This page is a record of specific works — music, books, films, recordings — that I find genuinely worth recommending. Not a comprehensive survey. Not a list assembled for completeness. These are things I have returned to, thought about, and can speak to honestly.

The organisation is by medium. Within each section, I have tried to say not just that something is good but why — what specifically it does, what it asks of you, and what you get back if you give it attention. Recommendations without reasons are not very useful.

A note on what I mean by “recommend”: I mean that I think the work repays serious engagement, not merely that I enjoyed it. Some of these things were not immediately enjoyable. Some required patience before they became valuable. The recommendation is for the value, not the comfort.

Books

I have written about specific books on other pages of this site — particularly on the history of mathematics and physics page and the inspiration from art and science page. What follows here is a more direct set of recommendations, with less analysis and more direct advocacy.

Mathematics and Physics

Paul Dirac — The Principles of Quantum Mechanics (Oxford, 1930, revised editions through 1958). The original presentation of quantum mechanics in the Dirac formalism. Not the easiest introduction — Dirac writes with a compression that requires real effort — but the most beautiful. The bra-ket notation, the treatment of observables as Hermitian operators, the handling of the uncertainty principle: all of it is in this book, and it is presented with a clarity that later textbooks have not always preserved in trying to be more accessible. Read this after you have learned quantum mechanics from a more conventional source. It will change how you think about what you learned.

Hermann Weyl — Symmetry (Princeton, 1952). A short book based on lectures Weyl gave at Princeton in 1951, connecting the mathematical concept of symmetry to its appearances in art, nature, and physics. The level is accessible — Weyl was writing for a general audience — but the ideas are deep. The final chapter on symmetry in physics is as good a brief account of why symmetry matters in theoretical physics as I have found. If you want to understand why physicists care so much about group theory and invariance, this is the clearest path in.

G.H. Hardy — A Mathematician’s Apology (Cambridge, 1940). I have written about this on the inspiration page. The recommendation stands: essential reading for anyone who wants to understand what mathematicians think they are doing. Hardy is wrong about some things — his claim that pure mathematics has no military applications proved incorrect almost immediately — but right about what matters, which is the aesthetic criterion. A short book. Read it in an afternoon.

Roger Penrose — The Road to Reality (Jonathan Cape, 2004). At over a thousand pages, it is a serious commitment. What it offers: a genuinely rigorous account of the mathematics underlying modern physics, from complex numbers through gauge theory and general relativity to speculative quantum gravity, written by someone who both created some of the mathematics and cares about its visual and conceptual character. There are diagrams throughout that are unlike the diagrams in other physics books — Penrose draws ideas, not just equations, and the drawings carry content. I do not know another single book that covers as much with as much genuine rigour.

Feynman, Leighton, Sands — The Feynman Lectures on Physics (Addison-Wesley, 1963–65), 3 vols. The best physics lectures ever delivered, transcribed and edited. Volume II on electromagnetism is the essential volume. Volume III on quantum mechanics is where Feynman’s path integral formulation is first accessible. Volume I is the broadest and in some ways the most remarkable, because Feynman manages to present classical mechanics and thermodynamics in a way that makes them feel like new discoveries rather than settled knowledge. Available free online at feynmanlectures.caltech.edu.

Literature

Fyodor Dostoevsky — The Brothers Karamazov (1880), translated by Richard Pevear and Larissa Volokhonsky. The best novel I have read. The Pevear-Volokhonsky translation is the one to read in English — it preserves the roughness and urgency of Dostoevsky’s prose where earlier translations smoothed it into Victorian propriety. The novel asks a single question from multiple angles across nine hundred pages: if God does not exist, is everything permitted? The Grand Inquisitor chapter (Book V) stands alone as a philosophical text. The answer the novel gives — if it gives one, which is debated — is earned over the full length of the book.

Leo Tolstoy — Anna Karenina (1877), translated by Pevear and Volokhonsky. Tolstoy and Dostoevsky are two completely different views of what fiction is for. Dostoevsky is interested in extreme states, in spiritual crisis, in the mind under pressure. Tolstoy is interested in ordinary life looked at with extraordinary precision. Anna Karenina contains the most accurate depiction of jealousy I have read, and the most accurate depiction of the way political opinions form and solidify in drawing room conversation. The Levin subplot — the landowner trying to figure out how to live — is the part I find most valuable on re-reading.

Albert Camus — The Myth of Sisyphus (1942), translated by Justin O’Brien. An essay, not a novel — Camus’s argument that the absurd (the confrontation between our demand for meaning and the world’s silence) does not lead to despair but to something more like defiance. The image of Sisyphus rolling his rock up the hill and watching it roll back down, over and over, and yet — as Camus insists in the final line — we must imagine Sisyphus happy. I find the argument imperfectly made but the conclusion right. Read alongside The Stranger (the novel that enacts the philosophy) and The Plague (the novel that tests it against communal suffering).

Jorge Luis Borges — Labyrinths (New Directions, 1962). The essential Borges collection in English. The stories that matter most to me: The Library of Babel (an infinite library containing every possible book), Tlön, Uqbar, Orbis Tertius (an imaginary world whose philosophical idealism begins to overwrite reality), and The Garden of Forking Paths (a murder mystery that is also a model of time). Borges treats mathematical ideas — infinity, combinatorics, topology, the structure of time — as literary material with an ease that no other fiction writer has matched. The stories are short, dense, and reward re-reading more than any other fiction I know.

