Books of 2025, part 2

Short reviews of the 43 books I have read since July

Happy New Year!

2025 was an OK year in reading for me. I read a lot of good stuff, but less great new-to-me stuff than in 2024. That year I gave three books the title of book-of-the-year (Plato’s Republic, Joyce’s Ulysses, Carson’s Autobiography of Red); and in hindsight there’s a fourth (Hutton’s Triumph of the Moon) that should also have made the list. Whereas this year, I only found two new-to-me books that I absolutely adored and are unequivocal five-stars; and one of them, Shakespeare’s sonnets, was not much of a discovery. Not that I didn’t enjoy it, I loved it, but obviously I was going to love it.

Probably the lack of discovery was because I did a lot of re-reading this year, more than was optimal. But still, every single time I read Macbeth it gets better; every time I read the letter to the Romans it gets better; and every time I read Shame and Necessity it gets better; so while I lost out on new discoveries, it’s not that much of a loss.

Maybe because of the amount of re-reading I did, I managed to read 78 books this year: 35 in the first half of the year, and 43 in the second half. You can find the review of the 35 books I read between January and June here, while the reviews of the books I read between July and December are below. Most of these were initially posted to my new Goodreads account, if you want to follow me there. You can use the side menu to the left to navigate this post, or just ctrl+F.

I mentioned above that I had two ‘books of the year’, one of which was Shakespeare’s sonnets (which all of you know you should read anyway, you don’t need me telling you). The other was the two volumes of Roy Foster’s W. B. Yeats: A Life. You may ask yourself: Could the life of a poet really deserve two volumes, some 1200 pages not counting appendices and endnotes, and physically large pages (with small font) at that? The answer to this question is unambiguously ‘yes’. This is not the overindulgence of an uncritical official biographer; it is the furthest thing from that. It’s the best book about Irish history I have read, and probably will ever read; it is also the best book about poetry I have read, and probably will ever read. If you enjoy biography, of any kind—even if you’re one of those Silicon Valley Canon types who only reads biographies of billionaires or relatively-minor political figures—I cannot recommend this enough.

• Poetry and drama

  • Catullus – Poems

  • Wendy Cope – Collected poems

  • Euripides – Bacchae

  • Seamus Heaney – New selected poems

  • William Shakespeare – A Midsummer Night’s Dream

  • William Shakespeare – Macbeth

• Novels

  • Ben Aaronovitch – Rivers of London

  • Ben Aaronovitch – Moon Over Soho

  • Anonymous – Tales of the Elders of Ireland

  • Raymond Chandler – The Long Good-bye

  • Agatha Christie – Crooked House

  • Charles Dickens – Great Expectations

  • Graham Greene – The End of the Affair

  • Terry Pratchett – Wyrd Sisters

  • Terry Pratchett – Guards! Guards!

  • Terry Pratchett – Night Watch

  • George Saunders – Lincoln in the Bardo

• Biography, criticism, and letters

  • Morris C. M. Chang – Autobiography vol. 1

  • Morris C. M. Chang – Autobiography vol. 2

  • R. F. Foster – The Arch-poet

  • Andrew Motion – Keats

  • Sharon Osbourne and Penelope Dening – Extreme

  • M. W. Rowe – J. L. Austin

• History

  • E. R. Dodds – The Greeks and the Irrational

  • R. F. Foster – The Irish Tale

  • Max Hastings – All Hell Let Loose

  • Ben MacIntyre – The Spy and the Traitor

  • Dwarkesh Patel and Gavin Leech – The Scaling Era

• Politics and ‘society’

  • Martin Gurri – The Revolt of the Public

  • Fintan O’Toole and Sam McBride – For and Against a United Ireland

  • Thomas C. Schelling – The Strategy of Conflict

• Philosophy

  • Daniel Dennett – Darwin’s Dangerous Idea

  • David Lewis – On the Plurality of Worlds

  • John Moriarty – Dreamtime

  • Plato – Five dialogues

  • Bernard Williams – Shame and Necessity

• Religion

  • Augustine – The City of God

  • John M. G. Barclay – Paul

  • Bruce W. Longenecker – The New Cambridge Companion to St Paul

  • Paul – Letters

• Pregnancy

  • Emily Oster – Expecting Better

  • Penny Simkin and Katie Rohs – The Birth Partner

• Miscellaneous

  • Robert Graves – The White Goddess

Poetry and drama

Catullus – Poems (c. 62–54 B. C. E.)

Be aware that I don’t have any Latin, so my thoughts on Latin translation are literally worthless, but nonetheless: my take on Peter Green’s Catullus is Yahtzee’s take on The Witcher 3—the act of translating always involves trade-offs and sacrifices, but by sheer bloody hard work Green has found ways to ‘have it all’ far more often than you’d think possible. His version of 2A is my favourite, brilliantly creative especially in its word choices (‘smidgin’ and ‘pecker’ are both inspired), in ways that clearly took a hell of a lot of effort. But you can’t push the Pareto frontier out to infinity, and sometimes the trade-offs Green makes at that frontier seem misguided to me, like his choice to open the famous 16 with ‘Up yours both, and sucks to the both of you’ (!). It isn’t that Green is prudish, he’s got ‘fuckfests’ and ‘queening’ elsewhere. The notes maybe suggest a misguided objection to making Catullus sound like Malcolm Tucker—notwithstanding that Green’s 15 sounds, correctly, like Method Man—but my guess is that Green just couldn’t get the literal meaning to fit his desired rhythm (perhaps because he didn’t know the English verb ‘to facefuck’?). Still, my quibbles are marginal (in the literal sense: on-the-margin).

