Google's Poetry Translation Software

The more my workload increases, the more I find myself dreaming of books, writing, reading, blogging, immersing myself in works of imaginative writing. But there simply is not enough time. Such is the irony of my life these days, but one result is the scarcity of posts here. I don't want to quit blogging, but sometimes I fear it's too difficult to keep it up. Ah well--it's a new year, so I'll keep trying.

***

Desparapluies, one of my brilliant former undergraduate students, works at Google (I think she's still there!), and I was thinking of her and her honors project, and of the countless works of undergraduate and graduate works I've read, as well as of the vast body of literature out there, including my own modest contributions, that would pose challenges to Google's new Poetry Translation software. Poetry, even the seemingly simplest of it, gives many readers a mental workout, so you need not extrapolate too wildly to consider how difficult it remains for artificial intelligence.

But why? Poetic language in almost every language has traditionally involved prosody, figuration, rhetorical devices, rhyme and other sonic devices, allusions and symbolic registers rooted in the language and culture in which it was produced, and the overall and often intricate interplay between all of these elements, in part because it arose out of orality, for which all of these aspects of poetry are required, and while computers have been increasingly able to perform extraordinary complex intellectual tasks, including readable, often idiomatic translation of prose, poetry and poetic language entails many more potentially insurmountable hurdles.  Even the idea of paraphrasing poetry, whether in translation or not, can present difficulties; what, for example, is the paraphrase--or, to put it another way, a précis or simple meaning rendered in prose--of Stéphane Mallarmé's famous poem, "Ses purs ongles très haut....," a sonnet most likely remembered for its dazzling use of the teleuton "-yx"?

Google software engineer Dmitriy Genzel and his team presented a paper at the Empirical Methods in Natural Language Processing (EMNLP) conference at MIT this past October, in which they focused on the "purely technical challenges around generating translations with fixed rhyme and meter schemes."  Part of the team's debate has centered on the importance of preserving form and meter in translating poetry, and in his blog post Genzel cites Vladimir Nabokov arguments about the impossibility of maintaining such features, while approvingly noting computer scientist Douglas Hofstadter's arguments on behalf of trying to do so.  As anyone who has read my many poetry translations on here or elsewhere knows, I agree wholeheartedly with Hofstadter.