Reading the last page first: spoilers, suspense and pleasure

Every book group I have known has at least one of them. Somebody who, on the first evening with a new novel, quietly turns to the back and reads the last page. They usually confess it in a lowered voice, as if admitting to eating dessert first. The rest of the group reacts with a mix of horror and curiosity, and then someone asks the real question: doesn't that ruin it?

My honest answer is that it depends on what the book is selling. Some novels are built so that not knowing is the whole engine. Many more are built on something else, and for those, knowing the ending changes how you read without spoiling much at all. Sometimes it makes the reading better.

Suspense and surprise are not the same thing

It helps to separate two pleasures that we tend to lump together.

Surprise is a single event. You did not see it coming, and then it arrives. It works once. You cannot be surprised by the same turn twice, which is why a twist can be destroyed by one careless sentence from a friend.

Suspense is a state you live in for pages at a time. It is the tension of caring about people while their situation is undecided, or while you know something they do not. Alfred Hitchcock famously explained this with a bomb under a table: show the audience the bomb, and a dull conversation becomes unbearable. The suspense comes from knowing, not from not knowing.

That is the case for the last-page reader. If you know a marriage ends badly, every tender scene early in the book carries a second weight. If you know a character dies, their ordinary afternoons become precious. Shakespeare understood this. The prologue to Romeo and Juliet tells the audience before the first scene that "a pair of star-cross'd lovers take their life." Nobody in four hundred years has walked out because the ending was given away. The play is not asking whether they die. It is asking how, and whether anyone could have stopped it.

What a spoiler actually takes away

A spoiler takes three things, and it is worth naming them, because most novels depend on only one or two.

1. The answer to a question the book poses openly. Who killed him? Will she leave? Does the letter arrive?
2. A reinterpretation the book has prepared in secret. The narrator was lying, the dead woman is alive, the whole story was told from the wrong side of the door.
3. The particular texture of a first encounter: the moment when you turn a page and feel the floor shift.

What a spoiler leaves alone is everything else. The sentences. The people. The way a writer handles time, or a minor character, or a single room. The thinking the book makes you do. A plot summary of Middlemarch tells you who marries whom, and the novel loses almost nothing by it, because its pleasures are in the watching, not the finding out.

Gabriel García Márquez made this into a structure. Chronicle of a Death Foretold (1981) tells you in its very first sentence that Santiago Nasar is going to be killed, and then spends the whole short novel showing how a town full of people who knew it was coming let it happen. Muriel Spark does something similar in The Prime of Miss Jean Brodie (1961), flashing forward to tell you, well before the end, which of Miss Brodie's girls will betray her. The questions that remain (why, and what it meant, and who was to blame) are harder and more interesting than the one Spark gave away.

Beatrix Hallam's field guide to how great novels end makes a related point: the kind of ending a book attempts tells you what kind of book it is. A quick look at the last page can be a way of reading the contract before you sign it.

Is there evidence?

There is a study that gets cited whenever this comes up, and it is worth describing carefully, because it is often overstated. In 2011 two psychologists at the University of California, San Diego, Jonathan Leavitt and Nicholas Christenfeld, published a short paper in Psychological Science with the cheerful title "Story Spoilers Don't Spoil Stories". They ran three experiments, each using short stories from a different genre, and gave some readers the outcome in advance. On the whole, the readers who had been told the ending reported enjoying the stories more, not less.

That is an interesting result, and it fits the experience of rereaders. But it is one set of experiments on short stories, and it has not gone unchallenged. In 2016 another team (William Levine, Michelle Betzner and Kevin Autry) published a close replication with different stories in the journal Discourse Processes, and found the opposite: spoilers given before a story reduced enjoyment, while spoilers given partway through made no clear difference. The fair summary is that the science is unsettled, and that the effect probably depends on the kind of story and the kind of spoiler. Nobody has proved that spoilers are good for you. Nobody has proved they always hurt, either.

When peeking is a bad idea

There are novels where I would beg a friend not to look. They share one feature: the ending does not just complete the book, it rewrites it.

