A reading list is a catalogue. This page is something different — it is an account of papers and books that actually changed how I think, or that I found so extraordinary that I kept returning to them long after I first read them. Some are foundational texts that every physicist knows. Some are things I stumbled onto at the right moment. All of them are things I can point to and say: this affected what I do.

The organisation is personal, not chronological or by subject. The papers and books are grouped by what they did for me: things that started the obsession, things that changed what I understood about the universe, papers that revealed the deep structure of the subject, books that changed how I read everything else, and a section on the pure aesthetic pleasure of exceptionally well-written science.

Note: Although this page is currently placed under the Astronomy section, it primarily serves as an archive of my earlier collection of reading materials.

As a result, some entries may not be directly related to astronomy. However, each of them had a significant impact on my intellectual development, which is why they have been retained here.

Most of these were originally part of the reading section on my earlier website, and they are preserved here for continuity.

More structured and topic-specific reading subsections are now available across different sections of the site.

I — The Books That Started Everything

Carl Sagan — Cosmos (1980)

I have already written about the role this book played in starting my interest in astronomy. But I want to say something more precise about what it is, because “Sagan’s Cosmos” has become a generic cultural reference point, and the specific thing that makes it extraordinary gets lost in that generality.

What Sagan does, better than anyone else I have read, is hold the scientific and the human simultaneously. He does not present the universe as a collection of facts to be admired from a distance. He presents it as something we are embedded in — that our atoms were forged in stellar interiors, that our planet is a tiny feature of a universe 13.8 billion years old, that the Voyager probes are already beyond the solar system and will drift between the stars for longer than civilisation has existed. And he does all of this while maintaining complete scientific accuracy.

The chapter “The Lives of the Stars” is where I first understood stellar nucleosynthesis — that every element heavier than hydrogen was made inside a star. The chapter on Kepler and the music of the spheres is where I first understood what it meant to look for mathematical structure in physical phenomena. The final chapter, “Who Speaks for Earth?”, is about the responsibility of beings who understand the universe to act with corresponding seriousness. That this was written at the height of the Cold War, aimed at the possibility of nuclear annihilation, does not diminish its relevance.

I have read Cosmos four times. It means something different each time.

Richard Feynman — The Character of Physical Law (1965)

Feynman’s Messenger Lectures at Cornell, delivered in 1964 and broadcast by the BBC. Seven lectures, each about an hour. Available in full on YouTube (the original recordings) and as a short book.

The first lecture, on the law of gravitation as an example of physical law, contains what I think is the best statement of what theoretical physics is doing: it is looking for the rules of the game, not the outcome of any particular match. The force between two masses separated by a distance is $F = Gm_1 m_2 / r^2$. That is a rule. What made Newton extraordinary was not that he measured the force between any particular pair of objects, but that he identified the rule that all objects obey.

The seventh lecture, “Seeking New Laws,” is the one I have thought about most. Feynman describes what it actually feels like to be a physicist working at the frontier — the uncertainty, the guessing, the way you have to be simultaneously rigorous and speculative, the fact that the rules of the game cannot be derived from anything more fundamental but are simply the way the universe is. The sentence that stayed with me: “When you get as fundamental as you can go… you can almost see the problem is the right one to be working on.”

Freeman Dyson — Disturbing the Universe (1979)

Dyson’s autobiography, covering his scientific life from post-war Cambridge through the IAS at Princeton, Project Orion (a spacecraft propelled by nuclear explosions), and his later speculations on life in the far future. Dyson was one of the architects of QED — he proved the equivalence of Feynman’s diagrams and Schwinger’s operator methods, and his formulation of the Dyson series is what modern QFT is built on.

What makes this book extraordinary is Dyson’s honesty about uncertainty — about not knowing whether a research direction will lead anywhere, about working on things because they are beautiful without knowing whether they are true. The description of sitting in a Greyhound bus for three days and nights and, by the time he arrived in Berkeley, having understood why Feynman’s diagrams worked — this is what mathematical insight looks like .

