The translator of literature encounters unique challenges not present in the translation of more workaday communication. Day-to-day speech is often plain and direct, so our interpreters, digital and human alike, prioritize speed over nuance. When we enlist Google Assistant to translate “where is the bathroom,” so that we may seek the assistance of a foreign companion, we are more interested in a ‘good-enough’ translation that quickly and efficiently results in directions to the facilities than in one which carries all the subtle music of the original phrase. But as Virginia Woolf argued in her broadcast “On Craftsmanship,” words can conjure dozens of diverse associations that color perception of a sentence. She presents an example: a train attendant announcing “passing Russell Square” may alternately arouse images of rustling skirts or British history— or both at once. But while Woolf’s attendant likely did not intend to shade his words with so many diverse meanings, a skilled writer who is aware of possible connotations and wordplay would.

In intentional works, each phrase and word contributes to an overall tone and gestalt that is the result of millions of these tiny associations. The literary translator is faced with the challenge of preserving these dense networks of association and meaning in a new language.

Artistic concerns aside, the literary translator must also determine how to translate elaborate sentence and syntactic structures that may not exist in the target language. Native speakers of different languages have different expectations and ability with regard to text complexity and readability. Other problems abound: how to deal with puns, jokes, poems, and rhythm— language functions that thrive on the euphony and idiosyncrasy of certain words?

The translator of literature must contend with all of these in addition to the normal challenges of translation, with their only reprieve being the additional time and study they are afforded. Clearly, the number of plates they must spin far outnumber those of the tourist guide or metro station ticket taker. As literary translators must render their new text both artful, accurate, and legible, they draw on many disparate cognitive skills. Meta-knowledge about the cognitive processes involved may enable better translation.

Cognitive science is a young field, with most scientific historians dating its theoretical origins to the 1950s and empirical study to only the 1970s. As such, a summary of its history may be quite brief. In “Cognitive Approaches” Alves and Albir provide an overview of the development and history of brain-based study of translation, from the field’s first tentative theoretical discussions to modern empirical studies that make use of eye-tracking and advanced computer data analysis. In their estimation, the cognitive approach to translation has yielded eight basic developing areas of study, some of them as broad as “stages related to understanding and re-expression” or “the role of retrieval”. The authors conclude that, as a young field, “[cognitive] Translation Studies lacks a tradition of empirical-experimental research,” and thus understanding of the cognitive processes involved in translation is rudimentary.

Given the limited research history and the insufficiency of current tools for empirical study, as well as the inherent difficulties in studying that which is “not amenable to direct observation” we must rely on broader conceptual work. (We will consider why conceptual research is important to the working translator shortly). As translation is a complex and abstract process, there exist a variety of models for conceptualizing it. In “Translation as a Cognitive Activity,” Albir and Alves describe six competing models of translation’s mental processes. Albir and Alves characterize Seleskovitch’s Interpretive Theory of Translation as the most foundational and “ground-breaking” of these. Seleskovitch defines three phases of translation; the first is understanding, followed by deverbalization, and then re-expression. This is the most enduring and simple formulation of cognitive translation; by contrast, Bell’s psycholinguistic model is a snarl of interlinking processes. In “Cognitive Processing Models of Translation in Pedagogy” Moghadas summarizes Bell’s model as dual-phase and three staged, where each original clause undergoes a process of syntactic, semantic, and then pragmatic processing. Each stage contains a number of sub-stages that attempt to describe the way the brain looks the clause’s components up in lexical memory, analyzes it semantically, and breaks down the clause’s context with regards to mode, tenor, and domain (although these are just a selection of sub-processes in Bell’s model). Albir and Alves synthesize a general model from the wider field and establish the baseline assumptions to be tested in further study. Attempts to interrogate these models in a laboratory setting are limited; Alves and Albir, in “Translation as a Cognitive Activity,” note these efforts do not “have a long-standing tradition in the field” and consequently have a “negative impact on the development and validation of research designs”. However, this does not mean that empirical study has been entirely fruitless.

Of particular interest to the discussion of literary translation are studies that quantify, assess, or otherwise compare translators of differing abilities and competence, as we can generally consider literary translators expert level. While there is no consensus definition for what a high skill translator is, in “Modeling Translator’s Competence,” Alves and Gonçalves propose a model that considers how aware an individual translator is of their own skill. This self-understanding of one’s specific translator competence is “directly proportional to the production … generated from two counterpart translation units” and is also proportional to the “maximization of [the] interpretative resemblance” between source language and target language. Crucially, Alves and Gonçalves conclude that translators at an expert level “draw more heavily on meta-cognitive processing,” which means the greater one’s own awareness of how translation works in the brain, the better one can translate. Expert translators are enabled by their “specialized knowledge” of cognitive translation processes to perform at the highest level.

This is the chief application of cognitive study to literary translation in practice. If a translator seeks to increase their skill, they may wish to read some translation studies journals. This conclusion seems unintuitive— how is awareness of the mental model supposed to increase efficacy? Study of Seleskovitch’s or Bell’s models may help the literary translator improve their self-understanding and self-conception of their translation approach. It is a matter of both intent and understanding. Perhaps it is not so different from a guitar player who is aware she struggles with barre chords, and thus takes greater care to think about how she moves her fingers as she plays them.

Translating literature is more difficult than most texts, and the end product is often more difficult to read than untranslated texts in the target language. Because of the high barriers to the production and comprehension of translated literature, any process that enables higher competency translation is facilitating communication. Literary translation’s raison d’etre is to share art across languages, cultures, and social gaps, and meta-cognitive knowledge of its processes helps translators fulfill their duty.