Wendy Cope – Collected poems (originally published 1986–2024)

She’s shallow, she’s English (derogatory), she’s all-too-often merely clever, and occasionally in later collections she’s downright insipid. (‘A Poem about Jesus’ springs to mind.) But when Cope is good, she is incomparable. I have learned ‘Waste Land Limericks’ by heart, it is a poem of such precise, gorgeous intelligence. The understated construction of ‘On Sonnet 18’ puts me in awe of Cope; coming up with the ‘“Yes”?’ on its own is an act of genius, but putting it casually in the middle of the third quatrain bespeaks total confidence and control. She has a greater mind than me or you; which makes it all the more depressing that she doesn’t always enjoy using it, and that she often allows the conventions and clichés of her class and nationality to tell her she needn’t bother. ‘Engineer’s Corner’ or ‘Kindness to Animals’ could be lifted directly from the shallowest dinner party conversation you could imagine, but for the rhymes. Ultimately, being demotic is not a virtue when the demos in question—middle-class, middlebrow middle England—is so uninspiring. Cope has the talent and the brains to rise above it, she often does rise above it, so she doesn’t get points for wallowing.

Euripides – Bacchae (405 B. C. E.)

I know this is a stupid thing to say, but I don’t think Euripides knew how booze works? Lots of what the chorus says only makes sense on the supposition that alcohol is actually like morphine, and much of the action seems to imply that alcohol is actually like meth. Obviously this makes no sense and Euripides would have actually consumed wine; but like, I have been drunk before, I know what it’s like, it’s a very distinctive experience, and it most definitely is nothing like this! This isn’t just nitpicking: a lot of what the play is trying to convey relies on specific claims about the type of experience Dionysus provides with wine (chaste people stay chaste, ll. 310–320; it’s compatible with quietness, common sense, and watching what you say, ll. 385–395; it takes away pain, passim), and I think Euripides is often straightforwardly wrong about that experience. Am I missing something?

Seamus Heaney – New selected poems (originally published 1988–2013)

Even though I know Heaney’s earlier career incredibly well, I’d never properly sat down with his later stuff before, and it was a revelation: these are amazing. (Well, not all of them—‘Audenesque’ sucks.) Many of the best are written by Heaney the classics student, who I feel very at home with: he was no scholar, but could use Latin poetry as common knowledge to express himself to others—exactly as I did with classmates aged seventeen, using Heaney’s own Field Work rather than Virgil. Of course, we were making shitty jokes about ‘mauve stars’, not composing works of brilliance like the first ‘Sonnet from Hellas’, ‘Mycenae Lookout’, ‘Route 110’… Earlier this year, I suggested that Heaney’s brilliance in Station Island came from the dialectic of striving after unattainable knowledge. But it is now apparent to me that in the late 1980s Heaney managed to take the brilliance he’d developed and leave the difficulties of the dialectic behind entirely—as he put it, he simply ‘straightened up’. How he did this I don’t know: the classics clearly helped, but he could do without them. I dream of my speech sounding exactly like ‘Whitby-sur-Moyola’, with its learnedness and its Ulsterisms reinforcing each other perfectly.

William Shakespeare – A Midsummer Night’s Dream (c. 1595)

William Shakespeare – The Tragedy of Macbeth (1606)

Every time you read him, even if you think you know how good he is, even if you’ve read it before, he hits you with ‘you are like a little baby, watch this’ and it’s better than anything you’ve read all year and better than you could have imagined. The fucker.

Novels

Ben Aaronovitch – Rivers of London (2011)

Ben Aaronovitch – Moon Over Soho (2011)

These are works of fantasy and, as such, contain such nonexistent things as magic, ghosts, vampires, river spirits, and officers of the Metropolitan Police who genuinely work to fight crime and protect the public. Ok, now we’ve gotten the obvious joke out of the way—the really unrealistic thing in Aaronovitch’s writing is how quickly everyone adjusts and moves on once magic is explained to them; but it works well, keeps the pace up and avoids the boring ‘adjustment’ scenes. Sadly, the plots are too messy: in book one there may or may not be two distinct main villains working together, or maybe they’re the same, the novel is insufficiently clear; and there’s a secondary plot about rivers tacked on; similar in book two. Good fun while they last but they don’t hang together. The audiobooks are really very well done and earn an extra star on this review.

Anonymous – Tales of the Elders of Ireland (c. 1200)

It’s slow to start: I’d maybe read the first bit until Caílte is baptised by St Patrick just for context, then skip ahead to around chapter 3, where Caílte has adventures with the Tuatha Dé. The prosimetrics are very stop-start, not helped by the fact that the verse in Dooley and Roe’s translation is middling at best. But many of the stories are exceptionally cool, and the whole imaginative world of medieval Irish literature (which is brilliantly exciting) is expressed clearly and precisely here. The issue is that, well, Matthew Arnold was on to something: that imaginative world is built out of materials taken from a context that we today don’t understand, a context that we see in the literature itself only through a glass darkly. Arnold may have been wrong when he made such confident assertions about that context, that it was pagan mythology pillaged and desecrated by ignorant monks (we know now that the monks were not so ignorant, the mythology perhaps not so pagan); but modern scholars err when they dismiss his perception that there is something here we know not what.

