In September of 1980, NBC ran a miniseries about a sixteenth-century English pilot who gets shipwrecked on the coast of Japan and spends the next several years trying to stay alive in a feudal society he doesn’t understand. My father watched the first two episodes from his La-Z-Boy and declared it confusing. I picked up the book the following week from the South Bend Public Library, a battered paperback of James Clavell’s Shogun that had clearly been through other hands, and I read all twelve hundred pages in under two weeks. I was fourteen years old.
I didn’t pick it up because I was interested in Japanese history. I picked it up because the television version was confusing and I wanted to understand what was happening. What I found instead was something I had no framework for yet: a novel that taught me more about the political architecture of feudal Japan, about the competing obligations of the samurai class, about the specific mechanics of how power gets organized in a society I’d never thought about, than anything in any classroom I’d sat in. Fiction was doing something that textbooks weren’t. I didn’t have a name for what that thing was. I’ve spent close to fifty years trying to work it out.
This isn’t a ranked list. If you want ranked lists, the internet has them in abundance: by century, by gender of author, by what wins awards, by what a particular algorithm thinks you’ll like next. What I want to do is something more particular: walk you through what historical fiction actually does when it’s working, the different things it can do, and the books that do each of them with some excellence. I have opinions. Some of them have changed over the years. A few of them I’ll defend to the wall.
The thing Clavell gets right in Shogun, the thing that makes it genuinely good despite its sprawl, is that he’s not explaining feudal Japan to you. He’s putting you inside a man who is trying to figure out feudal Japan in real time, under conditions of extreme personal danger, with no one who speaks his language and no shared framework for making sense of the rules. John Blackthorne, the English navigator, is our way in. His confusion is calibrated precisely to ours. We learn the system as he learns it, through consequence, through watching what happens when he gets it wrong, through gradually recognizing the patterns. You’re not reading about the culture. You’re experiencing the process of encountering it cold.
This is the particular gift of historical fiction at its best. Textbooks give you the structure. Fiction gives you the experience of being inside the structure without knowing it’s a structure. These aren’t the same thing, and the difference matters in ways that linger long after the book is closed.
Umberto Eco’s The Name of the Rose does this with fourteenth-century monastic life. Published in Italy in 1980 and in English translation in 1983, it’s a detective story set in a northern Italian abbey in November 1327, and it’s also a detailed education in how medieval monasteries actually worked: the hierarchy, the theological debates, the libraries and their politics, the relationship between the Church and the various heretical movements it was managing. But Eco doesn’t stop the narrative to explain any of this. The explanation is the narrative. The detective, an English Franciscan friar named William of Baskerville, investigates a series of suspicious deaths, and the abbey’s systems unfold because his investigation requires understanding them. By the time you’ve worked out what happened to the missing monks, you understand how knowledge was stored and controlled in medieval Europe better than a textbook chapter would leave you understanding it. You’ve inhabited the problem.
These are books that recreate a world. The measure is whether you feel the logic of the system from the inside, whether the choices characters make feel inevitable given where and when they are. When it works, you find yourself briefly disoriented when you set the book down, because your present-day surroundings require some adjustment. This isn’t a small accomplishment. It’s what the genre can do that no other form of nonfiction can replicate.
A different category entirely is what I think of as the reimagined person: the novel that takes a historical figure and gets so deep inside them that the line between reconstruction and testimony starts to blur.
Hilary Mantel’s Wolf Hall, published in 2009, is the clearest example I’ve encountered. She writes Thomas Cromwell, the man who served Henry VIII as chief minister through the English Reformation, from such close third-person proximity that the effect is almost vertiginous. The pronoun throughout is “he,” and Mantel trusts you to track which “he” is Cromwell and which “he” is everyone else, because Cromwell is always the center of gravity. You’re never not in his perspective. You feel his attention, his calculations, the grief underneath the calculations, his understanding of power that goes deeper than anyone else in the room has thought to develop.
What Mantel does that so few historical novelists manage is give Cromwell interiority that feels discovered rather than invented. She’s working from the historical record, the documents, the letters, the surviving accounts. She imagines the mind that produced those documents, and she imagines it with such specificity that the question of whether she got it right feels strangely beside the point. She got something right. Whether it’s Cromwell or a version of Cromwell that is actually more interesting than the provable man, I’m genuinely not sure the distinction matters as much as I once thought it did.
Toni Morrison’s Beloved, published in 1987, is the most emotionally demanding historical novel I’ve read. I’ve read it twice and I’m not certain I could read it a third time, which isn’t a criticism. Some books are supposed to cost something. Sethe, the protagonist, escaped from slavery in Kentucky but carries the weight of that slavery in ways that legal emancipation can’t touch. The novel moves between present and past in a structure that reflects how trauma actually works: not linearly, not at a safe narrative distance, but returning and returning to the thing that can’t be put down. Morrison is writing about what slavery did to people’s sense of their own humanity, and she’s writing it from inside, not from a documentary position, in a way that no historical account I’ve encountered has managed to do.
