This essay is a discussion of the popular imagination of “literal translations,” rather than more precise discussions of how the term should be used. Even in the latter scenario, however, it is likely that much of the critiques mentioned would be relevant for consideration. It is a rough draft, following from this discussion.
Translators and readers alike have been quietly swooned into a maddening passion for a deceptive lover. He has looked into our eyes with the utmost sincerity, promising us nothing but pure trust and picture-perfect truth. He has courted us with an enticing foreign accent, singing us songs of his far-away home and lifting our hearts to its tune as if to fly us there to see with our own eyes its grasses and groves, its people and places, its country and culture.
He is no other than the “literal translation.” He is the way for our uneducated spirits to touch the text, hot out of the linguists’ smelting fires, as undistorted as possible: not yet hammered into a twisted shape by the likes of a craftsman with their fantastic delusions. Or so we have believed.
Such “literal translations,” as conceived in the common imagination, often amount to nothing more than a linguistic gloss rendered, as minimally as possible, into a semi-meaningful sentence in the target language. In these works, native word usage is typically the first burdensome remnant of language to be torn away from its undeserving throne and hurled off the pages into exotic oblivion. Both in compounds and in stand-alone instances, idiomatic words are replaced for approximant foreign counterparts, and compounds are reduced to an arranged marriage between typically incompatible words.
On likely equal par to be swiftly sliced away is fluid syntax. In its inspired fury, linguistic gloss focuses its attention on the abstracted digits and bits of words, allowing the relationships and structure between them to be hacked away and placed into choppy discord. Sentences no longer flow, while smooth reading through paragraphs and passages dries to dust.
Glossing is the process by which a source text is broken down into its constituent parts, each of which is given a basic sign comprehensible to the target audience. The basic fundamental units of meaning will be divied out, subjectively of course on the basis of the needs and interests of the glossers and their audience. They will be assigned a simple word indicating their basic meaning. Then, the various morphemes and other bits of language will be broken off and indicated with technical terms for linguists to decipher and study the structure of the source language.
Two simple examples, taken from Wikipedia, of interlinear gloss:
(1) goa² iau²-boe⁷ koat⁴-teng⁷ tang¹-si⁵ boeh⁴ tng²-khi³
I not-yet decide when want return.
(2) Gila abur-u-n ferma hamišaluǧ güǧüna amuq’-da-č
now they-OBL-GEN farm forever behind stay-FUT-NEG
The purpose of a gloss is to help break down the components of the source language for students or linguists to understand how it works. Glosses like these can be found in reference grammars or language readers, that is, such works which have the specific aim of studying and teaching the source language.
Now, such sentences do not literally mean, in English, “I not yet decide when want return,” and so on. In most forms of English, such a sentence would, literally speaking, be an ungrammatical rendering which conjures up certain foreign-sounding connotations and intentional ambiguities in the mind of the reader. Rather, the literal translation of this first sentence would presumably be something close to “I have not decided when I want to return yet.”
Glosses are not English, nor any other natural human language. They are technical systems built out of select terms and specialized symbols to accomplish their particular task. Hence, rendering a text into a gloss is not translating it into a target language in the normal sense of the word. Unlike a gloss, the purpose of a translation is to render the source language into the target language.
Let us be clear: the gloss is a wonderful tool, which is perfectly capable of doing what it is intended to do. But it is a brutal, bitter, broken tool for doing what it is not intended to do: translate. The problem arises when glosses dress themselves up in the clothes of translation. It is all too obvious just what they are, but it is our desire for accuracy perhaps that colors our eyes and clogs our ears to see them for something else.
So long as the sentences are relatively straightforward and express themselves in relatively compatible ways with the target language, a gloss-based rendering can come across as clear enough. However, when the sentences grow in complexity, nuance, length, or really any other factor, they will not be able to be weeded away through minimal translation from glossing. To be properly translated, that is transposed into a proper form of target language, they will require significantly more expertise and finesse.
Target languages, like English, have their own syntax, semantics, and so on which are crucial to their functioning as coherent systems. The purpose of a gloss is to better understand a language, and emphatically not to reduce functioning languages into strings of morphemes with simplistic definitions and artificial syntax. Translations have as their basic purpose not the deconstruction of the target language but rather the enriching of it with new comprehensible texts.