Raymond Chandler – The Long Good-bye (1953)

Hmm. The story opens with a murder-mystery, but then most of the novel is taken up by a seemingly-unrelated plotline where our hero Marlowe is maybe helping an alcoholic get sober. Reading this, you know that the alcoholic will eventually help lead Marlowe to the killer, and so it turns out; but this doesn’t (initially) make any sense within the internal logic of the story, you just know it because you know how novels work. Thus the audience has information that the detective couldn’t possibly learn, breaking a key rule of the detective novel. But it’s great intrigue nonetheless, written by a man whose prose style has been justly influential. Maybe my theory is that Chandler knew he was smarter than his audience, but didn’t know how to deal with that. A few times he writes a very clever little prose construction, then explains it—thereby letting all the air out of it—seemingly because he was worried you’d not get it; too proud to take it out, too anxious to let it stand on its own. I think some of Chandler’s plotting and use of noir tropes at the macro level reflect the same dynamics as these micro-level constructions.

Agatha Christie – Crooked House (1949)

Thematic spoilers follow, though no explicit plot spoilers. Many early Christie stories had reactionary 1920s vibes. Not that she’d necessarily read Spengler or Fisher or whoever, but an obsessive (though deeply unsystematic) focus on heredity, shading seamlessly into race science, was ‘in the air’ and infected almost everyone; likewise, anti-semitism was ‘in the air’, and kneejerk anti-liberalism. Now, later in life, and especially postwar, the culture moved on and Christie moved on with it. But she announces in her author’s foreword that the idea for Crooked House sat in her mind for many years before she finished the book, and it shows. The solution comes down to sub-hereditarianism, and even—if I might make a small inferential leap—to phrenology. This isn’t me trying to cancel Agatha, it’s just that this is one of those ones where there are vanishingly few concrete clues and it really comes down to motive, to ‘psychology’; if the author’s assumptions about psychology come from phrenology, then the solution basically cannot be satisfying to the twenty-first century reader.

Charles Dickens – Great Expectations (1860–1861)

I decided that, if I was going to keep talking shit about Dickens, I had to at least read one novel that Jeff Daniels couldn’t dismiss as ‘minor Dickens’. But I could not finish it, which certainly meant it didn’t change my views. You know the way you have manifolds, spaces that locally look like ℝⁿ but can have a radically different global structure? Dickens novels are like manifolds: Great Expectations is locally homeomorphic to a good book, but then you zoom out… His sentences really could almost seduce me, and I could therefore enjoy many of the chapters built out of them. But I never wanted to pick it up after having put it down, because the overall plot is a complete slog, filled with completely interminable characters, jarring cliffhangers and melodrama (like the time Pip’s sister is raped for no reason whatsoever) and themes handled with the same subtlety and grace as the bombing of Hiroshima. (‘Wow, Boz, you’re saying that Pip’s unjustifiable snobbery leads him to completely misunderstand the situations he finds himself in? I must have missed that: why not bash me over the head with it twenty times in the next five pages.’)

Graham Greene – The End of the Affair (1955)

As I take it, this is a novel about how love is ultimately nothing but the love of God (note the use of second-person pronouns); how corruption by sin has left love so twisted that it is almost indistinguishable from hate; but also, how the flipside of this is that all human desire, even disbelief and hate, has something of love (and therefore, something of the love of God) in it, and can be used and reshaped by grace. That is all to say that it is an even more deeply Catholic novel than it has seemed to many, informed by the specifics of Catholic anthropology. Reading it in this way, and in light of the facts of Greene’s biography (and the way they diverge from this semi-autobiographical novel), The End of the Affair seems less characterful and invested in tension, more schematic and moralising: which is maybe missed by critics unfamiliar with the schema in question? In many ways Greene seems to have been an atheist who could not believe that atheism exists, victim to the greatest trick that Christian ressentiment ever pulled.

Terry Pratchett – Wyrd Sisters (1988)

This might have been a reread? I read a number of the Discworlds about a decade ago, but I can’t remember which; some elements of this book seemed familiar. But then, it’s a Shakespeare pastiche (and I love Shakespeare) which pokes lots of loving fun at neopaganism (and I’ve spent all year thinking about The Triumph of the Moon), so maybe that’s the whole reason it seems familiar. It is funny—at times very funny—and easily digestible, which is what I read it for. But I struggled to follow or care about the story, except insofar as it just ripped off Shakespeare; many of the techniques Pratchett deploys to create intrigue just took me out of it. The extended droit du seigneur joke doesn’t pay off the way it seemed to be heading, which was disappointing—there was a better, more thematic punchline that Pratchett seems (to me) to have been toying with, but gave up on. Final point: if you’re going to have one character called ‘Granny’ and another called ‘Nanny’, you’d better be damn sure that your copyeditors give enough of a shit to check, as Pratchett’s copyeditors clearly did not.

