Christopher Nolan’s “The Odyssey” sparks a debate about translation and transcreation

In 2017, Emily Wilson published an English translation of Homer’s Odyssey that became widely read inside and outside classics departments. Christopher Nolan cited that translation during the press tour for his own film adaptation, released in July 2026. Weeks after the film opened, Wilson published an essay in the London Review of Books calling the script “abysmal” and writing that she “would be ashamed to have written any part of this script.” The reaction split online audiences into two camps. One side defended Wilson’s authority on the source material. The other accused her of judging the film against a version of the Odyssey that exists only in her own head. Both camps are missing the more useful question underneath the argument.

Emily Wilson translated a poem, and Christopher Nolan transcreated a movie

Wilson’s translation and Nolan’s film solve different problems. A translation renders one language into another while preserving as much of the original meaning, tone, and structure as the target language allows. Wilson’s Odyssey keeps Homer’s ambiguity about Odysseus, a man whose cunning and cruelty never quite resolve into a straightforward hero. Nolan’s film adapts a poem composed almost 3,000 years ago into a theatrical release aimed at a global audience buying tickets in 2026. Those are different jobs. The second one is akin to transcreation, the work of taking a source and rebuilding it for a new medium, a new audience, and a new format, with license to cut, compress, and reshape that a translator does not have.

Localization teams give these jobs separate names for a reason. A patent filing needs a translator who will not introduce ambiguity the source document does not contain. A marketing campaign needs a transcreator willing to replace a joke that will not land in Portuguese with one that will, even if the words no longer match the English original. That distinction shows up in staffing decisions all the time among localization teams. Certified translation work and creative transcreation work go to people with different training, because a translator judged by transcreation standards drifts from the source, and a transcreator judged by translation standards produces copy too literal to land with the audience it’s written for.

Separating craft criticism from fidelity criticism

Wilson’s review mixes two different arguments. Most of it, the complaint that the film has “nothing convincing to say about time, memory, history, war,” that its narrative structure is gimmicky, that characters act without motivation the story itself provides, is craft criticism aimed at whether the movie holds together on its own terms. None of that requires her translation as a reference point. A critic with no attachment to Homer’s Greek could raise the same objections and be judging the film only against itself.

Wilson also argues that Nolan’s Odysseus lost the moral unpredictability Homer gave him, flattened from a complicated, ambiguous figure into a more familiar hero carrying guilt. That comparison does invoke her own reading of the character, and it is the one place where Nolan’s defenders have a point: there, she is measuring the film against her translation rather than against its own internal logic.

The trouble is that critics use that single strand to discredit the rest. Calling the whole review an attempt to judge the movie against her own version treats one fidelity-flavored complaint as though it explained all of them. It does not explain why the narrative structure is gimmicky or why characters lack motivation, since neither objection depends on comparing the film to her book at all.

That distinction, between critiquing a work on its own terms and measuring it against a source it publicly invoked, is one localization teams have to keep straight in their own judgment, not just in their clients’. A transcreated ad campaign gets judged by the client against the original brief. A dubbed film gets judged by fans against the subtitled version they already know. A reviewer, a translator, or a transcreator can drift between those two standards inside a single piece of feedback without noticing the shift. The useful question is which standard is doing the work in any given sentence of criticism, and whether that standard was agreed to before the work started.

Fidelity and expectation

A feature film, whatever its individual merits, cannot carry that same fidelity the way an entire translated book can, because adapting any long text for a single sitting requires cutting and compressing in ways a translation does not. Localization teams face a smaller version of this choice every week. A client might ask for a literal rendering of a tagline that will confuse a target market. Or they might ask for a transcreated version that keeps the intent and changes every word. Neither choice is wrong by itself. The failure happens when a client is told they are getting one and receives the other, and finds out after the work has already shipped. Before a single word gets adapted, the question to settle is which one was promised, translation or transcreation, and whether the finished work delivers on that promise instead of a different one nobody agreed to.