Any language which is being glossed is not reducible or equivalent to the gloss provided. Glossing is specific and limited to a particular set of goals. Unlike a gloss, the source language will not appear to the native speakers to have strange syntax and simplistic renderings of out-of-order morphemes. Unless the language has been intentionally altered to create an effect in the participants of the language, which a gloss would not be able to illuminate as it would appear as any other gloss would, then the source language will sound native, fluent, coherent, and meaningful to the audience. It will have rich variations, embellishments, stylistic choices, and many other turns of phrase or fun with language which will become clear only to one familiar with the nuances of the language as it is used on the ground.
A gloss simply cannot capture the nuances of language. Its task is simple: to gloss the morphemes of a text. Nonetheless, the morphemes of a text are not the totality of the text. The implications of the morphemes can hardly be addressed through a gloss. Moreover, the morphemes do not tell us all the possible ways the text could be said and the various shades of meaning and nuance in each of the variations; they do not tell us if they are arranged natively or incorrectly; they do not tell us the intended tone or intonation of the text; they do not provide the context for why a particular phrase is used and what effect it is supposed to have; they do not provide more detailed maps of syntax; and on and on we could list the many, many aspects of language which simple dictionary definitions of lexemes and morpheme analyses do not offer us.
We may distinguish translation from gloss in terms of what they communicate. A translation communicates sense or meaning (in the broadest use of the word). A gloss communicates linguistic abstraction. Each may be done in a variety of styles, with a variety of goals in mind, and a variety of differences in what and how they deliver their final work. Such variety is helpful and important. Still, to capture the sense of a text, one must look to translation. To capture a rendering of linguistic abstraction from that text, one may turn to gloss.
Almost no translator would dare provide a literal morphological gloss and call it a translation. What is unfortunately commonplace instead is that glossers assume the role of translators, placing themselves in the awkward position of constantly compromising their literal glosses for when they need to loathingly indulge in actual moments of translation. They sigh begrudgingly, hoping and griping that their audience would just understand the darn source text already! But at last, they have set themselves the task of providing text in a target language, and provide they do, sacrificing just about every bit of life in that language at the altar of their linguistic idols.
The language presented by glossers can seem to be little more than a fantasy, a dream, of how they wished language ought to be. True language cannot be reduced to a pile of morphological crumbs, dished out to the hungry hoping to taste the savour of nourishing meaning. It cannot be rendered into a string of misused words. These strings sell themselves as authentic garlands to tourists, being made all the while of little more than the scraps locals would never buy themselves. To many readers, they present themselves as representing their source texts, all the while giving much the opposite.
First, these “literal translations” are hardly translations; they are glosses which perform the minimal amount of gymnastics necessary to provide semi-grammatical, semi-meaningful sentences in an amputated form of the target language. They transform a gloss into the closest somewhat-sentence possible, stitching back together the members of the original carefully so as not to add or subtract more than they subjectivelly deem necessary. Square brackets or footnotes may be sprinkled about intuitively, feigning supposedly counterfeit words against what seem to be the cold-hard reality of the original words out in the open.
Moreover, “literal translations” are not literal either. Take a simple example like an idiom. We might encounter an English phrase such as “it’s raining cats and dogs.” A glosser in disguise, eager to provide the most literal and true meaning of the text, might decide to render this idiom in the typical quasi-meaningful form of their target language. But let us not be fooled. The last thing our English speaker intended to say was that a literal ‘it’ was literally raining out literal cats and, literally, literal dogs! To ‘literally’ render those words into their approximate form would, in the target language, be figurative. It would be an ambiguous and foreign-sounding turn of speech far from the basic meaning of the original, far from its native idiom jingle, and far from the connotations and contextual nuances transmitted in English.
The reader of these partially comprehensible glosses is not offered what they might believe or expect from the text, that is, a translation rid of loose and subjective renderings without the corruptions and distortions of artistic translators who alter the text at their whim. While subjectivist and free translations certainly can exist, the solution for clarity and precision should not be semi-fluent glossing.
“It’s raining cats and dogs” does have a precise meaning; it portrays a particular feeling in a native English context; and it will typically occur as part of a larger unit of thought which further clarifies its use and effect on the audience. This is even more the case for smaller units of meaning, such as words and phrases, as well as more complex sentences rather than fixed idioms.
There is no true safety to be found in translating mere gloss. While gloss-renderings may appear to offer a cop-out which protects the translator from over-interpretation, they are themselves one of many interpretive traps. Glossing is itself a foreign imposition on the text, based on foreign concepts and systems of thought, limiting the original in a very intentional and reductive way. It is just as idiosyncratic as any other free translation, in that it does not sets out with transferring the sense of the text in proper form. Instead, the author’s ideas and theories are given favor and transposed onto the text.