Terry Pratchett – Guards! Guards! (1989)

A weirdly good rebuff to the public choice / Charles Tilly / Anarchy, State, and Utopia theory of the state. Ankh-Morpork under Vetinari just is a perfectly efficient cartel of stationary bandits, and indeed it shares many similarities to real-life states. But the bureaucratic rigidness of Carrot and the dogged curiosity of (sober) Vimes also have their counterparts in real-life state institutions, yet the whole novel is about how these phenomena have no place in Tillyopolis. I’m not sure whether the dragon-summoning plot is supposed to give us Pratchett’s model of why states are not merely stationary bandits? (Probably not; I’m not sure if he ever actually read the stuff I’m taking him to be criticising.) But certainly he tells us his alternative theory, what he takes to be the actual grounds of the state: ‘the essential servility of mankind’. Still funny.

Terry Pratchett – Night Watch (2002)

Bernard Williams says somewhere that the main thing he learned from Anscombe was: doing good philosophy, being a good thinker, requires more than being ‘clever’. In Williams’ specific case it’s a rather unfair remark (Williams owed a lot more to Anscombe than that), but I think it gets at something important about writing and thinking. Reading Night Watch, it is obvious that Pratchett had an astonishingly rich cultural hinterland (probably richer than yours or mine), had thought hard about it, and had buckets of natural intelligence and imagination. And yet, there’s something missing. I don’t feel any inclinations to ‘think with’ Pratchett on any of the topics he touched, I don’t feel his ideas blossoming into real people or a vibrant atmosphere or a gripping narrative. You could blame the (surmountable) limitations of genre, you could blame the (nonexistent) limitations of comedy, but I think the right thing to blame is that Pratchett was ‘clever’ in the Anscombe-Williams sense. He had the raw brains and the reading/context, but lacked the other virtues necessary to turn his work into great writing, or (what Williams observed was sometimes the same thing) great thinking.

George Saunders – Lincoln in the Bardo (2017)

I once heard David Runciman speculate that Weber’s ‘Politik als Beruf’ was inspired more by Lincoln than Gladstone. Saunders is really good on a crucial aspect of the politician’s Beruf that is emphasised by Weber: the immense moral luck involved. In 1862 Lincoln was rather morally unlucky, and Saunders is brilliant in conveying how little the ‘sweetest, wisest soul of all my days and lands’ seemed either sweet or wise, even to himself. This deep insight into the greatest man of the nineteenth century mostly (though not entirely) makes up for other areas where Saunders is insufficiently careful about history; but certainly, don’t read this if you are mainly looking for a historical novel. Rather, it’s a religious novel loosely wearing a historical cloak, á la Judith, but instead of Judaism the book’s religion is what you might call ‘Western Buddhism’. And the greatest compliment I can pay the novel is that it’s convinced me that Western Buddhism—which I had previously dismissed as empty southern Californian nonsense—can actually be incredibly spiritually deep. But depth isn’t the same as truth, and I’m not converted: not least because Saunders’ assumptions are incompatible with the Irish wake practice.

Biography, criticism, and letters

Morris C. M. Chang – Autobiography of Morris C. M. Chang, volume 1: 1931–1964 (1998)

Morris C. M. Chang – Autobiography of Morris C. M. Chang, volume 2: 1964–2018 (2024)

Books about business and tech figures aren’t my usual fare, but hey, opportunities to read translations-in-progress don’t come around every day, ‘seize the carp’ and all that. This particular carp was quite slippery, and I’m not sure what I think of Chang. The standard model is that he is a brilliant visionary, whose idea of building a foundry that would only build other people’s designs showed immense foresight and wisdom. But Chang tells us that he focussed solely on manufacturing (at least in part) because 1980s Taiwan just didn’t have the economic capacity for any other part of the process. This suggests an alternative model: Chang was a not-particularly-talented semiconductor executive who lucked out by building a pure-play foundry right at the moment that fabless was taking off. This doesn’t seem right either (it doesn’t explain why nobody’s been able to replicate TSMC elsewhere, especially given the geopolitical incentives!) and I’m sure the truth involves synthesising these two models; but I don’t know how to synthesise them, and this biography didn’t help much. There’s a lot of score-settling and self-aggrandisement in here, and (relatedly) far too many pages given over to Chang’s time at Texas Instruments—not that interesting.

R. F. Foster – W. B. Yeats: A Life, volume 2: The Arch-poet, 1915–1939 (2003)

Together, the last few chapters of The Apprentice Mage and the first ~half of The Arch-poet are an unbelievably good portrait of a creative hitting his fucking stride, having a miracle period akin to Einstein c. 1905–1917 or James Brown c. 1967–1971. ‘You were silly like us: your gift survived it all’ is exactly wrong, and my main takeaway from Foster’s two volumes is the realisation that Yeats’ gift did not ‘survive’ in spite of his silliness but emerged from it: a big part of his greatness was the proportion (certainly not 100%, but much greater than chance?) of his silly obsessions and decisions which turned out in hindsight to somehow have given him exactly what he needed to scale the poetic heights even further. His most successful ‘silliness’ was proposing on the rebound to a woman half his age: George Yeats is quietly revealed by Foster to have been the greater mind in her sadly-lopsided marriage, and an incalculably important influence on her husband. She is also extremely sympathetic on top of being smart and witty; on numerous occasions when W. B. asked George to channel the spirits, the message from the other world to her ageing husband amounted to ‘make sure to fuck your wife tonight’.