The chapter on Oppenheimer and the culture of post-war theoretical physics is the best account I have found of what that world was like: the extraordinary concentration of talent, the combination of camaraderie and competition, the awareness that physics had just changed the world and the uncertainty about what that meant.

II — Papers That Changed What I Understood

C.N. Yang and R.L. Mills — “Conservation of Isotopic Spin and Isotopic Gauge Invariance,” Physical Review 96 (1954)

The paper that introduced non-Abelian gauge theory. Yang and Mills asked: what if the gauge invariance of electromagnetism — the freedom to change the phase of the wavefunction independently at each point — were generalised to a non-Abelian symmetry group? What if you could rotate the proton and neutron into each other in isospin space, and the rotation could be different at different points in spacetime?

The answer required introducing gauge fields that self-interact — unlike the photon, which does not interact with itself, the Yang–Mills gauge bosons interact with each other because the gauge group is non-Abelian. This self-interaction is the source of the richness (and the difficulty) of Yang–Mills theory.

The paper is short — eight pages — and worth reading in its original form. The appendix, where Yang describes presenting the paper to Pauli (who had apparently worked on the same idea independently and had been stopped by the difficulty of the mass problem), is a remarkable document of scientific tension. Pauli kept interrupting to ask: “What is the mass of this field?” Yang did not have an answer. The mass problem — giving mass to the gauge bosons while preserving gauge invariance — is what the Stueckelberg mechanism, the Higgs mechanism, and my own research are ultimately about.

P.W. Anderson — “More Is Different,” Science 177 (1972)

Arguably the most important essay in physics philosophy of the 20th century, in four pages. Anderson argues against the reductionist program — against the idea that finding the laws of elementary particles tells you everything. He shows that at each level of organisation (elementary particles → nuclei → atoms → molecules → materials → living systems), genuinely new phenomena emerge that are not derivable in any practical sense from the level below. More is different.

This paper is the intellectual foundation of condensed matter physics as a discipline, but its implications are much broader. The mass gap problem is, in some sense, about emergence — about why strongly coupled Yang–Mills theory, which at high energies looks like a gas of massless gluons, at low energies produces massive hadrons. The emergent mass is not obvious from the Lagrangian. It emerges from the collective dynamics of the strongly coupled theory. Anderson’s paper is why I think the emergence perspective on the mass gap is not merely a philosophical preference but a physical one.

R. Penrose and S.W. Hawking — Singularity Theorems (1965–1970)

Two related sets of papers:

Penrose, “Gravitational Collapse and Space-Time Singularities,” Physical Review Letters 14 (1965) — The first singularity theorem. Penrose introduced the concept of a trapped surface — a surface from which no light can escape outward — and proved that any spacetime containing a trapped surface must contain a singularity (in the sense of geodesic incompleteness). The implication: black holes, if they form at all, must contain singularities. General relativity predicts its own breakdown.

Hawking and Penrose, “The Singularities of Gravitational Collapse and Cosmology,” Proceedings of the Royal Society A314 (1970) — The general singularity theorem. Under very general energy conditions (which are satisfied by all known classical matter), any spacetime satisfying the Einstein equations and containing either a trapped surface or a closed spatial section must contain a singularity. The Big Bang singularity follows from the same mathematics as the black hole singularity.

These papers are not just technically important — they changed the ontology of general relativity. Before Penrose and Hawking, singularities were thought to be artefacts of special symmetry assumptions (the Schwarzschild solution is perfectly spherical — real stars are not). After them, singularities are generic predictions of general relativity under very general conditions. The need for quantum gravity — for a theory that can describe what happens at the singularity — became unavoidable.

