Every translator faces the same choice many times on every page. Should I follow the words and structure of the original as closely as my language allows, so that readers can see what the author actually wrote? Or should I say what the author meant in the way a native speaker of my language would say it, so that readers understand without effort?
Both answers have strong arguments behind them, and the debate between them is more than two thousand years old. This article traces it, from ancient Rome to modern translation theory, and explains how Open Word Bible approaches it.
An example
Take 1 Peter 1:13. A word-for-word rendering of the Greek begins, “having girded up the loins of your mind.” The King James Version kept the picture: “gird up the loins of your mind.” The image is of a man tucking his long robe into his belt so that he can work or run. To an ancient reader it was vivid. To most modern readers it is puzzling. The World English Bible gives the sense instead:
Therefore prepare your minds for action, be sober, and set your hope fully on the grace that will be brought to you at the revelation of Jesus Christ— (1 Peter 1:13)
The first approach keeps the image and asks the reader to understand it. The second gives the meaning and loses the image. Neither is simply right.
Cicero, Horace and Jerome
The question was already familiar in ancient Rome, where educated people translated Greek literature into Latin. Around 46 BCE the orator Cicero wrote that, in translating two Greek speeches, he had worked “as an orator,” not as a mere interpreter, keeping the force and style of the words rather than counting them out one by one (On the Best Kind of Orator 14). The poet Horace, in his Art of Poetry, advised writers not to render word for word like a slavish “faithful interpreter.”
Four centuries later, Jerome, the translator of the Latin Vulgate (see Jerome and the Latin Vulgate), quoted both. In his Letter 57, defending a translation he had made, he declared that he rendered “not word for word, but sense for sense” (non verbum e verbo, sed sensum exprimere de sensu). But he made one exception: Holy Scripture, he said, “where even the order of the words is a mystery.” So the most famous early statement of meaning-for-meaning translation also contains the most famous argument for word-for-word translation of the Bible. In practice, scholars note, Jerome’s own Bible translation was often freer than that exception suggests.
Luther and Tyndale
At the Reformation, the question became urgent, because translations were now meant for ordinary people. Martin Luther, in his Open Letter on Translating (1530), said that to learn how to speak German one must not ask the Latin letters but the mother in the house, the children in the street and the common man in the market, and watch how they speak. He defended his addition of “alone” in Romans 3:28 partly on the grounds that natural German required it, although his critics saw the addition as theological (see Luther and the German Bible). Yet Luther also said, in a defence of his translation of the Psalms, that sometimes he had kept the original wording exactly, even against natural German, where much depended on it.
William Tyndale wanted a ploughboy to understand the Bible, and his New Testament was written in plain, strong English (see William Tyndale and the English New Testament). The translators of the King James Version, in their preface “The Translators to the Reader,” explained that they had deliberately not tied themselves “to an uniformity of phrasing, or to an identity of words,” using different English words for the same Greek or Hebrew word where the sense allowed. Many later translators, including the revisers of 1881, thought this a mistake; the debate over consistency of terms continues today.
Schleiermacher’s two paths
In 1813 the German theologian Friedrich Schleiermacher gave a lecture in Berlin on the different methods of translating. He argued that there are really only two. Either the translator leaves the author in peace as much as possible and moves the reader towards the author, or leaves the reader in peace and moves the author towards the reader. Schleiermacher preferred the first: a translation should let readers feel that they are reading something foreign.
His distinction has been revived in modern theory, for example by Lawrence Venuti, who calls the two paths “foreignising” and “domesticating” and argues that domesticating translation hides the strangeness of other cultures.
Eugene Nida and dynamic equivalence
The most influential modern statement of the meaning-based approach came from Eugene Nida, a linguist with the American Bible Society (see Translating for every language today). In Toward a Science of Translating (1964) he distinguished formal equivalence, which focuses on the form and content of the message itself, from dynamic equivalence, which aims at the closest natural equivalent, so that readers today respond to the message in substantially the same way as the original readers did. In The Theory and Practice of Translation (1969), written with Charles Taber, he argued that a translation should be judged by whether ordinary readers understand it correctly.
