Translating Truth: The Case for Essentially Literal Bible Translation is a collection of papers originally presented to the Evangelical Theological Society in 2004. Each of the five authors was part of the Translation Oversight Committee for the English Standard Version (ESV). This post is part 3 of a series, evaluating C. John Collins’ paper, “What the Reader Wants and the Translator Can Give: First John as a Test Case.”
What is a Translation?
C. John Collins opens his paper recognizing a very real problem that characterizes debates about English Bible translation. He observes that both sides tend to talk past each other, basing their arguments on their own presuppositions or definitions of terms. The most notable example is the definition of “translation” itself. Dynamic or functional equivalence translations are somtimes labeled as skillful free interpretations (Grudem), and essentially literal translations are sometimes called transcriptions. Each side appears eager to portray the other as not really doing translation at all.
Rather than engaging right away with specialist definitions, Collins considers what the lay person would consider a translation. For him, this is a good place to start, because “a good philosophy will start from everyday rationality and build on it, and refine it.” His conclusion is that the average lay person would expect neither a woodenly literal rendering nor one where the renderings do not directly correlate with the words, but one that, in the words of R. C. Van Leeuwen, “conveys as much as possible of what was said, and how it was said, in as near word-for-word form as the target language allows, though inevitably with some difference and imperfectly.” This kind of translation allows someone to “listen in” on a foreign communication without the translator getting in the way.
In Collins’ view, this definition does not have a place within the popular method of placing translation philosophies on a continuum between literal and dynamic. He draws special attention to the fact that essentially literal translations, unlike “woodenly literal” ones, favor recognized linguistic operations, resulting in a translation that sounds like good literary English. In other words, such a translation recognizes that a shift is indeed taking place from one language to another, but interpretation is limited to these recognized linguistic operations.
Because his translation philosophy is apparently so distinct from others, Collins suggests abandoning the single-line continuum for something a little more complicated: a triangle, with “literal” on the left, “dynamic” on the right, and “essentially literal” at the top. I find this suggestion unconvincing and unhelpful. First, he hasn’t actually demonstrated that essentially literal translations are categorically different and don’t take a place on the line-continuum model. All he has done here is highlight the difference between his model and the two extremes. Second, this new diagram doesn’t really alter the line-continuum at all. It simply bends it upward so that essentially literal translations are in a prominent place at the top center. There is an irony to this, as earlier in the paper he draws attention to the NIV’s portrayal of itself as the ideal balance between the two extremes on the continuum. I think he is correct in saying that translation is much more complicated than the line-continuum model makes it appear, but his solution only makes the nature of those complications more confusing and reflects his own bias.
From here, Collins moves on to discuss another complexity of translation: the difference between the text and the message (or in terms used in my previous post, the difference between saying and meaning). Collins handles this issue with far more competence than Ryken. He points out that a text is “a means by which the speaker (or author) operates on [a] shared picture of the world to produce some effect (the message) in the audience.” The primary difference between essentially literal and dynamic equivalence translations is that the former stops at the text level, allowing the reader to simply “listen in” on a foreign communication. The latter, on the other hand, seeks to convey the meaning, producing the same effect that the communication had on the original audience. I think this is a fair assessment in theory, so long as we recognize that essentially literal translations regularly go beyond the text, and dynamic equivalence translations often stick close to it. That said, this generalization highlights a fundamental difference in the definition of “translation” by these two philosophies. Consider Van Leeuwen’s above definition of translation and its emphasis on what was said, and compare it with a definition of translation from relevance theory: “A receptor-language utterance is a direct translation of a source-language utterance if, and only if, it presumes to interpretively resemble the original completely” (emphasis mine) (Gutt, Relevance Theory: A Guide to Successful Communication in Translation).
Translations for Particular Uses
Having established the wide gap between these translation philosophies, Collins pulls back from his stated preference and proceeds to ask the important question, “What kind of translation might suit…various contexts for the English reader?” I’m glad that he is at least entertaining the possibility that different kinds of translations might be best suited for different contexts. Abandoning Nida and Taber’s division of contexts by the social status of the audience, he proposes three basic uses for a Bible translation:
1) Church. He posits that a church Bible should be intelligible, ecumenical, orally rhythmic, preachable, poetic and dignified.
2) Family reading. He sees no reason a church version can’t be used in the family so long as it is intelligible. Parents should be willing to explain things to their children (and the church should equip them to do so), and children will live up to what is expected of them.
3) Outreach. He admits that a simpler, more readable Bible may be better here, but it needs to be clear this is only for introductory purposes.
Case Study on 1 John
With these three uses in mind, Collins proceeds to examine how different translation philosophies perform on the text of 1 John. He focuses on the following features of the Greek text:
1) Repetition of Greek words. One notable example is the appearance of menō 24 times within 18 verses. Essentially literal translations do a good job transparently showing this repetition with the word “abide.” The NLT on the other hand uses no fewer than ten different renderings for the same word. Collins cites contextual consistency, naturalness, and ready intelligibility as motivations for the NLT’s variety.