S.W. Hawking — “Particle Creation by Black Holes,” Communications in Mathematical Physics 43 (1975)

Hawking’s derivation that black holes emit thermal radiation at temperature $T = \hbar c^3 / (8\pi G M k_B)$ — the Hawking temperature. This is arguably the most important result connecting quantum mechanics and general relativity ever obtained. It shows that:

  • Black holes are thermodynamic objects with a temperature and an entropy ($S = A/4G\hbar$, where $A$ is the horizon area)
  • Black holes must eventually evaporate completely
  • The evaporation, if unitary, must encode all the information about the infalling matter in the outgoing radiation

The last point is the black hole information paradox, which has occupied theoretical physicists for fifty years. Hawking himself initially believed the information was genuinely destroyed; he later conceded this position. The paradox is still not fully resolved, though recent results on the Page curve and the island formula represent significant progress.

The paper is technically demanding (it uses Bogoliubov transformations to relate the vacuum state of a quantum field before and after gravitational collapse) but the physical argument is clear. The result is one of the most beautiful in all of physics: it connects the black hole area (geometry), the temperature (thermodynamics), and $\hbar$ (quantum mechanics) in a single formula.

Juan Maldacena — “The Large N Limit of Superconformal Field Theories and Supergravity,” International Journal of Theoretical Physics 38 (1999) — hep-th/9711200

The most cited paper in physics. Maldacena conjectured that string theory on $AdS_5 \times S^5$ is exactly dual to $\mathcal{N}=4$ super-Yang–Mills theory on the 4-dimensional boundary. A quantum gravity theory in the bulk is equivalent to a gauge theory (without gravity) on the boundary. This is the AdS/CFT correspondence.

The paper changed theoretical physics in the 1990s and has not stopped changing it. Its implications include:

  • Gauge/gravity duality as a tool for computing strongly coupled gauge theory results (including properties of the quark-gluon plasma)
  • The holographic principle as a precise statement rather than a philosophical speculation
  • A concrete example of how gravity (including black holes) can emerge from a non-gravitational theory
  • The most precise framework for studying the black hole information paradox

For my own work, AdS/CFT is important as the context in which the connection between entanglement and geometry is most concretely understood — the RT formula (Ryu–Takayanagi, 2006) gives the entanglement entropy of a boundary region as the area of the minimal bulk surface, connecting entanglement directly to geometry. QGET attempts to generalise this beyond the AdS/CFT setting.

Mark Van Raamsdonk — “Building up Spacetime with Quantum Entanglement,” General Relativity and Gravitation 42 (2010) — arXiv:1005.3035

A short paper — 13 pages — that gave me the conceptual foundation for the QGET framework. Van Raamsdonk argues, using the RT formula and the structure of the thermofield double state in AdS/CFT, that spacetime connectivity is equivalent to quantum entanglement: two regions of spacetime are connected if and only if their dual boundary descriptions are entangled. Disentangling the boundary theory causes the bulk spacetime to pull apart. The geometry of spacetime is held together by entanglement.

I read this paper in grade 9, after the first Yang–Mills paper. The connection between QGET’s entanglement matrix and Van Raamsdonk’s picture was the starting point for the QGET framework.

R. Bousso — “The Holographic Principle,” Reviews of Modern Physics 74 (2002) — hep-th/0203101

The comprehensive review of the holographic principle — the idea that the information content of a region of space is bounded by its surface area, not its volume (the Bekenstein-Hawking bound, $S \leq A/4G\hbar$). The implications for quantum gravity are profound: a fundamental theory of quantum gravity must be in some sense lower-dimensional than it appears. Bousso’s review synthesises ’t Hooft’s original holographic conjecture, Susskind’s development of it, and the AdS/CFT evidence. A review paper rather than an original result, but the best single document for understanding why holography matters.

Gross and Wilczek; Politzer — Asymptotic Freedom (1973)

Two papers published simultaneously in Physical Review Letters:

D.J. Gross and F. Wilczek, “Ultraviolet Behavior of Non-Abelian Gauge Theories,” Physical Review Letters 30 (1973) H.D. Politzer, “Reliable Perturbative Results for Strong Interactions?” Physical Review Letters 30 (1973)

These papers proved that non-Abelian gauge theories are asymptotically free: the coupling constant decreases at short distances (high energies) and increases at long distances (low energies). This explained why quarks inside protons behave as if they are free at high energies (as seen in deep inelastic scattering) while being confined at low energies. The result earned Gross, Wilczek, and Politzer the 2004 Nobel Prize in Physics.

