I noticed an interesting translation issue as I was reading Jerome Walsh's commentary on I Kings. The longstanding debate between favoring the grammatical form vs. favoring the sense of a text comes up full force in I Kings 11:1-4. Consider the NRSV translation of these verses:
King Solomon loved many foreign women along with the daughter of Pharaoh: Moabite, Ammonite, Edomite, Sidonian, and Hittite women, from the nations concerning which the Lord had said to the Israelites, "You shall not enter into marriage with them, neither shall they with you; for they will surely incline your heart to follow their gods"; Solomon clung to these in love. Among his wives were seven hundred princesses and three hundred concubines; and his wives turned away his heart. For when Solomon was old, his wives turned away his heart after other gods; and his heart was not true to the Lord his God, as was the heart of his father David.
Walsh makes the following comment in a footnote (p.134, n.2):
Hebrew uses the same word (nasim) where English has two different ones, "women" and "wives." The NRSV tries to capture the proper nuance to translate each case. My discussion tries to reflect the way nasim becomes a motif word in the Hebrew text.
Some people favor the sense over the form, most noticeable in translations sometimes called dynamic equivalence (e.g. in Bible translation, the NLT is a good example, and the NIV and TNIV tend in that direction often). One good thing about this kind of translation in cases like this is that you get to capture the nuance of the word in different contexts. The same Hebrew word can mean both "wife" and "woman". In different contexts, it might have the flavor of one of those rather than the other, and here it has each flavor a verse apart. If you translate them both the same way, that's harder to capture. In particular, if you talk about Solomon's women rather than Solomon's wives, in English you get the sense that it's talking about his harem. But then with Solomon you actually are talking about his harem, so maybe it's not that big a difference in his case. Still, one might argue for translating the word as "wives" in all of its occurrences so as to avoid that sense instead of translating it consistently as "women" the way Walsh does. You lose something either way, but you lose something if you translate it differently in different instances also.
I think it's easier to tell from the context what the sense might be, so it's less necessary in these verses to seek to distinguish between the senses the word can have by translating as the NRSV does, as one in one verse and the other in the other verse. What Walsh points out, though, is that you miss something important about this passage if you emphasize sense over form. The repetition of the word conveys something in the Hebrew that you lose in an English translation if it distinguishes between different senses the word can have throughout this passage. There's a literary element of the passage that the NRSV translates away.
This sort of thing often happens in the so-called dynamic translations. Translations that emphasize form, while sometimes missing elements that a sense-for-sense translation will convey, does capture some elements like this that you won't see in a translation like the NLT and often won't see even in the NIV or TNIV. There are those who regularly deride translations like the ESV or NASB as if they have no positive features as translations, seeing them as wooden artifacts of archaic language that barely make sense as English and are too hard for the average English speaker to understand. Whatever element of truth there is in that characterization, there are certainly things that the ESV and NASB preserve that you don't find in the sense-for-sense translations, and it's one reason I always like to have one around.
(The ESV does translate them the same way the NRSV does, as "women" and then "wives", I should note. This is a theoretical point about Bible translation, not an argument for a particular translation as a whole.)