A fluent English-speaking reader of, say, Charles Dickens will not pick up his tome and find themselves stumbling through an exotic-seeming soup of clunky word clusters. No, they will find themselves floating down a quaint river with beautiful scenery that is easy on the eyes. The sentences will be rich and packed with detail. The syntax will be meaningful and contextual. The turns of phrase and use of compounds will evoke particular, select associations. And if the author intends the text to sound foreign, he will employ literary devices familiar to his reader to do so, such that the authors intended meaning and effect will come across. If we compare this to a gloss masquerading as a translation, very little of the above will be the case.
If we grant the glossers their misconstrued syntax and isolated morphemes, we might still move on to consider some nuances of poetry or conscious composition. The intentional acoustics and rhythms found in poetry are no less “literal” than the choice of slang words over prestigious ones might be. While we may be tempted to scoff off poetry as notoriously difficult to render “literally,” in the sense that the glossers would have us believe, the same issues prove equally true of prose. Alliteration, assonance, onomatopoeia, rhyme, and other phonological features are no less a part of an original text than morphemes and syntax. And the same can be said for the social dimensions of a text, or the countless other original, inseparable dimensions of language that are literally there, shining off the words on the page, calling out to their audience.
Another case of misleading “literalism” is etymological translation. Etymology is the study of a word’s history. Etymologically, ‘assassin’ comes from the Arabic meaning ‘hashish-user.’ Clearly, the literal meaning of the English word ‘assassin’ is not ‘hashish-user.’ But gloss translations, which reduce language to a small selection of abstractions, often fall prey to the assumption that a word’s etymology is a “literal” meaning, and any other rendering is a more intuitive, free, interpretive rendering. But despite how much one might wish that morphology and etymology were the pure, unscathed, literal meaning of a word, we shall not cede the most basic sense of words to abstract theories: assassin means assassin, not hashish-user.
The same is true for any etymology or etymological rendering. Etymology should be of interest primarily to linguists and philologists, not to translators. Most native speakers of a language will be largely unfamiliar with the vast majority of their words’ etymology unless they receive formal education in the matter, just as they may be ignorant of the history of any other subject. But not knowing the history of words does not render native speakers incapable of using them.
The carpenter may not know the history of the wood he shapes or tools he uses, though he may be a perfectly skilled carpenter. Likewise, a specialist in technological history or evolutionary biology may be utterly incapable in the face of basic carpentry. History is history; use is use. And language, in its most primordial ground, is use.
The use of language is embedded in the world. It is embodied, given to us in all its fullness. At this fundamental level, language as lived is pre-reflective, unconscious. It shapes the kinds of questions we ask and the solutions we propose. Language is embodied in its speakers and participants. But so too are we embodied in language. To interact with it at the level of translation, we must listen closely and allow ourselves to be invited through the door of its surface appearance and into its worldly enactment, whether ancient or modern.
Native speakers are the ideal point of reference for understanding a language as it is used and in the sense it communicates. And yet for the majority of human history and in the majority of cases, native speakers are unfamiliar with the linguistic abstractions derived from their languages. Just as in the case of etymology, they may not be able to draw a syntax tree, form a reliable theory of generative grammar, break down morphophonological nuances, or reconstruct the proto-forms of words. Yet they are the de-facto, gold standard for understanding how to form sentences and how to understand them in their language. That is, the job of a translator.
Language in its natural environment is blind to any constraints or academic systems that researchers impose on it. It is fluid and innovative; it is complex and variegated. It soars through the wind of its speakers, communicating rich nuance without regard for how that might be codified. This is even more the case when, through abstraction, we look to find correspondence between our artificial categories imposed on language.
The word is a simple thing. But what precisely counts as a word is already a point of debate, and different linguists with particular frameworks in mind will use the concept in different ways. Granting ourselves just the concept of ‘words,’ what is thought of as a word in one language may correspond to a sentence or series of words in another. A one-word sentence in English is very different from a one-word sentence in Kalaallisut. We cannot simply map concepts from one language on to another and exchange them as cheap goods in a marketplace.
The same is true of grammatical particles, syntax, declensions and conjugations. How these are categorized in any given language may be a point of contention, and the particulars of them will be unique to a language community. While there certainly is overlap between languages, even still, to find abstract correspondence between these categories is the job of a linguist. The underlying signification or sense of language is primary for translators, not the blocks of unit or formal structure into which languages organize it. The craft of translation is not in mapping theoretical correspondences, but instead in finding meaningful correspondence at the level of use, of life.