For my own work, asymptotic freedom is fundamental: the one-loop beta function of the Stueckelberg-extended Yang–Mills theory preserves asymptotic freedom (the Stueckelberg sector does not contribute at one loop). Verifying that a proposed mass generation mechanism is consistent with asymptotic freedom is a necessary condition for taking it seriously.

’t Hooft and Veltman — Renormalisability of Non-Abelian Gauge Theories (1972)

G. ’t Hooft and M. Veltman, “Regularization and Renormalization of Gauge Fields,” Nuclear Physics B44 (1972)

The proof that non-Abelian gauge theories are renormalisable — that the infinities that arise in perturbation theory can be absorbed into a finite number of physical parameters. This was not obvious: the self-interacting nature of non-Abelian gauge fields produces new ultraviolet divergences at each loop order, and it was not clear that they could all be cancelled. ’t Hooft and Veltman proved they could, using dimensional regularisation (which they invented for this purpose). The result made the Standard Model viable.

’t Hooft was 25 when he proved this. The combination of mathematical power and physical insight in this paper is extraordinary. The dimensional regularisation technique — replacing the 4-dimensional space-time integrals with $d$-dimensional integrals and working in $d = 4 - \epsilon$ — is now the standard tool of loop calculations in QFT. It is what I used in computing the beta function of the Stueckelberg-extended theory.

III — Books That Changed How I Think

Roger Penrose — The Road to Reality (2004)

1,136 pages. A complete account of the mathematical and physical understanding of the universe, from the natural numbers through quantum field theory and string theory, written by one of the deepest mathematical physicists of the 20th century. Penrose does not simplify — he uses genuine mathematics throughout — but he builds everything from foundations, so the book is accessible to anyone willing to follow it carefully.

The parts I have returned to most: the chapter on Riemann surfaces and complex analysis (which gave me my first intuition for complex geometry), the chapter on fibre bundles (which I read alongside Schuller’s lectures), the long section on quantum mechanics and the measurement problem (which is the most careful treatment of the foundations I have found anywhere), and the final chapters on quantum gravity and twistor theory.

Penrose’s view on the measurement problem — that wavefunction collapse is a real physical process driven by a difference in the quantum superposition of spacetime geometries — is heterodox and probably wrong. But the precision with which he states the problem, and the honesty about what the standard interpretation does and does not explain, is a model for how to think about foundational issues.

Steven Weinberg — The First Three Minutes (1977)

Already mentioned in the cosmology notes. Worth noting here for a different reason: it is the best example I know of writing that is simultaneously technically accurate and narratively engaging. Weinberg does not hide the equations — he includes them when they are needed — but he writes with a historian’s sense of how ideas developed and a novelist’s sense of pacing. The opening chapter, which describes the universe in the first few minutes, is structured as a drama. The final chapter, “Epilogue: the Prospect Ahead,” ends with the sentence: “The more the universe seems comprehensible, the more it also seems pointless.” That sentence caused me to think for a week.

G.H. Hardy — A Mathematician’s Apology (1940)

Hardy’s defence of pure mathematics — the argument that mathematics is worth doing for its beauty alone, independent of any application. Written when Hardy was old and his creative powers were fading, it has the clarity of someone who has spent a lifetime at the frontier and knows exactly what he found there.

The specific claim I find most interesting is Hardy’s assertion that the best mathematics is characterised by unexpectedness combined with inevitability — that a good theorem surprises you and then, once you understand the proof, seems like it could not have been otherwise. This is exactly what I find in the best physics as well. The Hawking temperature formula. Asymptotic freedom. The proof that a manifold with a trapped surface must contain a singularity. All of these are unexpected on first encounter and, on second encounter, feel inevitable.

Hardy’s remarks on the permanence of mathematics — that a mathematical truth, once proven, is proven forever — have a particular resonance. The ephemerality of many things I care about contrasts sharply with the permanence of a mathematical result. Once you prove something, nobody can take it away.

