In her recent review of the reissue of Andrey Platonov's Happy Moscow (NYRB Classics), Christiane Craig observes that "The work of translating Platonov must demand an almost inhuman attention to the particular structures of the Russian language and to how the subversion of these structures might serve to transform thought." This is because Platonov "seeks to demonstrate the natural limits of his own language." Craig quotes Joseph Brodsky on Platonov:
He will lead the sentence into some kind of logical dead-end. Always. Consequently, in order to comprehend what he is saying, you have to sort of “back” from the dead-end and then to realize what brought you to that dead-end. And you realize that this is the grammar, the very grammar, of the Russian language itself.
Craig adds that translating Platonov "should also require superior imagination, a sense of how to remake Platonov’s 'supermarket' in English, to reconstitute the same 'variety' that Brodsky observes in the original." I would say that, if Brodsky's description of Platonov's practice is accurate (I assume that it is), then all the imagination in the world isn't going to provide those of us who must read a writer like Platonov in translation with anything like the actual work the author composed--or, perhaps more to the point, is only going to give us something that is "like" that work. (The problem seems to me particularly acute with a language such as Russian, which is arguably farther removed from English than, say, German, or even the Romance languages.)
Perhaps that is an obvious enough point, and represents a state of affairs we simply have to live with unless we're going to make the effort to learn the language used by every writer we want to read writing in a language we don't currently know. However, I do think that in reviews and criticism of translated fiction this reality is sometimes not acknowledged, or is even ignored, especially in those reviews that claim to speak of the writer's "style" or of the effects of language more generally. If a writer's style interrogates the "very grammar" of its language, if doing justice to the writer requires attention to the "particular structures" of that language, then shouldn't the critic know something about that grammar and those structures in order to assess the writer's ambitions and accomplishments for us? How can we appreciate a writer who explores "the natural limits of his own language" if we don't know what those limits are and the critic can't show us?
I am not posing this as a problem with translation, as if we could do without translated works because they will inevitably fall short of fully representing the "real" text of which they are a version, but with criticism. In our fallen world, translation of fiction (and poetry) from languages other than our own is both imperative and what we're stuck with. But unless the critic is also fluent in the language from which the work is translated, and can perhaps delineate in a general way how the translator has provided "structures" that do or do not adequately approximate the structures of the original, discussions of style or of specific linguistic devices seem to me pointless, if not outright deceptive. At best they confuse the skill of the translator with the skill of the author, while at worst they encourage the notion that all features of the literary work that originate in the distinctive features of a particular language can be subsumed to some universal practice of "good writing" that just doesn't exist, thus erasing linguistic differences.
In my reviews of translated fiction, I confine myself to commenting on observable formal features that likely are transferable beyond linguistic barriers, although of course some elements of form--point of view, for example--are inextricably linked to the linguistic resources of a specific language. I don't believe I can responsibly go beyond describing my experience of reading the work being attentive to formal qualities not obviously dependent on the discrete properties of an unfamiliar language. If this suggests that criticism of translated literature is inherently incomplete, it seems to me necessary to acknowledge this limitation in order to resist making claims that simply can't be supported.The reality of such a limitation does not make criticism of translated literature less important, but it does require the critic to be more scrupulous in reaching conclusions and rendering judgments.