What connects Mantel and Morrison across very different projects is a refusal to use the past as costume. The historical setting isn’t background. It’s the substance. When you finish Wolf Hall, you understand something about power and compromise and survival in the Tudor court that didn’t exist in your understanding before. When you finish Beloved, the phrase “formerly enslaved” no longer carries its usual abstract, documentary weight. It carries weight. The genre at its best does this. The genre at its worst does the opposite: it puts period clothing on characters who think and feel like people from your own era, and calls the result historical.
The third thing historical fiction can do is use the past to say something about the present that can’t be said directly without sounding like an argument.
Pat Barker’s Regeneration, published in 1991, is set at Craiglockhart War Hospital in Scotland in 1917. The psychiatrist W.H.R. Rivers is treating soldiers suffering from shell shock, including the war poet Siegfried Sassoon, who has been sent to the hospital not because he’s having a breakdown but because he published a statement refusing to return to the front and the military needed somewhere to put him that wasn’t a court-martial. The novel is about what war does to men who are asked to survive it, and it’s also about what institutions do when confronted with inconvenient human cost, and Barker uses 1917 to make both arguments without the reader ever feeling lectured to. Rivers is a figure of genuine moral complexity: he helps men recover well enough to be sent back to the conditions that destroyed them, and he knows this, and he does it anyway, and Barker lets the contradiction exist without resolving it for you. That’s a harder thing to do than it looks.
Colson Whitehead’s The Underground Railroad, published in 2016, does something formally bolder. The underground railroad was a real network of people and routes helping enslaved people escape to the north. Whitehead makes it literally real: an actual railroad, underground, with stations and conductors and schedules. The fantastical element lands differently than you’d expect. It doesn’t soften the horror of what Cora, the protagonist, is escaping. It makes the infrastructure of terror more visible, because a literal railroad requires literal infrastructure, and the infrastructure of slavery was always the point. Whitehead isn’t writing about a period. He’s writing about systems that generate and sustain atrocity, and he’s using the past as the clearest possible lens onto a pattern that doesn’t stay in the past.
Viet Thanh Nguyen’s The Sympathizer, published in 2015, takes a different angle again. The narrator is a communist spy who spent the Vietnam War embedded with the South Vietnamese side, and the novel follows him from the fall of Saigon through refugee camps to Los Angeles and eventually back to Vietnam. It’s a novel about colonialism and about what it means to hold two contradictory worldviews inside a single mind, and about what America does to the countries it encounters and the people who survive those encounters. The freshness is in the vantage point. We’ve read Vietnam War narratives told from the American side, the confusion and the casualties and the gradual moral reckoning. We haven’t read many told from the perspective of the people who were the subject of that narrative. Nguyen writes with intelligence and a controlled fury, and enough dark humor that the fury doesn’t collapse into polemic. It’s the kind of book that reorganizes what you think you already understood.
There’s a failure mode the genre produces regularly, and it’s worth naming directly.
It’s the novel that uses history as backdrop rather than substance. A present-day sensibility transplanted into period clothing. The protagonist who holds views that no person of that era could have held without extraordinary context that the novel doesn’t provide. The historical detail accurate enough to create atmosphere but not deep enough to make you feel the logic of the world from the inside. These novels aren’t dishonest exactly. They’re just doing something different from what they appear to be doing. They’re giving you the comfort of history without the instruction of it.
What distinguishes the books worth your time is that the historical circumstances shape the characters rather than merely surrounding them. People think thoughts that are specific to where and when they are. The novel asks you to sit inside that specificity, rather than importing your modern judgment as a corrective. Mantel’s Cromwell doesn’t apologize for the assumptions of the sixteenth century. He operates within them. That’s why he teaches you something.
The books I’ve mentioned don’t agree with each other about what the genre is for. Clavell is interested in how power works across cultural difference. Eco is interested in how knowledge gets controlled and by whom. Mantel is interested in what it costs to be indispensable to the powerful. Morrison is interested in what slavery does to a person’s sense of her own humanity. Barker is interested in the institutional management of inconvenient human cost. Whitehead and Nguyen are interested in what America’s relationship with its own history looks like from the outside.
None of them are comfortable books, in different ways and to very different degrees. The genre has plenty of comfortable entries if that’s what you’re after. But comfort isn’t what the best of it is doing.
If you’ve never read serious historical fiction and you’re wondering where to start, I’d probably press Wolf Hall and Beloved into your hands first. One will show you how much a novel can do with a historical figure you thought you knew. The other will show you how much the genre can cost, which is also a way of understanding how much it can give.
For a broader look at the books that have actually changed how I think, across genres and decades, I’ve written about that elsewhere. And if you want the same question approached from the nonfiction side, about who decided which historical figures we remember and how we remember them, the piece I wrote a few weeks ago about Marcus Aurelius gets at the same machinery from a different angle.
The test I use for any historical novel is simple enough. When I set it down, do I see the present differently? Not the past. The present. If yes, the book did what the genre can do at its best. If I just enjoyed the trip and left the same way I arrived, the book was probably fine. Fine isn’t what I’m recommending here.
Warren Holt writes about ideas and culture for The Sunday Evening Review. He is the author of two books, including “The First Answer Is Usually Wrong.”