Richard Feynman — QED: The Strange Theory of Light and Matter (1985)

The most remarkable exercise in popularisation I have encountered. Feynman explains quantum electrodynamics — the complete quantum theory of light and electrons, accurate to one part in a billion — without a single equation, using only arrows rotating in the complex plane. The conceptual framework (path integrals, probability amplitudes, the sum over histories) is conveyed exactly correctly, but in a language that requires no mathematical prerequisites.

The reason this is inspiring is not the accessibility but the precision. Every sentence in QED is exactly right — not an approximation, not a metaphor that captures the feel but misses the substance, but the actual content of the theory stated in ordinary language. That this is possible is itself a fact about physics: the structure of QED is sufficiently clear that it can be stated without mathematics, even if computing with it requires the mathematics. The structure and the computation are different things.

C.W. Misner, K.S. Thorne, J.A. Wheeler — Gravitation (1973)

1,279 pages. The encyclopaedia of general relativity. I have not read all of it — nobody reads all of it at once — but I have spent considerable time in specific sections: the treatment of geodesic deviation and curvature (Chapter 11), the discussion of the initial value problem for GR (Chapter 21), the treatment of gravitational waves (Chapters 35–37), and the section on quantum gravity (Chapter 44, which is now historically interesting as a document of where the field stood in 1973).

What makes this book different from every other physics textbook I have used is the visual richness. MTW includes hundreds of figures, many of them geometric diagrams drawn by Kip Thorne that convey the structure of curved spacetime in ways that equations alone do not. The book also includes philosophical discussions throughout — boxes with titles like “What is the role of the equivalence principle?” and “How can space tell matter how to move while matter tells space how to curve?” — that are not found in more modern treatments but that reflect the way the authors actually thought about the subject.

IV — Papers of Particular Technical Beauty

P.A.M. Dirac — “The Quantum Theory of the Electron,” Proceedings of the Royal Society A117 (1928)

The paper in which Dirac derived his relativistic wave equation for the electron by the requirement that the equation be first-order in both space and time derivatives (so that it transforms simply under Lorentz transformations). The derivation forced the introduction of four-component spinors and predicted:

  • The electron’s spin as an intrinsic property, not added by hand
  • The electron’s magnetic moment, with the correct $g$-factor
  • The existence of antiparticles (the positron, discovered experimentally in 1932)

All of this from the requirement that the wave equation be Lorentz covariant and first-order. The economy of the argument — one constraint producing three non-trivial predictions — is a model of theoretical physics at its best.

E. Noether — “Invariante Variationsprobleme,” Nachrichten von der Gesellschaft der Wissenschaften zu Göttingen (1918)

Noether’s theorem: every continuous symmetry of the action corresponds to a conserved quantity. Spacetime translation symmetry gives conservation of energy and momentum. Rotational symmetry gives conservation of angular momentum. $U(1)$ gauge symmetry of electromagnetism gives conservation of electric charge.

The theorem is the deepest result connecting symmetry to dynamics, and it underlies the entire structure of modern physics. Every conservation law that is used in theoretical physics — not just the classical ones but the conserved currents of QFT, the Ward identities, the charges associated with BRST symmetry — is an instance of Noether’s theorem.

Faddeev and Popov — “Feynman Diagrams for the Yang–Mills Field,” Physics Letters B25 (1967)

The paper introducing the Faddeev–Popov procedure and ghost fields for quantising non-Abelian gauge theories. Without this paper, the Standard Model could not have been quantised. The ghost fields — fermions with the wrong statistics, introduced to cancel the unphysical gauge degrees of freedom — appear everywhere in BRST quantisation and are essential to the structure of my own work. Short (two pages) and foundational.

Bekenstein — “Black Holes and Entropy,” Physical Review D 7 (1973)

Bekenstein’s argument that black holes must have an entropy proportional to their horizon area. This was controversial when published — Hawking initially disagreed, arguing that a thermodynamic temperature for a black hole required the existence of Hawking radiation, which he had not yet discovered. Bekenstein was right. The paper is a model of physical intuition working ahead of formal derivation.

