's take this drive together.

The questions are meaningful, we'll all learn things we don't know now, they're probably gonna be interesting to other people we don't even know yet. You know, and so being open to collaboration and to ask them questions that sound hard and interesting. Getting that balance can be tricky, but where you and other people are going to learn something,

maybe it doesn't pan out and you'll learn from that too. Maybe the effect goes away and that's cool. No one saw that coming, whatever it might be. What I'm interested in now, I think we touched on a few of them. One is really trying to learn more and dig in more on this kind of fundamental black hole physics, things like cosmic censorship,

which is a hot, hot topic and many experts who know tons about that. And I'm trying to, you know, again with Alba and some other colleagues, trying to say what can we contribute to that? What questions were they maybe not focused on about things like, say, what could we contribute to that? What questions were they maybe not focused on about things like, say, a black hole in a medium or a dynamical space

or other things that are on some of their minds, but what can we bring to that as well? And I'll learn a ton from that. I already have from even the efforts so far. So one of those is, again, very kind of abstract, theoretical, mathematical. And I'm having a great time with it. And the other is going more like what we're talking about near the end.

Instead of only relying on Mars being perturbed or not, what would it take really, really, is it feasible to build purpose-built, inexpensive satellites and put them where we want them? So could we optimize their orbits so they're not only in the ecliptic? Could we instrument them not only so we can arrange them, but could they have little inexpensive cosmic ray detectors? How would we do with that? These CubeSats. Right, like CubeSats and so on. That's right. What would it take to do a fleet? And could we use them to detect other exotic gravitational effects like a gravitational time delay? If we really know the fleet of them and they have really disciplined clocks, could we do other beautiful gravitational tests that would be

consistent with a black hole but not a typical asteroid? So suddenly, it invites conversations with people who know a lot about things like CubeSats, about very fancy clocks, about ranging, about computer simulations. Again, I know very little about each of those, but I sure look forward to learning more. It's in the service of what's the world made of and how might we know? And are these primordial black holes really here or are they merely a very pretty idea?

And maybe they're just that, and we've already learned a lot from chasing the pretty idea, but wouldn't it be something if they're also part of our universe? David, what drives you other than curiosity? A couple things. Curiosity is a big one. Another one, it really is trying to build a space where younger folks can learn a lot of stuff and take those ideas and those skills where their imagination takes them. And I take the role of being in education really seriously.

It's a great privilege and a responsibility. And again, that sounds very, I don't know, it sounds like it's on a hallmark greeting card, but I really mean it. I mean, to be able to watch a younger person who has questions like I had when I was, you know, a young person once back in the day, and watch them develop incredible skills,

incredible skills, but also help them foster their own, and maintain their own curiosity. What do they want to do with their skills? It doesn't have to be in academia. What do they want to do in the world? What do they want to do to do and make and build and learn in whatever setting that excites them. You know, that's pretty amazing. And being surrounded by people from undergraduate through

PhD, through postdoc, who are excited and eager young junior faculty to really see people. Let me share one last story. One of my very dear colleagues on this black hole journey is Ray Weiss, who's merely 93 years old, Nobel Prize winner who helped dream up, design, build, and lead the LIGO project for decades, that first successfully detected gravitational waves almost exactly 10 years ago. Ray is just a treasure. He's a treasure. He's incredibly humble and down to earth. I have these meetings with a first year undergraduate, and Ray Weiss, and every stage in between. If that's not a source of inspiration

to get me out of bed in the morning, I don't know what would be. That's the community of people who are still learning. Ray is still learning stuff, and asking us questions, and we're learning from him too. That's a pretty great gig, to be involved in that kind of journey together to try to say, again, what's the world made of and how would we know?

Professor, thank you for spending so much time with me. Kurt, it was really a pleasure. Thanks so much for having me on. I appreciate it. All right. That was fun. I've received several messages, emails, and comments from professors saying that they recommend theories of everything to their students, and that's fantastic.

If you're a professor or a lecturer and there's a particular standout episode that your students can benefit from, please do share. And as always, feel free to contact me. New update. Started a sub stack. Writings on there are currently about language and ill-defined concepts concepts as well as some other

mathematical details. Much more being written there. This is content that isn't anywhere else. It's not on theories of everything. It's not on Patreon. Also, full transcripts will be placed there at some point in the future. Several people ask me, hey Kurt, you've spoken to so many people in the fields of theoretical physics, philosophy, and consciousness.

What are your thoughts? While I remain impartial in interviews, this substack is a way to peer into my present deliberations on these topics. Also, thank you to our partner, The Economist. Firstly, thank you for watching, thank you for listening. If you haven't subscribed or clicked that like button, now is the time to do so. Why? Because each subscribe, each like helps YouTube push this content to more people like yourself,

plus it helps out Kurt directly, aka me. I also found out last year that external links count plenty toward the algorithm, which means that whenever you share on Twitter, say on Facebook or even on Reddit, etc., it shows YouTube, hey, people are talking about this content outside of YouTube, which in turn greatly aids the distribution on YouTube. Thirdly, you should know this podcast is on iTunes, it's on Spotify, it's on all of the audio platforms. All you have to do is type in theories of everything and you'll find it.

Personally, I gain from rewatching lectures and podcasts. I also read in the comments that hey, toll listeners also gain from replaying. So how about instead you re-listen on those platforms like iTunes, Spotify, Google Podcasts, whichever podcast catcher you use. And finally, if you'd like to support more conversations like this, more content like this, then do consider visiting patreon.com slash Kurt Jaimungal and donating with whatever you like. There's also PayPal, there's also crypto, there's also just joining on YouTube. Again, keep in mind it's support from the sponsors and you that allow me to

work on toe full time. You also get early access to ad free episodes, whether it's audio or video, it's audio in the case of Patreon, video in the case of YouTube. For instance, this episode that you're listening to right now was released a few days earlier. Every dollar helps far more than you think. Either way, your viewership is generosity enough. Thank you so much. you