We may recall again the experience of a reader with a text in their native language, hoping to translate it to a language they know equally well. Assuming that they understand the meaning and are familiar with the subject matter at hand, they should be perfectly capable of translating the text provided as is. They need not rely on glossing to understand the text. Their understanding will instead be much more holistic and rich in their interpretation of the language. They will generally be perfectly capable of rendering the meaning without naive literalism provided they do not have a misunderstanding of translation or a misleading theory of language.
It is likely the that gloss-translations of ancient texts are largely a product of their didactic role. Many translators learn ancient languages through translation and glossing, generally side-by-side. If one is familiarized with a language through gloss-based translation, this way of relating may remain throughout one’s relationship with the language. Students learn from former gloss-translators to carry on their methods, and readers become acquainted with and convinced of this same theory of translation.
Perhaps the best translators would be those people who set aside linguistics and etymology when all but absolutely necessary, and who strive only to accurately capture the sense of the passage and render it well into their chosen translation style. Of course, a grasp of etymology and other branches of linguistics can be of immense benefit when translating a text, especially an ancient one where native speakers are no longer accessible. But we must not forget that in most cases, such texts did at one time have native speakers, and their audience was mostly such native speakers. Such secondary sources of knowledge must remain as such: away from our primary relationship to language.
True translation is a difficult craft which demands familiarity with all the languages involved. The fruits of its labor are significantly more nutritious and sweet, and therefore, all the more precious. It comes in a variety of flavor, with some renderings attempting to stick close to the original text, and others aspiring to their own literary feats beyond the original’s goals. Nonetheless, a translation is a translation. It is not an inspired original work, nor is it a hardly grammatical gloss. It walks a fine line between the two. The middle it finds is not merely a point between two ends; it is closer to the point of a triangle between the line beneath: in the middle, yet off the scale.
Those blessed readers and translators alike who expect to touch an accurate sense of the original text should no longer fall prey to the whispers of adapted glossing. In believing these to be faithful translations, they lose not only the original but also their own language in the process. As opposed to common conceptions, the so-called “literal translations” produce radically distorted compositions because of their commitments to limited theories of language.
There is an ironic and sad turn of events when the ideology behind such “literal translations” and their convinced adherents is taken too literally. The result is that specialty “translations” are produced in specialized, rarified forms of the target language which are created for the sake of translation. Often, many new word compounds will be coined, along with syntactical and morphological features brought over, perhaps even phonological ones, to essentially pass the source language through a code to produce the same string of morphemes in the guise of what appears to be the target language. The meaning of these passages becomes so obscure and foreign to the average reader, that familiarity with the underlying language and training in the target code-language is required to decipher their meaning.
Given that the resulting target text will be lacking much of the nuance of the original and easily misconstrued, stacks of commentaries and explanations must be provided in order to make sense of the “translation.” These commentaries will, of course, be provided in different times and places, in a more native and fluent form of the target language, from a variety of different perspectives on the source text. They may explain the meaning in such a radically different way from the original than any subjectivist or freer translation of the text would have offered had they attempted to properly translate the text. The result is an obscure and ambiguous scholastic code with an exegetical tradition that tries to explain a text it no longer has native access to.
We should be clear in our critiques and allow for repetition: interlinear glossing and gloss-like grammatical renderings are important, useful, and informative endeavors. They help contribute to the study of a language and its corpus. They are the best at doing precisely what they are designed to do. The same is true for various styles and methods of translation. There is ample room for various translators to adopt different ideologies and approaches to a text, each making a particular contribution and shedding light on the original in important ways. We should not expect there to be a single, one-and-only method of correct translation. There are most certainly many fine and diverse translation ideologies.
Confusion arises, instead, when participants in translation confuse semi-grammatical glossing for not just true translation proper, but also the most accurate and precise form of translation. When they feel that they have touched the essence of language, as if a building were only its scaffolding and nothing more. When they feel that any other form of rendering is not faithful to the text, not accurate, incorrect, interpretive, subjective, and that the only objective and pure way of relating to a text is in their quasi-translation glossing. This is the harmful misconception.
It may very well be the case that readers are those who fall most prey to this error. Translators themselves are often aware of the complexities of the craft and the differences of approach. But it is readers who often take to translating, and translators who are readers of other works. The gloss-literalism ideology is capable of silently tangling its coils into the subconscious of the literate population. And from there, more outspoken or intentional works take on its methods with a false sense of confidence. It is to these issues that we hope to point.
So: let us leave the gloss to be well and at peace, and allow proper translation its welcome place back! Wishing all the best to all the translators out there. May you be happy and well!
.