Penington; Almheiri, Mahajan, Maldacena, Zhao — Island Formula Papers (2019)

Two papers published simultaneously:

G. Penington, “Entanglement Wedge Reconstruction and the Information Paradox,” JHEP (2020) — arXiv:1905.08255 A. Almheiri, N. Mahajan, J. Maldacena, Y. Zhao, “The Page Curve of Hawking Radiation from Semiclassical Geometry,” JHEP (2020) — arXiv:1908.10996

These papers showed, using a combination of AdS/CFT and the replica trick, that the entanglement entropy of Hawking radiation follows the Page curve — it rises initially, reaches a maximum at the Page time, and then decreases. This is consistent with unitary evaporation and represents a resolution of the black hole information paradox, at least within the AdS/CFT framework. The key ingredient is the island formula: the entropy of the radiation includes contributions from an “island” region behind the horizon, connected to the exterior via a replica wormhole.

These papers are the most important recent results connecting quantum information, holography, and black hole physics. They are also directly relevant to the QGET programme — the island formula suggests a specific way in which entanglement structure encodes spacetime connectivity, which is exactly the claim QGET makes.

V — Books on the Culture of Physics

Abraham Pais — Subtle Is the Lord: The Science and the Life of Albert Einstein (1982)

The definitive scientific biography of Einstein. Pais was a theoretical physicist and a personal friend of Einstein’s; his account of Einstein’s science is technically precise in a way that no general biography can be. The treatment of the development of general relativity — the eleven years from the equivalence principle (1905) to the final field equations (1915) — is the most detailed account of a major scientific development in the historical literature.

The title is from Einstein’s remark about God: “Raffiniert ist der Herrgott, aber boshaft ist er nicht” — “Subtle is the Lord, but malicious He is not.” Einstein believed nature was subtle — hidden — but fair. The laws, once found, would be elegant and economical. Whether he was right about this remains a question.

James Gleick — Genius: The Life and Science of Richard Feynman (1992)

The best biography of Feynman, by the author of Chaos. Gleick had access to Feynman’s papers and interviews and had the technical background to describe the science accurately. The account of Feynman’s derivation of the path integral from Dirac’s idea about the Lagrangian — the moment in his wartime Princeton office where it all came together — is the best account I have read of what mathematical discovery feels like.

G.H. Hardy and E.M. Wright — An Introduction to the Theory of Numbers (1938)

Strictly a mathematics book, not a physics book. But I include it here because it represents something about the culture of mathematical work that I want to preserve: the plain, serious, unhurried prose of people who are clearly enjoying what they are doing. Hardy and Wright do not perform enthusiasm; they simply work through beautiful mathematics together. Every theorem is stated correctly; every proof is complete; nothing is left vague. This is the standard I try to hold my own writing to.

A Note on What Inspiring Means

I want to be precise about what I mean when I say a paper or book is inspiring. I do not mean motivational in the sense of making me feel good about science or convinced that great things are possible. Most genuinely inspiring scientific work makes me feel the opposite of that, at least initially — it shows me how much has already been done and how far the frontier is from where I currently stand.

What inspiring means, for me, is: it showed me something I could not have seen otherwise. It made a connection I did not know existed. It asked a question I had not thought to ask, or answered a question I thought was unanswerable, or showed that a question I thought was answered was still open.

The papers I have listed above did one or more of these things. The Yang–Mills paper showed me that symmetry could be gauged in a non-Abelian way and that this had extraordinary consequences. The Hawking radiation paper showed me that quantum mechanics and general relativity, when brought into the same calculation, produce something completely unexpected. The Van Raamsdonk paper showed me that the connectivity of spacetime might be equivalent to quantum entanglement. The Penrose singularity theorem showed me that geometry could be used to prove that something must be singular without specifying what kind of singularity.

Each of these was a form of illumination — a moment when something came into view that had been invisible before. That is what I am looking for when I read, and it is what the best papers and books consistently provide.

Last updated March 2026.