In November 1998, Joan Didion was not happy. It had recently been announced that Ernest Hemingway’s last novel, True At First Light was to be published in the coming months, some thirty-eight years after the author’s death. Rather than being delighted about the prospect of another work of fiction from the titan of American letters, Didion was up in arms. For her, the publication of Hemingway’s posthumous works — the novels Islands in the Stream (1970), Garden of Eden (1986), and a wealth of non-fiction including his memoir of his years in Paris, A Moveable Feast (1964) — were part of a ‘systematic creation of a marketable product’: ‘a discrete body of work different in kind from, and in fact tending to obscure, the body of work published by Hemingway in his lifetime.’

Rather than adding to our understanding of the author, the publication of his manuscripts and literary detritus was, for Didion, nothing more than a ‘process of branding’ that detracted from the meticulous nature of Hemingway’s other works. In her words: ‘You care about the punctuation or you don’t, and Hemingway did. You care about the ‘ands’ and the ‘buts’ or you don’t, and Hemingway did. You think something is in shape to be published or you don’t, and Hemingway didn’t.’

But how could something by Hemingway be published and not have his trademark attention to detail, sparsity, and punctuation? The answer is, rather disconcertingly, that many of the posthumous works with Hemingway’s name on the cover are not, truly, ‘by Hemingway’ in the sense that readers might expect.

The Garden of Eden — published in 1986, and recently the subject of Sonia Overall’s novel Eden (2022) — stands at a mere 247 pages in its first edition. But the manuscripts left behind — three, in total — vary dramatically in length between 400 pages and 1,500 pages, with one as long as 200,000 words. The manuscripts were, according to Hemingway’s biographer Kenneth S. Lynn, ‘irreconcilable’ in plot and subject.

So how did The Garden of Eden, which readers can pick up in a bookshop today, come into being? It could hardly have been Hemingway’s ghost who exercised control over what the publisher’s note rather disingenuously describes as ‘cuts in the manuscript,’ ‘routine copy-editing corrections,’ and ‘minor interpolations for clarity and consistency.’

This act of literary surgery — rather than being ‘routine’ edits, the finished book excised two-thirds of the manuscript material — was the work of editors at Hemingway’s publishers, Charles Scribners and Sons. In an interview in the New York Times in 1985 just before the book was published, one editor, Tom Jenks, admits to removing a whole subplot he didn’t believe Hemingway had ‘integrated into the main body of the novel.’

Without having access to the manuscript, it is hard to say whether Jenks’s judgement was correct. But no matter how good an editor is or, as Jenks puts it, how much they try to make the ‘same sorts of decisions’ as the (dead) author would, there is still the unavoidable —and disconcerting — fact that these edits were made without Hemingway’s knowledge. If you’re feeling kind, the book is a brilliant, hard-fought work of literary necromancy: bringing an author back to life by ruthlessly editing their dead prose. If you’re feeling cruel, the book is little more than an editorial version of Frankenstein’s monster: something sewn together using patches of manuscript, without the hand of the author to make it truly whole.

This ghoulish editorial work has ramifications beyond simple squabbles over literary style and who has the right to edit a dead author (Hemingway’s ‘Africa manuscript’ was edited into two books, True At First Light, by his son, Patrick Hemingway, and Under Kilimanjaro, by Hemingway scholars). Towards the end of his career Hemingway was exploring love, affection, and sexual relationships in new ways: Garden of Eden is, amongst other things, a novel of bisexuality and fluid sexual attraction. One of his characters, Catherine, cuts her hair to look like her husband David, and has him dye his so they appear like ‘twins’ when having sex. But, for some critics, the edits made to the manuscript denied the full possibility of Hemingway’s new thinking. The author Barbara Probst Solomon, writing in The New Republic in 1987, did not hold back:

He meant Eden to have been his final summation on art and literature, on the nature of love and the body, on the possibilities of human life. But you won’t find any of these strong conceptions in the book that Scribner’s has published in his name. […] in almost no significant respect is this book the author’s.

It is, again, hard to judge without access to the manuscript — and, indeed, the relationship that is left is stronger than Solomon would admit (‘But now I’m a boy too and I can do anything and anything and anything…’). But her claim that the book is in ‘no significant respect’ Hemingway’s is hard to forget.

But Hemingway is not the only author whose work has suffered this fate after his death. During his lifetime, Vladimir Nabokov was notoriously controlling of his literary output. His wife, Vera, had to prevent him from burning unfinished drafts of his masterpiece Lolita (1955). And, when he died in 1977, he was no less fastidious: he left precise instructions for his literary executors (his son Dmitri and Vera) to burn the remains of his work on The Original of Laura. But, by 2009, the book was on sale.

Dmitri Nabokov was eloquent on the difficulty of publishing his father’s work: he did not want to be a ‘literary arsonist,’ but he also did not want to betray his father’s wishes. Whether it was a desire to give the literary community a last gem or, more prosaically, a desire to make a last buck out of his paternal estate, his adherence to his father’s desires did not last.

The case of The Original of Laura is different to the many posthumous Hemingway publications: the book is not edited and is, instead, a facsimile of the index cards Nabokov wrote on. But this fidelity did not stop the volume from being lambasted: Martin Amis was typically succinct when he wrote ‘When a writer starts to come off the rails, you expect skidmarks and broken glass; with Nabokov, naturally, the eruption is on the scale of a nuclear accident.’

More recently, David Foster Wallace’s The Pale King was published in 2011 — three years after his death. In this case, the finished book is the work of his editor and friend Michael Pietsch, who had meticulously worked with him on the many edits needed for his other works — amongst them, Infinite Jest (1996). Foster Wallace had left the material for The Pale King in a computer file where it could be found, but ‘had not indicated a plan for the novel’s overall structure.’  Wallace was known for his non-linear, non-plot based writing — ‘overall structure’ was always going to be a challenge, even for the author himself.

Pietsch’s editing is quite different to either that which was foisted upon Hemingway or Nabokov: it was clear that Foster Wallace wanted his last novel to be published, and Pietsch had a long relationship with him. To one request for a cut to Infinite Jest, Foster Wallace replied with ‘p. 785ff –  I can give you 5,000 words of theoretico–structural arguments for this, but let’s spare one another, shall we?’ This back-and-forth was important. When editing his later work, Pietsch said ‘I didn’t feel I had the liberty to edit his words without him there to respond to them.’

Pietsch aimed for ‘completeness’ whereas other editors of posthumous works aimed for concision. But, despite the differences between them, these books by Hemingway, Nabokov, and Foster Wallace exist in the same kind of hinterland: novels by authors who died before they could either finish them or discard them. These spectral texts give us an insight into the last years and thoughts of our beloved writers, but we should remember the other hands that have shaped them.