Nida’s ideas shaped hundreds of translations in languages around the world, and in English the Good News Bible. Because “dynamic” was often misunderstood as permission to paraphrase, he and Jan de Waard later preferred the term functional equivalence (From One Language to Another, 1986).
Purposes and contexts
Later theorists widened the question. Skopos theory, developed by Katharina Reiss and Hans Vermeer in the 1980s, argues that the right method depends on the purpose of the translation: a text for public reading, a study edition and a children’s Bible may rightly differ. Christiane Nord added the principle of loyalty: whatever its purpose, a translation must not misrepresent the author’s intentions.
Ernst-August Gutt, a translation consultant who applied relevance theory to translation, argued that meaning always depends on context, which readers of an ancient text do not share. A translation can either try to give the meaning in terms of the reader’s own context, or try to preserve as much of the original as possible and expect readers to learn the context, perhaps with the help of notes. Gutt called the second “direct translation.” On his account, the choice is not between accuracy and freedom but between two different ways of communicating.
The critics
Meaning-based translation has had strong critics. Leland Ryken, a literary scholar, argued in The Word of God in English (2002) for what he called “essentially literal” translation. He contended that dynamic equivalence interprets for the reader, removes ambiguities that the original leaves open, flattens literary features such as metaphor and repetition, and makes it impossible to tell where the text ends and the translator begins.
Defenders of meaning-based translation, such as Gordon Fee and Mark Strauss, replied that every translation interprets, that literal translations can mislead by using English words in ways no English speaker would understand, and that a reader who misunderstands an accurate-looking sentence has not received the original. Each side has a point. Much depends on the reader and the purpose, as skopos theory says.
One meaning, three levels
Open Word Bible does not try to settle this argument by choosing a side. It gives every sentence at three levels: Literal, which follows the original’s wording and order wherever English allows; Balanced, which reads as ordinary modern English while staying close to the original; and Readable, which uses plain language and explains idioms (see Literal, Balanced, Readable: three levels, one meaning). Here is 1 Peter 1:13 in the current drafts:
Therefore, having girded up the loins of your mind, staying sober, hope wholly on the grace being carried to you at [the] revealing of Jesus Christ.
So brace your minds for action and stay clear-headed. Put your hope entirely on the grace that is coming to you when Jesus Christ is revealed.
So get your minds ready to act, and think clearly. Put all your hope in the grace God will give you when Jesus Christ is revealed.
What holds the levels together is the rule that they never differ in meaning. Decisions about what the Greek or Hebrew means are made once, recorded with their alternatives and reasons, and followed by every level. Only wording changes. That meets one of Ryken’s chief concerns: a reader can always see where interpretation has happened, because each decision is listed in the sentence’s reasoning record (see Showing our work: what a reasoning record is). And it meets Nida’s: a reader who needs plain language can have it without a different meaning being smuggled in.
The approach has limits. Three levels cannot capture every possible purpose, and deciding what counts as meaning and what counts as wording is itself a judgement that reviewers may dispute. The drafts are AI-assisted and none has yet been reviewed. But the project’s hope is that, by showing its reasoning, it lets readers take part in the oldest argument in translation rather than simply receive its results.
Sources
- Cicero. De optimo genere oratorum (On the Best Kind of Orator) 14–23.
- Jerome. Letter 57, to Pammachius, “On the Best Method of Translating.”
- Luther, Martin. Sendbrief vom Dolmetschen (Open Letter on Translating), 1530.
- “The Translators to the Reader.” Preface to the King James Version, 1611.
- Schleiermacher, Friedrich. “Über die verschiedenen Methoden des Übersetzens.” Lecture to the Royal Academy of Sciences, Berlin, 1813.
- Nida, Eugene A. Toward a Science of Translating. Leiden: Brill, 1964.
- Nida, Eugene A., and Charles R. Taber. The Theory and Practice of Translation. Leiden: Brill, 1969.
- Gutt, Ernst-August. Translation and Relevance: Cognition and Context. 2nd ed. Manchester: St Jerome, 2000.
- Ryken, Leland. The Word of God in English: Criteria for Excellence in Bible Translation. Wheaton, IL: Crossway, 2002.
- Bible quotations are from the World English Bible (public domain), except where the article quotes a historic translation or gives its own literal rendering.