Our thinking about the ethical questions regarding posthumous publication (something I've been thinking about since the publication of Nabokov's notecards for/as The Original of Laura) might benefit from a categorization of the various types of posthumous books. I think we need six categories:
1. Books finished and prepared for publication by the author but published only after his death. (Forster's Maurice, the English translation of Sebald's Nach der Natur.) No ethical questions should arise in this case, provided the book is published as the author intended.
2. Books finished by the author but not edited and published until after her death. (The later volumes of Proust's Recherche; Ellmann's biography of Wilde.) There is no significant ethical problem here, but such books will always be read under a thin cloud of suspicion about possible changes the author might have made had she lived through the editing stage.
3. Books left unfinished at the author's death and published as unfinished works. (Kafka's The Castle, Musil's The Man Without Qualities, Ellison's Three Days Before the Shooting, Wallace's The Pale King.) This is a much more troubling category. The ethics of publication seems to be directly proportional to the manuscript's level of completion. Kafka's The Trial seems a much more ethically acceptable publication than Nabokov's The Original of Laura. (However, for both books, see category 5 below.)
4. Books posthumously edited from much longer, unfinished manuscripts. (Hemingway's The Garden of Eden; Ellison's Juneteenth). This category is ethically problematic, to say the least. I loved The Garden of Eden, but a compelling argument can be made that it should never have been published except in an edition of the complete manuscript Hemingway left at his death, a manuscript that reportedly continues the story beyond the scope of the published 'novel.'
5. Books published contrary to the express wishes of the author. (Virgil's Aeneid; Kafka's works; Nabokov's Laura). I would not wish to have lived in a 20th century without Kafka's fictions, but the publication of those works against the author's deathbed insistence that they be destroyed was clearly unethical. (None of Max Brod's excuses convince me, but the world owes him an unpayable debt for his betrayal--the same, sadly, cannot be said of the late Dmitri Nabokov.)
6. Books neither prepared for publication by nor envisioned by the author. (The Letters of James Joyce; Flaubert in Egypt.) I enjoy leafing through the published letters and diaries of great writers, and part of that enjoyment surely arises from the ethically questionable nature of my snooping. Again, I would not have wanted to live in a 20th century without Joyce's erotic letters to Nora, but the publication of these letters is almost certainly unethical--provided that we base our ethical judgments (as I have throughout this post) on the apparent intentions of the writers. Given both the highly problematic nature of determining intentions and the number of great works this basis would eliminate from the canon, it appears that we require a different basis for judgment.