Marcus Aurelius — Meditations, translated by Gregory Hays (Modern Library, 2002). Not a book Marcus intended to publish — it is his private journal, written in Greek during military campaigns in the 170s AD, as a record of his attempts to live according to Stoic principles. The Hays translation is the most readable modern version. What I find in it: an unusually honest account of the difficulty of living by principles you believe in, from someone who had the authority to do whatever he wanted and chose instead to hold himself to account. The philosophy is Stoic — act virtuously, accept what you cannot control, treat every moment as potentially the last — and it is presented not as a system but as a practice, repeated daily. I re-read sections of it regularly.

History and Civilisation

Edward Gibbon — The Decline and Fall of the Roman Empire (1776–89), abridged by David Womersley. The full text runs to six volumes and is not something most people will read cover to cover. The Womersley abridgement in one volume preserves the essential narrative and, more importantly, Gibbon’s prose — one of the great prose styles in English, ironic, cadenced, comprehensive. The thesis — that Christianity and barbarism together brought down Roman civilisation — is contested and in some respects wrong, but the way Gibbon handles evidence, motive, and contingency is a model of historical writing. Reading him teaches you something about how to construct an argument over a long span.

Barbara Tuchman — The Guns of August (Macmillan, 1962). The best single-volume account of how the First World War began. What Tuchman demonstrates — and this is the reason the book remains important beyond its subject — is how decision-makers can be trapped by plans, assumptions, and institutional momentum into courses of action that none of them individually would have chosen. The military plans that existed in 1914 (the Schlieffen Plan, the French Plan XVII) were so elaborate and so dependent on timing that once mobilisation began, reversal was nearly impossible. The book is a study in how systems that are individually rational produce collectively catastrophic outcomes. This is a pattern that recurs.

William Dalrymple — The Anarchy: The Relentless Rise of the East India Company (Bloomsbury, 2019). The East India Company as a case study in corporate power: a joint-stock company that acquired, in the space of a century, an army larger than most European nations and effective sovereignty over most of the Indian subcontinent. Dalrymple is a narrative historian of the highest quality, and the story he tells — using Mughal and Maratha sources alongside the standard British colonial record — is the most accurate and the most disturbing account of that particular period I have read. The financial structure of the Company, and the way it separated the costs of conquest (borne by the conquered) from the profits (borne by the shareholders), is directly relevant to any thinking about corporate governance.

Film

I watch less film than I read, and my recommendations here are more selective. I am listing films that I think reward the kind of attention I am recommending throughout this page — not films that are merely enjoyable (though several of these are also enjoyable), but films that do something that most films do not attempt.

Andrei Tarkovsky — Stalker (1979). A Soviet science fiction film based loosely on a novella by the Strugatsky brothers. Three men — a writer, a scientist, and a guide — travel to a mysterious zone where the laws of physics do not apply normally, seeking a room that is said to grant the deepest wish of whoever enters it. What actually happens in the film is almost entirely indirect: long, slow, nearly undramatic scenes in an environment of extraordinary visual texture (pools of water, rusting machinery, long grass in the rain), punctuated by philosophical argument between the three characters about whether they want to reach the room at all.

The film is slow in a way that rewards patience rather than punishing it. The question it asks — what would you actually wish for, if you had to reveal your deepest wish to something that could see through all your self-deception? — is not answered. The film earns the lack of an answer.

Solaris (1972) is the other essential Tarkovsky. Both are long, both are slow, both are about consciousness and the limits of scientific rationality in confronting phenomena that do not reduce to mechanism.

Christopher Nolan — Interstellar (2014). I am aware this is a less prestigious recommendation than Tarkovsky. I am making it because Interstellar is the film I have seen that takes the physics seriously in the most direct way. The representation of the black hole (based on Kip Thorne’s calculations), the dilation effects, the five-dimensional library sequence — these are not decorative; they are the film’s actual subject. The film has real flaws: the ending is narratively convenient in ways the physics cannot fully support, and some of the emotional material in the first half is thin. But the sequence around the black hole is the most accurate cinematic representation of what the mathematics actually describes that I have seen. For that, the film deserves its recommendation despite its imperfections.

Stanley Kubrick — 2001: A Space Odyssey (1968). The film that understood, before almost anyone else in cinema, that the encounter with genuine alien intelligence would not be narrative — it would not have dialogue, motives we could recognise, or a resolution that restores human categories. The final sequence of 2001 is forty minutes of pure visual and sonic experience, with no dialogue and no explanation. The film refuses to explain itself, which is either a failure of nerve or the only honest response to the question it is asking. I think the latter.

Ingmar Bergman — The Seventh Seal (1957). The medieval knight playing chess with Death, seeking an answer to whether God exists before he dies. The film is often described as gloomy, which is inaccurate — it is full of moments of genuine joy (the family of travelling players, the strawberries and milk) and the Death figure is treated with a dark irony that reads as comedy as much as tragedy. The theological question it raises is not answered in the film, but the terms of the question — direct, stripped of modern hedging — are clearer than in most contemporary treatments of the same problem.

A Note on How to Use This Page

These recommendations are honest rather than comprehensive. I have not tried to cover every major work in each category. I have tried to describe, with some precision, what specific things I find in specific works and why I think those things are worth finding.

The underlying principle, if there is one: works that reward sustained attention over casual engagement; works that are not depleted by repeat encounter; works that make demands of you and justify those demands by what they offer in return. This standard excludes a great deal. It is the right standard.

I update this page when I have something to add that I can speak to with the same level of specificity. I do not add things to be comprehensive.

Last updated March 2026.