Andrew Motion – Keats (1997)

If your job is bashing ignorant literary critics over the head with ‘obviously Keats was a liberal, he repeatedly turned to camera and said “I am a liberal now” then specified what kind of liberal he was, you idiots’, then I could maybe be willing to forgive derivative and somewhat-shallow social and cultural history: you’re taking the ‘baby’s first’ approach. But Motion seemingly failed to get Keats critics to internalise ‘Keats was a liberal’, maybe because of the crudeness and shallowness of much of his analysis. (Helen Vendler was one who failed to grasp Keats’ liberalism, smugly declaring in her review of Motion’s biography that Keats was after ‘transhistorical truth’; but for God’s sake she was right about the fucking bees!) And anyway, I came to this not mainly for the poetry, but because I’m interested in late–eighteenth / early–nineteenth century liberalism, and I thought Motion’s focus might translate into relevant lessons. Keats was articulate and intelligent, and left a lot of stuff in writing, so he could be a rich case study of how people who were not politicians or political theorists arrived at and conceptualised liberalism. For this purpose, shallow cultural and social history is fatal.

Sharon Osbourne and Penelope Dening – Extreme: My Autobiography (2005)

Sharon Osbourne’s life is one of the most interesting of Britain’s twentieth century, a story anyone with any serious interest in biography should be familiar with. What we need is a good biography, but in its absence this extremely trashy mass-market ‘celebrity autobiography’ will have to do. Alas, Osbourne’s ghostwriter has no interest in making sense of the aspects of her life that are hard to understand: at most, these aspects are just displayed to satisfy celebrity-culture voyeurism, often with noticeable gaps where an inquisitive biographer should have been asking ‘what?’ or ‘why?’ (the crucial decade of the 90s receives particularly cursory treatment). Key themes left underanalysed include the weird aspects of Osbourne’s psychology (she shits on objects a lot?), her work ethic and ‘production function’, and her relationship with her husband. Even the plates raise unanswered questions: a picture of her husband captioned with ‘in real life Ozzy is the least scary person in the world’ faces a picture of Sharon with a black eye. I fully believe that this juxtaposition reflects her true feelings; but the only progress I’ve made towards understanding those feelings is of the Socrato-Rumsfeldian kind.

M. W. Rowe – J. L. Austin: Philosopher and D-Day Intelligence Officer (2023)

The story behind this book is brilliant. Rowe was writing an intellectual biography of a philosopher, and came across a couple of hearsay anecdotes about how he did important intelligence work before D-Day. He wanted to check this, started digging, and before long found himself committed to the years-long task of piecing together the story of a man who was absolutely integral to the Allied war effort and never told a soul. Austin seemed comfortable with keeping the secret, in part thanks to a personality that most readers will not warm to but which made me go ‘he’s just like me fr fr’. Further thoughts here.

History

E. R. Dodds – The Greeks and the Irrational (1951)

Alternative title: A Commentary on Auden’s ‘In Memory of Sigmund Freud’. In the pages of this book you can watch, as if in real time, as a kind of late-Victorianism radically transforms itself into so-called ‘Cold War liberalism’ (one of the greatest misnomers in all intellectual history). You can see, too, just how important Freud was to this process, thanks to Dodds analytically separating the different logical steps that were indeterminately melded together in Auden’s ‘whole climate of opinion’. But also, for modern readers with historical distance from psychoanalysis, Dodds’ argument accidentally helps you see ways in which Freud himself did not totally stand outside the process he influenced. As such, the present reviewer came away from the work with the message that, if you truly wish to advance Dodds’—and Auden’s—project of being ‘enthusiastic over the night … because it needs our love’, you need to revert from Freud to his predecessor Nietzsche. Which was exactly what Dodds’ student Bernard Williams (below) did. (Also Dodds has some things to say about ancient Greece, maybe, idk.)

R. F. Foster – The Irish Story: Telling Tales and Making It Up in Ireland (2001)

The famous essay about Angela’s Ashes and Gerry Adams deserves its reputation and justifies the book’s entry price by itself. The rest of the volume might be entertaining to someone who is less familiar with (and jaded about) the ‘revisionist’ debates than I am; but at the highest-stakes moments in his arguments, Foster creeps right up to the line between opinionated (good for popular and introductory writing) to tendentious (very bad for those same types of writing). The four essays focussed on Yeats entirely escape this, thanks to Foster’s incredibly deep familiarity with the material; but that same material is treated in greater detail in his biography of the poet, which I had just finished reading, so (again) I just wasn’t the right audience. Still, I repeat that the Angela’s Ashes piece really is excellent.

Max Hastings – All Hell Let Loose: The World at War 1939–1945 (2011)

The title and marketing understate the degree to which Hasting’s narrative is structured around contemporary diary entries, letters, reportage, and (in general) eyewitness testimonies. This ‘eyewitness’ structure is the most distinctive aspect of the book, and it was a really good choice. The second World War was the biggest event in human history (a superlative I feel no hesitation in using), that involved almost every single person alive on the globe; even a long book like this simply cannot provide you with a deep or detailed understanding of the whole causal process. (Many books exist which try and are shallow for it.) Hastings has his own opinions on the causal process, but he doesn’t try to make them bear the weight of the book; they are peppered in with restraint. Instead of a work that tries to systematically map the arrows in the directed acyclic graph of the war, Hasting tells you what each node of the graph looks like from ground level (to mix metaphors). Visceral, and (indeed) makes you even more interested in where the arrows go and which way they point; a helpful aid for resuscitating my interest in the war.