• Classic whodunits. In Agatha Christie's The Murder of Roger Ackroyd (1926) the solution is the point. Knowing it turns the book into a technical exercise in how she did it, which is fun, but only the second time.
• Novels that turn on a late revelation about the narration itself. Ian McEwan's Atonement (2001) is the example most book groups know. I will not say what happens, only that the last section changes what you think you have been reading, and that this works best if you arrive there unprepared.
• Books whose world is a secret the narrator holds back. Kazuo Ishiguro's Never Let Me Go (2005) releases the truth about its characters slowly and in a deliberately muffled way. The pleasure, and the dread, is in piecing it together alongside Kathy, who knows more than she says.

A rough test: if you could describe the book's ending in one sentence and a friend who had not read it would gasp, leave the last page alone. If the one-sentence ending would make them shrug ("they don't end up together," "he dies old"), you can probably look without losing much.

Rereading shows the same effect

Anyone who rereads already knows what last-page readers are talking about. On a second reading of a novel, nothing can surprise you, and yet the best books get better. Vladimir Nabokov said it bluntly in his Lectures on Literature: "one cannot read a book: one can only reread it."

What happens on a reread is that your attention, freed from the plot, goes elsewhere. You notice that a character's first line already contains her last choice. You see the author planting things. You stop reading for what happens and start reading for how it happens, and that is where most literary fiction keeps its gold.

A long, crowded novel can be easier on the second reading for exactly this reason. Our reader's guide to Salvatore Scibona's The End suggests reading its opening section twice before going on, and the novel itself is built to be understood backward: it starts on a single carnival day in 1953, travels back through decades, and returns to that day with everything you have learned. Knowing roughly where it is heading does not hurt that book at all. It helps you hold the threads.

A worked example: one evening, two readers

Here is how the difference plays out in practice. Take a group reading Never Let Me Go. Say one member, Amira, has read the last chapter first, and another, Joe, has not.

At the first meeting, covering the Hailsham chapters, Joe is reading for the mystery. He is asking what the school is for, why the guardians behave so oddly, what "donations" means. His questions are good ones and he is fully absorbed.

Amira already knows the broad answer. So she is reading for something else: how a child absorbs a terrible fact without quite hearing it. She notices how often Kathy says that they were "told and not told." She notices the small kindnesses that look, in the light of the ending, like grief.

Neither reader is having the wrong experience. But they cannot have the same conversation, and a host who does not notice that will end up with one of them bored and the other spoiled. Which brings me to book groups.

Advice for book groups

Groups fight about spoilers more than about anything else except snacks. A few habits keep the peace.

1. Decide the rule out loud at the start. "We discuss the whole book at each meeting" and "we only discuss up to this week's page" are both fine. Unspoken assumptions are what cause trouble.
2. Split long books into marked sections, and put the boundary in the meeting invitation: "Thursday, up to the end of Part Two."
3. Give last-page readers a role. They make excellent observers of foreshadowing. Ask them, near the end of a meeting, "What did you notice early on that the rest of us will see later?" and let them answer without details.
4. Hold twist-dependent books to a stricter rule. For Atonement or a Christie, nobody discusses the ending until everyone has finished, and the last meeting starts with a show of hands.
5. Flag spoilers in your group chat the way reviewers do: a short line before the discussion, and a few blank lines underneath. It costs nothing.
6. For a rereading group, drop the rule entirely. If everyone has read it before, the ending is common ground, and the conversation can start there.

So, should you read the last page first?

For a novel whose pleasures are in its sentences, its people and its patterns, peeking is a reasonable choice and sometimes a generous one. It lets you read like a rereader the first time through. For a novel built around a single reversal, it is a waste of the one thing that book can only give you once.

The odd thing is that most last-page readers already know this. They look at the end of a family saga and leave the end of a mystery alone. The habit is less a vice than a way of choosing which kind of attention to bring. If you want more on that kind of attention, the essays on our reading life shelf keep returning to it. And if you are the one in your group who peeks, you are in good company. You are just reading the book twice, starting with the second time.