Starting a Memoir: The Difference Between What Happened and What It Is About

The most useful sentence I have read about this form is Vivian Gornick's, in The Situation and the Story: the situation is the context, the plot, the circumstance — and the story is the emotional experience that preoccupies the writer, the insight, the wisdom, whatever it is the writer has come to say. Nearly every failed memoir draft I have read, including my own attempts at writing about my own life, fails at exactly that seam. The situation is all there. The story is missing.

This matters because the situation feels like the hard part. You lived it, you remember it, you can reconstruct the sequence. And then you have forty pages in which everything is accurate, nothing is untrue, and no reader has any idea why they are being told it.

Autobiography answers a different question

The distinction people draw between memoir and autobiography is usually presented as one of length or scope, which undersells it. An autobiography is organised around a life and implicitly claims that the life is interesting. A memoir is organised around a question and claims only that the question is.

Joan Didion's The Year of Magical Thinking covers roughly a year and is about what grief does to a rational person's relationship with cause and effect. Helen Macdonald's H is for Hawk is about her father's death and about training a goshawk, and it is the collision between the two that makes a book rather than a diary. Sonali Deraniyagala's Wave covers the loss of her entire family in the 2004 tsunami and refuses almost all of the consolations the form usually offers, which is itself the shape of the book.

None of these is a life story. Each is a piece of life put under a specific pressure.

So the first question is not "where do I begin." It is: what do I not understand about this yet? If you can already state the lesson in a sentence, you probably have an essay-length anecdote rather than a book, because a memoir written from a settled conclusion reads as a lecture with scenery.

Memory is unreliable, and that is material

New memoirists tend to panic about accuracy. Did the conversation happen in the kitchen or the hall. Was it raining. Did he say that, or is that what I have made him say over twenty years of retelling.

The panic is reasonable and the usual response to it is wrong. Trying to write only what you can verify produces prose that is both thin and faintly dishonest, because the certainty it projects is fake anyway. Memory is reconstructive; each recall rewrites slightly; the version you have is the version you have rehearsed.

The better move is to put the uncertainty in the text. "I have told this so many times that I can no longer find the original underneath it" is a true sentence, and it is more interesting than a confidently reported dialogue you are quietly inventing. Some of the best memoirists make this a structural principle — the narrator's doubt about their own record becomes part of the subject.

What the reader is owed is a stable contract. If you have compressed three conversations into one, or changed a name, say so somewhere, usually in a brief note. Readers are generous about compression and unforgiving about discovering it by accident.

The therapeutic claim, handled honestly

You will often see memoir sold on its psychological benefits, usually with the expressive-writing research attached and usually overstated.

The underlying literature is real. James Pennebaker, whose experiments defined the paradigm, has reviewed decades of work on writing about emotional upheavals, and across a large number of studies there are measurable if modest associations with improved wellbeing and some health measures. Pennebaker himself is notably careful about the size of those effects and about what is driving them, and the paradigm he studied is short, private, unstructured writing — fifteen or twenty minutes, nobody reading it — which is not the same activity as spending two years shaping a publishable book about your childhood.

In fact the two can pull against each other. Private writing lets you be incoherent, which appears to be part of what helps. Writing for a reader forces you to be shapely, and shaping a painful event into a satisfying narrative is not always the same as processing it.

So: writing about your life may well be good for you. Writing a memoir is a different undertaking with different demands, and going into it expecting it to be therapy is one of the more common ways people get hurt.

What to write first

Not the beginning. The beginning is the part you understand least, because you have not yet worked out what the book is.

Write the scene you cannot stop thinking about. Write it badly and at length, with everything in it: the room, what people were wearing, what was said, what you thought and did not say, what you were embarrassed by. Do not decide yet whether it belongs in the book. You are looking for heat, and heat is more reliably found than planned.

Then write four or five more. At that point something usually becomes visible — a preoccupation you did not know you had, a person who keeps appearing, a question that every scene is circling. That is closer to your story than anything you would have produced by outlining.

A few things reliably make those scenes better. Specificity of the physical world, because it is what convinces: not "the house smelled familiar" but the particular thing it smelled of. Dialogue, even approximate, because people reveal themselves in speech faster than in description. And the willingness to render your younger self as a character with limited understanding, rather than as a vehicle for what you know now. The gap between what the narrator knows and what the younger self knew is where most of the emotional force of the form lives.

The people in it

This is the part nobody prepares you for, and it is not a craft problem.

Everyone you write about is a real person who did not agree to be a character. Some of them will read it. Some of them remember it differently and are not wrong. The standard advice — change names, wait until they are dead, write as if nobody will read it and decide later — is all reasonable and none of it resolves the underlying discomfort, which is that honest memoir involves saying things about people that they would prefer you did not say.

Two things I think are defensible. First, apply the same unflattering light to yourself that you apply to anyone else; the memoirs that curdle are the ones where the narrator is the only person behaving well. Second, distinguish between an account that is unflattering and an account that is settling a score. Readers can tell, and the second one makes them side with your antagonist.

There are also situations where the honest answer is that the book should not be published, or not yet, or not while a particular person is alive. That is a legitimate outcome and it does not make the writing wasted.

On not needing permission

The framing that everyone has a story worth telling is well meant and slightly unhelpful, because it locates the value in the events. A childhood is not interesting because it was difficult, and an ordinary life is not disqualified because it was ordinary. Annie Dillard built An American Childhood largely out of the sensation of waking up into consciousness, which is available to everybody.

What makes a memoir worth reading is attention: the willingness to look at something you would rather glance past, and the patience to find language for it that is more precise than the language you have used before. That is a skill you can practise, on material nobody else would call dramatic, starting with a single scene you have never quite managed to explain to anyone.