Ben MacIntyre – The Spy and the Traitor: The Greatest Espionage Story of the Cold War (2018)

The story of a British spy inside the KGB in the 1970s and 80s. It is quite obvious that the point of this book, indeed the reason MI6 ran Oleg Gordievsky in the first place, is so he can be ‘Phim Kilby’, an Uno ‘Reverse’ card to spiritually undo this country’s deepest twentieth-century humiliation. ‘The Gordievsky case compensated, in some measure, for the embarrassments of the Philby years,’ writes MacIntyre; elsewhere, ‘[n]ow MI6 was doing to the KGB what the KGB had once done to it.’ Alas, this is nonsense, for two reasons. First, Gordievsky’s intel just let MI6 know that the KGB had already descended from ‘hypercompetent secret army’ to ‘paranoid security arm of a collapsing empire’; I’m sure this was nice to hear, but (unlike Philby’s intel) I struggle to believe it changed anything. The Soviet system died all on its own. And, second, in telling Gordievsky’s story MacIntyre is also forced to talk about Aldrich Ames, an equal and opposite humiliation inflicted on the West at the same time; Ames, indeed, told the Soviets about Gordievsky. Even as communism slowly collapsed around it, the KGB was still a net-humiliator of the West; Philby can’t be undone.

Dwarkesh Patel and Gavin Leech – The Scaling Era: An Oral History of AI, 2019–2025 (2025)

Misleading subtitle, the book cuts off in 2024. My main takeaway is how little any of these people know what the fuck they’re doing. I don’t blame Patel or Leech for this; indeed, I’m really glad that Patel knows enough about the field, and is clever enough, to ask the questions that reveal the real limits. He opens the very first chapter with the most obvious thing that any intelligent person would want to know about the field: given that we are in the ‘scaling era’ of artificial intelligence, why does ‘scaling’ lead to better artificial intelligence? The best answer comes from one interviewee who posits some kind of totally unknowable ‘data manifold’ made up of who-knows-what components that might answer the question if it even exists; everyone else is reduced to mystic burbling. If ‘knowing what the fuck you’re doing’ were a prerequisite to ‘changing the world’, we could just write the whole field off; alas, it is not, and we cannot.

Politics and ‘society’

Martin Gurri – The Revolt of the Public: and the Crisis of Authority in the New Millennium (2014)

Here’s a man who was onto something deep, but stumbled—in large part because he’d not internalised Arrow’s impossibility theorem. Yes, the internet and social media have hugely changed the landscape. Yes, elites are reactively ‘fighting the last war’. Gurri was right about all that, and early. But he himself doesn’t understand the new conditions either, because he’s not heard the Arrovian news. He’s halfway there, and understands that a mass diffuse public can’t be for anything, can’t have a governing agenda. But he thinks that this means the public can only be against things, when the truth is that in itself the public doesn’t have any coherent preferences. As a result, his model is full of fallacies of composition, and makes some obviously incorrect predictions: how do you fit Trump’s many, many embarrassing failures into a model whereby the Trump movement is driven by outrage at failure? It’s remarkable how pessimistic Gurri is: he seems to believe, crudely, that there is no point trying to rebuild state capacity because it was a mirage in the first place. The book is too long and of uneven quality: it definitely shows that it was self-published and didn’t get a proper editor’s going-over.

Fintan O’Toole and Sam McBride – For and Against a United Ireland (2025)

Short and punchy. McBride’s chapters are better than O’Toole’s. The elephant in the room, the thing I kept thinking about, was Fergus’ question: ‘Is Northern Ireland a failed state?’ McBride especially knows more than most the dysfunction of our institutions, but there was little focus on institutional reform either in the Union or in a new Republic: lots on schools, healthcare, pensions, taxation and fiscal policy; but political and civil service reform, representation, ministerial power, etc. are mostly just referred to quickly, and usually under the rubric of a unitary vs devolved United Ireland. This is understandable based on the topics that will probably be on most swing voters’ minds in a future border poll. But without solid political institutions, even the best planning in the world won’t implement the kind of pension / health / tax / school systems that voters want. This might be thought to push towards change, and thus a United Ireland, but I’m increasingly unsure of that logic. Anyway: these kinds of questions clearly informed the writing of the book, especially McBride’s chapters, but don’t surface often enough.

Thomas C. Schelling – The Strategy of Conflict (1960)

I think Schelling is weirdly akin to his contemporary J. L. Austin (see Rowe, above). What both men thought they were doing was building up a fundamental base-layer of systematic theory which could become the foundation of a productive collective enterprise going forward. But in both cases, it has turned out that the task of actually applying the theory demanded fine-tuned individual judgment calls—almost every case is an ‘edge case’ in some way—which can’t serve as a suitable basis for the scientific division of labour. In Austin’s case, this was realised almost immediately after his death; in Schelling’s case, this still hasn’t been realised, and still today people try to get on with applying his work, usually by just sanding off the difficult bits to make it tractable. But like with Austin, Schelling’s own individual judgment was so amazingly fine-tuned as to make the theory in his hands almost magically insightful. (Does this analogy make Tyler Cowen into Stanley Cavell?)