This observation reveals a legitimate weakness of the NLT. The NLT often does not allow the reader to see the repetition of words. However, it reads far more naturally and understandably. Collins acknowledges this “trade-off between literal precision and readability,” and this is why he holds to essentially literal rather than woodenly literal translation. What he does not explain is why he thinks that word repetition is always more important to preserve than contextual consistency, naturalness, etc. Repetiton might be ideal to preserve in a Bible used for preaching, but in my assessment it’s not going to be the most important factor to consider in a Bible for other uses.
2) Puzzling Ambiguities. One example of Greek ambiguity that Collins points out is the Greek genitive construction translated “love of God” in the ESV. Does this mean “love for God” (objective genitive) or “God’s love” (subjective genitive)? He argues that the ESV maintains ambiguity by using this ‘of’ construction, saying, “The essentially literal approach will be to pass the responsibility on to the reader to decide.” I would suggest that in some cases, the ESV is successful in retaining ambiguity in the text, but this is not one of them. I think it is fallacious to say that the English ‘of’ construction is equivalent to the Greek genitive and retains its full exegetical potential. The phrase “love of God” could be interpreted as “love for God,” but I suspect that most readers would naturally assume a subjective reading. Often when essentially literal proponents claim they are preserving ambiguity, they actually are not achieving their aim to the extent they think they are. Unless people are schooled in the biblical languages and have learned to flag English ‘of’ constructions as ambiguous, I suspect that most English readers would find many of them unnatural more than ambiguous. They will simply assume the interpretation that involves the least cognitive effort. Having said this, I do agree with Collins that essentially literal translations more often do a better job retaining ambiguity than their dynamic equivalence counterparts.
3) Old Testament Evocations. Collins argues that essentially literal translations do a far better job retaining OT evocations in the text. My assessment here is similar to what I presented above regarding puzzling ambiguities. I would agree that essentially literal translations tend to retain OT evocations better, but that they do not often do it as well as they claim. For example, Collins points out the repeated use of the Greek word tēreō, which the ESV translates ‘keep.’ He argues that this rendering conveys more than simple obedience to God’s commands, but an attitude of carefully attending to, or even treasuring them. (The NIV and NLT often prefer the word ‘obey’ among other options.) Looking at the word ‘keep’ in an English dictionary, I don’t see any of the nuances that Collins is pointing out relating to being careful or treasuring. He seems to be defining ‘keep’ in a way that most English speakers do not.
To summarize, Collins argues that how we define translation is of great importance. His own starting point is the lay understanding of translation. He asks, “What does the reader want, and what can the translator provide? An opportunity to listen in on the original foreign language communication, without prejudging what to do with that communication.” Such a translation, he argues, is best suited for all types of uses. He does concede that meaning-based translations might be suitable for new believers as an introduction, and he also admits that essentially literal translations may place a heavier burden on the reader to learn about the world of the text, but suggests that people are (or should be) up for the challenge.
I see two closely related problems with Collins’ conclusions:
1) His definition of translation is far too narrow and one-sided. Much of his argument highly favors “text” over “meaning” (using his words). In praising the virtues of essentially literal translations (many of which really are virtues), he focuses on its transparency to the original Greek, word-for-word correlation, etc. His selection of features of the Greek text in 1 John are carefully chosen to highlight the advantages of essentially literal translations, while neglecting other elements that are vital to successful communication of the text. He acknowledges illocutionary force, implicatures, etc., but doesn’t seem to hold these in high regard as part of the communicative event, relegating them to appropriation or application. A more healthy, complete view of translation would recognize that aspects of the text like word repetition and OT evocations are equal to, and even overlap with illocutionary force and implicatures as part of the whole communicative process. All of these elements are important and intertwined, yet no translation can capture them all. While acknowledging the trade-off between “literal precision and readability,” Collins elevates one over the other, which is unwarranted and presents a very lopsided and incomplete picture of communication and translation.
2) He presupposes that one translation or type of translation is better than many. Because of his lopsided approach to translation and communication, Collins downplays the benefits of using multiple translations. He assumes that “what the reader wants” is the same for every person. There are readers who want a text that speaks to their heads and allows them to wrestle with the text. But there are also readers who want a Bible that speaks to their hearts. And then there are those who want both, because the original text spoke to both head and heart.
Collins attempts to shoehorn essentially literal translations into every kind of use, even though they might not be ideal for that use. Where he does concede that another type of translation might be better, he argues that this should be seen as a temporary solution until the individual is mature enough to graduate to a better Bible translation. Here I will share briefly from my personal experience. I have been a Christian for nearly 30 years. I have been through Bible college, seminary, Greek study, translation work, etc., and I still greatly benefit from translations like the NLT. They are not something I have outgrown. They complement the other translations in my arsenal because they reveal aspects of the original communication that I would not have previously seen and that do not come across in essentially literal translations. I agree with Collins that the church needs to equip people to study the Bible critically and learn about its historical and cultural backdrop. But instead of seeing meaning-based translations as a useful tool to help people, he views them as the opposite, a danger that prevents people from understanding.
Who gets to decide what a translation really is? I don’t think there will be a consensus any time soon, but I think we all agree that a translation is one that does “justice to the original act of communication” (Collins). I only wish that Collins and others would recognize that no translation fully does justice in every respect, and elevating some aspects of communication over others by insisting on only one type of translation hinders the church from understanding God’s full communicative intentions.