A short screed: Anita Brookner and the writing ‘House Rules’

Something I often say in Creative Writing classes is this: a story (or poem, or novel, or essay) should contain the rules for its own reading. If you have intentions for a piece of writing, then those intentions should be embedded, or encoded in the piece. You should establish your own house rules, give clues as to what you’re trying to do. A dead body and a detective will tend to suggest the likelihood of a crime novel. But it goes beyond this. Take irony. You can’t be ironical without establishing that you’re being ironical. The regular use of the comma splice could be a stylistic choice, or it could be editorial sloppiness or grammatical ineptitude. If it is a choice, I want some implicit indication of this, perhaps in the relationship between content and form, or between narrative voice and character. I want to know you know what you’re doing.

This works well enough inside the Creative Writing course, where the idea of authorial intention is not only viable, but necessary. You can’t mark anything without objective criteria. In Creative Writing courses the universities set out the (quite broad) definitions of what ‘good writing’ is, and students are encouraged to present work that makes sense within that rubric. Make sure I know what it is you’re trying to do, and I’ll mark you accordingly. Embed your own house rules in your writing. Goethe’s three questions of constructive criticism very much apply:

What was the author trying to do?

Did they succeed?

Was it worth doing?

(There is a further complication produced by the current system of teaching, which is that often you already have an idea of what a particular student is trying to do, from discussions in the seminar, and so the encoded intentions, the house rules, go by the bye. This can cloud the marking – and shows the need for second marking, and moderation. Rewarding a piece of work primarily for its intention is clearly academically unsound, no matter that it is, critically speaking, unsounder still.)

Things get more tricky outside of academia, where the critical paradigms are more varied and confused. Two months ago I read my first novel by Anita Brookner, and yesterday I finished my second. The first was her debut, A Start in Life (1981), the second her eleventh, A Closed Eye (1991). I was struck by both books having the same narrative construction, one that I found frustrating, and I found myself asking myself: is this what she intended? It must be, if she did it twice. (I’ll explain what it is in a moment.) But why?

I’m afraid that, if I’d been ‘marking’ it, I’d have marked her down.


“Is it true that Anita Brookner writes exactly the same novel every year?” James Wood has VS Naipaul saying in an interview, in 1994. It is true, Wood replies, and Naipaul says: “How awful, how awful.”

Of course, there’s no problem with writing the same book over and over. After all, nobody complains when Patrick Modiano does it. Modiano’s books, mostly written in the first person, are about uncovering enigmatic aspects of an opaque past, personal and political. Brookner’s books – the two I’ve read, and the third on my shelf seems the same – are written in the third person. They have a central character, but the narrative doesn’t exclusively take their part; others’ points of view are also offered. And here’s what I find odd, structurally, about the books. (Again, I’ve only read two of her novels: perhaps they all work like this; perhaps I’ve just stumbled on the two, or two, that do.)

The two books both seemingly open in media res, with ‘present day’ chapters that introduce the central character. A Start in Life has the famous opening line

Dr Weiss, at forty, knew that her life had been ruined by literature.

A Closed Eye, equally arrestingly, opens with a letter, from 40-something Harriet, a widow living in Switzerland, to the daughter of her oldest friend, long since dead, inviting her to come and visit, and then offers a chapter of her life in France. Chapter 1 of A Start in Life ends:

[Dr Weiss] silently, on certain evenings, let the dusk gather in her small sitting room, propped her head on her hand, and thought back to the play of which she had been entrusted with such an exacting part.

Chapter 2 then opens:

She remembered herself as a pale, neat child…

Similarly, chapter 3 of A Closed Eye opens with:

Unlike Monsieur Papineau, her only friend in this curious aftermath to a life, she could not recall her childhood with anything like the quality of affection which he lavished on the past. And yet she had been quite happy…

What then follows is an extended analepsis, in which the character’s life is gone back over, more or less chronologically, to bring you back up to date. (Analepsis, rather than flashback – it’s impossible to think that the main character, in the novel, is remembering all of it, and in any case, as I’ve just mentioned, the narrative goes into the interior life of characters now dead that she can know nothing of.) The analepsis takes up the vast majority of the novel. A Start in Life returns to forty-year-old Dr Weiss for a mere half a page. A Closed Eye, having set up the idea of young Lizzie coming to stay with Harriet, does bring this to fruition – for a total of seven pages.

What I find strange, or disturbing, about this set-up is that the frame narrative moves forward so little. The usual function of analepsis (or flashbacks) in fiction is to fill in the backstory of a character, so that we can understand them in the main narrative, and so react and interpret when they go on to act in it. Although, of course, in Brookner, nothing does happen; nothing is allowed to happen. Not in that nothing happens, but in that the novel ends.