Philosophy

Daniel Dennett – Darwin’s Dangerous Idea: Evolution and the Meanings of Life (1995)

Dennett wrote philosophy piecewise, using a series of small-scale models that each serve a local and limited, if often very interesting, end. The point (according to Dennett himself) is to use small models to prime the audience for some larger-scale way of thinking: in the case of this book, an ambitious systematic and ‘synthetic’ theory of Darwinian naturalism that applies across all of philosophy and science. You have to slowly build up a picture of this grand theory using the smaller models. But how Dennett thinks his smaller models relate to each is obscure on a first (or second, or third) read, and to grasp the details you need to extend to him a level of philosophical charity and close attention that goes way beyond the ordinary. I really do have an exceptional amount of faith in Dennettianism, so I have tried my best, and I still feel slightly lost at key points. This difficulty is disguised by Dennett’s folksy style, which makes him seem like a less technically-demanding writer than he really is, and by the fact that the models have a consistent ‘vibe’; but if Darwinian vibes are all you’re reading for, read The Blind Watchmaker.

David Lewis – On the Plurality of Worlds (1986)

Lewis was the twentieth century’s Hegel, a comparison that goes much deeper than just ‘highly systematic philosophers’ or ‘nuts’. Indeed, I think the ‘nuts’ point is overblown, or at least misunderstood. Modal realism falls out naturally from the ‘mosaic’ picture of Humean supervenience if you don’t assume that all particulars are spatiotemporally related; chapter one of Plurality is thus essentially an exposition of the mosaic worldview. Modal realism must inherit its craziness directly from Humean supervenience. And indeed, the more you pick at Humean supervenience, the weirder Lewis’ whole way of looking at the world comes to seem. Why does Lewis allow for worlds with ‘unHumean whatnots’, not just alien-but-Humean natural properties? When the inhabitants of such worlds say e.g. ‘cause’, are they referring to the unHumean whatnots or to healthily Humean counterfactual dependence? Why bother in the first place with a prolonged quasi-military campaign for a view that you admit amounts to ‘reactionary physics’? I have long harboured the suspicion that the Hegel comparison might illuminate all this; Wolfgang Schwarz has offered me some useful hints here (without mentioning Hegel). But rereading Plurality has not helped me turn this into a proper theory of David Lewis—not yet, anyway.

John Moriarty – Dreamtime (1994)

This is a bad book. Do not read it. Highlights include: literally the very first page, where the author writes ‘the answer implied but not overtly stated in this book is that…’ and then overtly states the answer.

Plato – Five dialogues (originally published 4th century B. C. E.)

Coming back to these after reading the Republic, I find many Republic-y themes jumping out at me. The argument for divine omnibenevolence is hinted at in Euthyphro; the foundational concern with division of labour is 90% there already in Apology’s brief discussion of craftsmen; that gets turned into an antidemocratic argument throughout Crito; etc. I’m also pretty sure that I disagree with the claim that Meno and Republic have incompatible epistemologies because the objects of knowledge are different. (The one continuity I don’t see is the standard story that Socratic definition-hunting turned into the theory of Forms: it’s not clear to me that the thing Socrates wants is definition.) But for all that this tells against a simple Vlastosian theory, I don’t think I’m a unitarian, because the model of reincarnation and the model of the Forms seem to change between (respectively) Phaedo and Phaedrus, and Phaedo and Republic. (I don’t actually think Plato makes the one-over-many argument in Phaedo, as tempting as it is to read him that way!) As far as I can tell my takes here are novel so I want to think about them more before committing them to writing. I should probably also read Theaetetus.

Bernard Williams – Shame and Necessity (1993)

Oh, come on. If you know me at all you know what I think of Shame and Necessity; is there any point reviewing it? Fine, I will give you a two-word review: tolle, lege.

Religion

Augustine – Concerning the City of God Against the Pagans (413–426)

I heard someone somewhere talking about teaching Augustine’s Confessions to smart teenage boys, who tend to see in him a kindred spirit. This is the right way to think about Augustine: he was history’s smartest teenage boy for his whole life. There’s an uncaring, at times seemingly almost psychopathic streak in the City of God, coexisting with an obvious capacity for kindness, which I struggled to understand—until I realised that Augustine was being brutal and cruel in exactly the way teenage boys are brutal and cruel, and that he alternated it with kindness in precisely the same unpredictable way as teenage boys do. Likewise, Augustine was smart the way teenage boys are smart: clever but not deep. Too clever for his own good, perhaps; certainly too clever for Western civilisation’s good, since we adopted his ideas wholeheartedly for more than a millennium, and still haven’t properly let go. As such, I am sorry to say that this one really is a ‘must-read’: sorry because it is just a meandering series of digressions that doesn’t have any real structure (the ‘city of God vs city of man’ thing is a bit fake), but it is too important.

John M. G. Barclay – Paul: A Very Brief History (2017)

Well, uh, at least the title isn’t a lie: it is very brief, just under 90 small pages of pretty large text, which I hadn’t quite realised when I bought it. I’m not quite sure who the target audience for this is? To be sure, they are ninety good pages—this book helped prompt me to take Augustine’s interpretation of Paul more seriously than I had before, by linking the former’s ideas about ‘pride’ to the latter’s obsession with ‘boasting’—but when the topic is the life, works, and complete reception history of one of the greatest intellectual figures of antiquity, it’s inevitable that the book seems stretched far too thin. You’d get much more out of the In Our Time on Paul, which features Barclay as a guest, or especially the New Cambridge Companion (below).

