Conflicting Translation Procedures (Rom 2:27, 29)

I was asked the other day about the translation of γραμμα in Romans 2:27. “In verse 27, the ESV went with ‘written code’ but it verse 29 they went with ‘letter.’ ‘Written code’ is what the NIV has and is more functional, where ‘letter’ is more formal. Can I ask why the difference there? Thanks so much!”

Good question (and nicely asked). Here is the text of the ESV.

Then he who is physically uncircumcised but keeps the law will condemn you who have the written code (γαμματος) and circumcision but break the law.… But a Jew is one inwardly, and circumcision is a matter of the heart, by the Spirit, not by the letter (γραμματι). His praise is not from man but from God.”

I don’t remember this actual discussion, but I suspect the answer is in the footnote to “written code” in v 27. It says, “Or the letter.” In other words, the ESV did not miss the concordance; we saw that the same Greek word was used in both verses. And one of our policies was to try and use the same English word for the same Greek word in the same context. Some people ridicule this policy, accusing the ESV of not understanding the nature of language, and that one word can change its meaning from context to context. When I started on the ESV, I held this position (but without the ridicule and questioning of a person’s competence). Semantic range and all that stuff.

However, what I soon learned was that there is something gained by following this rule — again, not every time a word is used regardless of context, but certainly when the same word is used in the same context, as we have here. What is gained is that the English reader can see another clue as to the train of the author’s thought. In Greek hermeneutics class we teach our students to watch repeated words; this translation procedure helps the English reader do the same. The conflict with English usage is that English prefers to vary words in the same context for stylistic reasons; using the same word over and over again is boring, which is why the RSV often varied its translation of the same Greek word in the same context, something that we changed in the ESV.

So why did the ESV not use “letter” in both places? The answer is that translation is not that simple. In any one verse there are several translation procedures at work, and sometimes they conflict with each other. Then the decisions become more difficult. What are the translation procedures at play here?

1. Concordance. As much as possible, use the same English word for the same Greek word so the user can follow the author’s train of thought, as long as doing so does not misrepresent the semantic range of the Greek word.

2. One for one. Prefer a single word translation for one Greek word.

3. Less interpretive. While all translations are interpretive, the ESV prefers the less interpretive. “Written code” is more interpretive; “letter” is less.

4. Euphony. The single word “letter” provides a nice poetic balance to the singe word “Spirit” in 2 Cor 3:6. (The NIV/TNIV do the same.)

5. Must make some sense. But wait! There’s more! (Sounds like a Greek infomercial.) Why does the ESV use “code” in Rom 2:27. Because saying “you who have the written letter and circumcision” makes no sense. Now granted, the ESV is content to make its readers work a little to understand the text, just as Paul was content to make his readers work a little to understand the text. But “letter” just sounds weird.

6. Open to misunderstanding. The ESV is especially sensitive to this problem, a problem all formal translations share. If the ESV read, “you who have the letter and circumcision but break the law,” would people unfamiliar with Paul’s theology think of an actual letter?

It is interesting that there are three similar uses of γραμμα in Paul. In Rom 7:6 the ESV translates as “written code,“ and in 2 Cor 3:6 (2x) the ESV uses ”letter.” The NIV/TNIV do the same.

As you can see, it gets complicated. When you put it altogether and shake it up, the ESV felt (in line with its translation procedures) that “letter” was too difficult in v 27 and went with “written code.”

In conclusion, we should be very slow to point our fingers and say that a translation is “wrong,” as so often happens. Translation is not simple. On any one verse there are many principles of translation shouting to be heard, and sometimes their shouting is in conflict with each other. Contained in all this is an appeal for humility.

When someone says that a translation is “wrong” and that the translators “should” have done it another way, in most cases (in my experience) this comes out of a lack of respect to the complexities of translation and a lack of knowledge of all the procedures being followed by a translation committee.