When I was reading A Start in Life, while my reading brain was taking in and assimilating the whole backstory to Dr Weiss’s life, it was doing so in a functional, forward-looking way; I was telling myself that these scenes were equipping me for understanding Dr Weiss in the present, when we got back to her, either in a prolepsis, cutting short the flashback and jumping back through time, or when the flashback naturally caught back up with the present. Only it didn’t, or not for a long time.


Or, to put it another way: how, when we start reading a book that opens in media res, do we know it is so? I’m not sure of the answer, but I do believe that analysis – which I don’t have the time to do now, trust me – would yield it up. It is, you might say, a tenet of my critical thinking: you know you’re reading a story that opens in media res even before it slips into analepsis. It’s there, encoded in the story. It’s in secret ‘how to guide’ that the author embeds in their writing.

(I’m not saying, by the way, that you can’t actively mislead readers, or break your own rules, but to do that you have to have solidly established rules to start with. The pscyhological thriller relies on twists, but twists only work because the rules of induction, of establishing what the reader is supposed to expect based on what they are given, are there from the off.)

To go back to A Closed Eye. Because of my misreading (or Brookner’s miswriting) of the opening section,  which had Harriet writing a letter to Lizzie and setting up her possible visit, I thought: ah, these scenes in the past will allow me to understand the scenes in the present, to come. Which they do – but if that’s the modus operandi of the novel, then Brookner is giving us 240 pages of backstory to explain seven of ‘story’. No, clearly that’s not the primary purpose of the flashback, but at the time of reading I didn’t know that. I’d made assumptions based on the rules I’d intuited from the opening section, and was, you might say, reading them wrong.


People often complain that nothing happens in Brookner’s novels, which is a) not true and b) invalid. There is passion, of sorts, and illness and drama and death, though certainly these appear less frequently than in most other novels. (You could compare these books to Flaubert’s Un Couer Simple, in which a small life is filled out to the length of a conte, but here the lives are filled out more, the pages roll on and on with the low-level hum of trivial existence. I wrote about Un Coeur Simple here)

And, to answer the second complaint, literature needn’t depend on drama alone for its effects.

Nevertheless, what I find unnerving about these two books is that they seem to be setting up a particular operation – introduce a character in a brief sketch; show how they became that person; introduce them into a plot that will allow the reader to test their assumptions and expectations of that character in a particular situation – but they then only carry out the first two parts of it. It seems to push against the established rules of the novel, when so much of the rest of the book conforms to them, often splendidly.

I am forced, then, to ask myself: Why did she do this? If the analepsis, the backstory, is the ‘meat’ of the novel, why not just begin from the beginning? Why the frame? Why start in media res?

Obviously there are books that operate on such a ratio of frame to flashback: death-bed confessions, or retold tales like Heart of Darkness, but they have different sets of instructions encoded in their openings. The frame is the perfunctory; a setting only, a ground to give context to the figural work to come. Compare, for instance, Elena Ferrante’s Neapolitan Quartet. There, Lenu sets out quite directly and demonstrably to remember and write the story of Lila. The narrative stance of the frame is inescapably backward looking. There will be no moving forward from the present. (There is, but not significantly.)

In Brookner, there is none of this: the implication is that the ‘frame’, the present day scene in which the character is introduced to us, is a vital, living moment, a place in which and from which things might yet happen. The backwards look seems – to me – to be presented as if it were to be followed by a recuperation, a coming back to consciousness, and then a turn of the head and a movement into the future.

Perhaps in the end it’s the static-ness that is the point. By ending her books as she does, Brookner seems to be saying that a life’s narrative can progress to a point at which nothing further will happen, or at least – as with Zeno’s arrow approaching the target – that the progress will be so incremental as to defy observable measurement. In a way, this is an ideal counterargument to those who say that ‘nothing happens’ in Brookner. Nothing does happen, but only after the end of the novel. She picks up her characters at the very point at which they have reached the point at which nothing further will happen, and she then backtracks to trace quite how that comes to be. The novels don’t really close, as frame narratives usually do – instead they end on a single open, sustained chord, projected unchanging into the future. This moment is the same as all the moments that follow.


(What Brookner does with these two novels is, in narrative terms, counterintuitive, even unsatisfactory. I feel the need to ask why she might have done it. So I suppose this essay is the attempt to work out what that reason might be. Two things to do: go back and look again at those opening chapters: how does an author encode the instruction to the reader to read their book opening as happening in media res? And, naturally, read more Brookner, to see if this is a trick she plays compulsively, elsewhere.)


One comment

  1. Pingback: A short screed: Anita Brookner and the writing ...

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s