Bruce W. Longenecker (ed.) – The New Cambridge Companion to St Paul (2020)

Really good, will become my new go-to ‘if you had to pick only one’ recommendation for the secondary literature on Paul. Successfully written so as to assume very little knowledge, down to the chapter titles which are very ‘baby’s first’, and yet it takes you right up to date on the big questions and the figures in the dialectic (there’s an ‘index of modern scholars’ at the back). Two standouts: Paula Fredriksen’s chapter has a really interesting and quite compelling précis of her argument for the so-called ‘radical new perspective’, focussing on the identity of Gentile godfearers (though I’m still not convinced); and Bruce Longenecker’s chapter on death and hamartía is spot-on in its daimonic and apocalyptic elements, and was helpful in getting me to think about Paul’s apocalypticism in ways that allow one to bring in moderns like Schelling (above) or Williams (above), even if Longenecker’s talk of ‘cultural power’ and ‘relational dysfunction’ sometimes seems a bit ‘1990s English department’.

Paul – Letters (c. 50–65)

Although I regularly find myself dipping into him, I have not read all of Paul straight through in years, and it was an unblemished joy to give myself the excuse this year. Possibly the most remarkable human being to ever put pen to paper, Paul ‘defies comparison’ in the specific sense of allowing for so many fruitful comparisons that their diversity makes you realise that you’ll never grasp his depths—both Timothy Leary and Achilles come up in my notes as useful analogues for understanding the apostle. He is so tactile, so immediate, so human: the real historical individual who wrote the text stands before you as you read, as with no other author I know of. And yet at the same time, so much of the basic message of Paul’s letters is unknowable and bound to be endlessly contested because of how much of their context is lost, making them an immensely fun interpretative game for a certain type of reader (like me). Unquestionably, he is one of the all-time greats.

Pregnancy

Emily Oster – Expecting Better: Why the Conventional Pregnancy Wisdom is Wrong and What You Really Need to Know (2013)

It’s an exaggeration, but an enlightening one, to say that Oster treats pregnant bodies as black boxes, sets of input-output relations. She looks in the first instance to statistical data, not biological mechanisms (‘I am an expert in a particular thing—in data’). This style of reasoning obviously has its problems (Oster began her career doing Freakonomics-style contextless econ-imperialism, though with more-than-typical epistemic virtue); a common take is that looking to input-output relationships is second-best at best, something we should rely on only when we can’t understand the system itself. But Oster shows that you always need this type of reasoning: else, like so many doctors, you’ll risk (a) a complete disregard for the context in which the systems (in this case, pregnant bodies) are embedded and (b) plain ol’ scope neglect—and be complacent about it. The complacency means that when Oster disagrees with official advice, the response is often incoherent rage that still contains the same shockingly stupid errors. I can only recommend Expecting Better qua pregnancy book (rather than qua epistemological fable) if you’d be happy regularly deferring to Oster’s judgment; the gaps are many and the style is breezy. But better deferring to her than anyone else.

Penny Simkin and Katie Rohs – The Birth Partner: A Complete Guide to Childbirth for Dads, Partners, Doulas, and All Other Labor Companions (originally published 1989)

This was the best of any of the books we bought when Rachel was pregnant, to the point that she regularly grabbed it off me even though she wasn’t the intended audience. Respectful, smart, full of detail, pitched at the right level—Simkin assumes that you’re an intelligent adult who knows what biology is, which shouldn’t be rare but is—and importantly, really clear on when you should stand your ground and when you should fuck off. (I love the way that the book essentially says ‘if there’s a cord prolapse, shut the fuck up and stand the fuck back’—I’m paraphrasing, but I think faithfully.) Simkin obviously hates doctors exactly the correct amount, which is rare and impressive. But, I have to confess that I really was not a great birth partner despite my best efforts: you may take this to be a mark against Simkins, but I take it to be a mixture of my own failings and of the limitations of booklearnin’.

Miscellaneous

Robert Graves – The White Goddess (1948)

Graves was a crackpot and all the hallmarks of the crackpot method are here. ‘Evidence’ that isn’t properly cited so you can’t check what it says, ‘evidence’ that Graves has tweaked or altered, ‘evidence’ that Graves cites properly but which makes the complete opposite point, many claims simply asserted with nothing even resembling evidence, and (of course) an obviously psychosexual current running through the whole thing. Now, it is true that among the cracked pottery are fragments of potentially very interesting speculative fiction. But I came to The White Goddess hoping that I’d get those materials ‘pure’, but in fact they are even more ‘impure’ here: for every one paragraph of Goddess stuff you get ten pages of Graves analysing the esoteric meanings of medieval Irish poetry. If his topics (historical linguistics, medieval literature, religious anthropology) bore you, you’ll be bored; but even if the topics interest you, you’ll also be bored, because you’ll immediately notice Graves is just bullshitting. The only two things about this book worth remarking on are (a) someone made a religion out of this and (b) Graves generally conducts his crackpottery with panache, it’s not a resentful or paranoid screed.