By Susanna Clarke

All Posts Contain Spoilers

I think ‘Jonathan Strange and Mr. Norrell’ is a masterpiece.

I’ve read it several times over the years – it is one of those rare books that gets better every time you read it. It’s magnificent: subtle and funny and sad and totally ingenious, one of those books so enjoyable to read that you look forward to rereading it.

‘Jonathan Strange’ came out in 2004, Susanna Clarke’s first novel. We, her fans, waited fifteen years for her to publish another, and when ‘Piranesi’ finally landed, in 2020, I think it was…thinner than we had all hoped. ‘Jonathan Strange’ is a great honking book (my copy clocks in at 850 pages) – the kind of book you can stretch out into. ‘Piranesi’ is a tight 275 pages, clearly not the leisurely stroll of Clarke’s first novel.

I often save books I’m particularly excited about for treats – I have saved “Piranesi” for a year now. Holding off has allowed me to gather a sense of the public’s response to it, and, while most people seem to love it, there is one thing I have heard over and over again: it is very different from ‘Jonathan Strange’.

Which is obviously emotionally complicated news for everyone who loved ‘Jonathan Strange’. The relationship between readers and authors is very, very complicated. On the one hand, you, as a reader, want the authors that you love to be able to grow, to experiment and try new things and develop. Some people (Tom Clancy’s fans) want to read the same exact book over and over again; everyone else appreciates some variety.

On the other hand, though, you love the authors you love for a reason, and it is not in your interest, as their reader, to encourage them to grow out of doing the thing that you loved in the first place. And authorial growth is rather like random mutation: a few changes are beneficial; most are deleterious.

The answer to this problem is trust. Over the course of their authorial life, writers earn their readers’ trust. They demonstrate reliability in whatever traits have made them loved, and readers learn that, even as the author changes and grows, their work will be still worth reading. The amount of trust is contingent on two things: how reliable the author is, and how outstanding they are, relative to other authors, at the thing which has made them loved in the first place. As a equation, it might look something like this:

Trust = consistency x unique value

The unique value is an expression of the thing that this author does that makes you, the reader, love them especially, that sets them apart from other authors. It might be understood as something like

Unique value = rarity among authors x execution

Where how uniquely prized an author is, is a function both of how good they are at what they do, and how many other people do it.

As an example, let’s take Agatha Christie. Christie is among the most trustworthy authors, probably, who ever lived. But let’s break her down by the components of her Trust Equation:

Trust = consistency x (rarity x execution)

Rarity: Low, basically non-existent. Murder mysteries are common as dirt.

Execution: Medium. Strong but not astonishing. She has a few masterpieces, it’s true, but most of her books are solid rather than shining.

Consistency: 100%. This is the variable doing the heavy-lifting for Christie. She just doesn’t write bad books, and she doesn’t go off-script. If you like what she’s selling, you know you’re going to get it every single time. The trust, in this case, is basically entirely a function of Christie’s consistency.

Let’s take another example: Jane Austen. What do the variables in Austen’s equation look like?

Consistency: Medium. ‘Pride and Prejudice’ is one of the greatest books every written, but if all Jane Austen had written was ‘Persuasion’, no one alive would know her name.

Rarity: Medium-high. The satire of manners is not uncommon, but the gently-biting, feminine perspective of Austen’s works belongs to her and her alone.

Execution: Stellar. This is where Jane Austen shines. She doesn’t pull it off every single time, but when she does, there is literally no one better. This is why Austen has high reader trust: it’s worth reading any of her books because sometimes they will turn out to be ‘Pride and Prejudice’.

Anyway, why am I talking about this in an article about ‘Piranesi’?

New authors are, obviously, deeply handicapped according to this equation: their consistency is N/A. Worse: you don’t even know yet what their traits will be. It may be that the things you loved about their first work were completely incidental to their project. This is what makes the release of an author’s second book so suspenseful: it is only in that second book that you really begin to know who they are.

On one level, ‘Piranesi’ is a very difficult book to describe. Our narrator, called Piranesi although he is pretty sure that it’s not his real name, lives in the House. The House is vast: an endless series of huge rooms filled with statues. Some rooms are flooded; some are open to the air. Great tides sweep through the House, flooding whole chambers and receding. Piranesi loves the House; his days are devoting to fishing for food and mapping the endless halls, charting the tides, and caring for the dead. As far as he knows, there have only been fifteen people in the House, ever: himself, the thirteen skeletons he cares for, and the Other. The Other is aloof and strange, but, as the only other living person in the house, he is also Piranesi’s only friend.

One day, though, the Other warns Piranesi: another person has arrived in the House, Sixteen. Sixteen is wandering the halls of the house, trying to find Piranesi, and, according to the Other, destroy him.

‘Piranesi’ is a good novel, but it’s not, I think, a great one. And here is the thing that surprised me: though it is very different from ‘Jonathan Strange’, that is not the problem with it. In fact, ‘Piranesi’ would have been a better novel if it had been less like ‘Jonathan Strange and Mr. Norrell’.

‘Jonathan Strange’ is, at heart, a cozy novel. It is Dickensian: it lives in detail, in the specific, closely-observed eccentricities of its characters. It is tangible, concrete, minute. It is not a mystery, and it is not other-worldly. That is exactly its charm: it is a story of magic set in our world, one we know well. It is the juxtaposition of quotidian humanity with the supernatural.

‘Piranesi’ is the opposite: it is a spooky, atmospheric novel, alien and vague. ‘Jonathan Strange’ inhabits fire-lit drawing rooms in London townhouses; ‘Piranesi’ echoes in vast, cold spaces. And those echo-y bits are the best parts of the novel: Piranesi, wandering alone through his House, showing it to us.

As I said, on the one hand, ‘Piranesi’ is a difficult novel to describe. On the other hand, it is very simple: it is a Whodunit. What is the House, who is Piranesi, who is the Other, and how did they get there? These are the animating mysteries of the book, and Clarke answers every single one of them. She answers them clearly: concretely, specifically and very unmagically.

And it’s a let-down. The imaginative premise of ‘Piranesi’ is wonderful, majestic and unnerving and grand. When it is solved, it is just a novel of people being terrible to each other in strange places. It is as though Clarke’s mind, which is so rigorous and thorough, could not inhabit the mystery of the House: she needed to solve it for us, but in solving it, she diminished it.

Ever since I finished ‘Piranesi’, I have been wondering whether I would have liked it more or less if I had not read ‘Jonathan Strange’. It feels terrible to hold the goodness of a first book against an author, but it’s impossible not to compare.

I think I would have liked it more, but, after way too much thought, I’ve realized that that is not the right question to be asking. The question I should have been asking is: does ‘Piranesi’ make me more or less likely to read Clarke’s next book?

The answer is definitely yes – I will read Clarke’s next book. Whatever else, both of these books are products of a capacious and thorough imagination – I’m interested to see what she comes up with next.

Perhaps the capacity to surprise is an authorial trait. Perhaps you can learn to trust that an author is capable of re-inventing herself with expertise, that her mind is solving different problems in different ways, that she is building new and better worlds in which to solve those problems. Perhaps, the capacity to surprise will end up being Clarke’s authorial project, the special trait which will make her beloved.

I kind of hope it is. Maybe she’ll never write another ‘Jonathan Strange’, but that’s OK, I think I like her anyway. I don’t trust her totally yet, but we’re moving in the right direction.