KJV, ESV, and Robert Alter pursue similar translation strategies. Note how KJV, ESV, and Alter translate the following passages:
The Lord is a man of war:
The Lord is his name. (KJV)
The Lord is a man of war:
The Lord is his name. (ESV, Alter)
וַהֲרֵמֹתָ מֶכֶס לַיהוָה מֵאֵת אַנְשֵׁי הַמִּלְחָמָה הַיֹּצְאִים לַצָּבָא
levy a tribute unto the Lord of
the men of war which went out to battle. (KJV)
[A]nd levy for the Lord a tribute from the men of war who went out to battle. (ESV)
[A]nd you shall raise a levy for the Lord from the men of war who went out to the army. (Alter)
עַד־תֹּם כָּל־הַדּוֹר אַנְשֵׁי הַמִּלְחָמָה מִקֶּרֶב הַמַּחֲנֶה
all the generation of the men of war were wasted out of them. (KJV)
[U]ntil the entire generation, that is, the men of war, had perished from the camp. (ESV)
[U]ntil the whole generation, the men of war, came to an end from the midst of the camp. (Alter)
1 Samuel 17:33
כִּי־נַעַר אַתָּה וְהוּא אִישׁ מִלְחָמָה מִנְּעֻרָיו
thou are but a youth, a he a man of war from his youth. (KJV)
[F]or you are but a youth, and he has been a man of war from his youth. (ESV)
[F]or you are a lad and he is a man of war from his youth. (Alter)
KJV, ESV, and Alter render איש מלחמהand אנשי המלחמה by way of "a man of war," and "the men of war," respectively. The advantage of preserving concordance across the relevant passages is that one is able to "hear" Ex 15:3 against the background of the other passages. The disadvantage is that "man of war" and “men of war” are calques - albeit intelligible ones - not natural English equivalents. The calque is clear and forceful. It is used with some frequency today. But it is not standard terminology.
DE (“dynamic equivalence,” more often referred to as “functional equivalence,” a translation strategy pursued by, e.g., CEV and NLT) and middle-of-the-road translations (e.g., (T)NIV and REB) often translate a recurrent expression like the one in question every which way. CEV has “a warrior,” “the soldiers,” “the men who had been in the army,” and “a soldier.” NLT has “a warrior,” “the army,” “the men old enough to fight in battle,” and “a man of war,” respectively. (T)NIV has "a warrior," "the soldiers," and "the fighting men," and “a warrior,” respectively. REB has “a warrior,” “the warriors,” the fighting men,” and “a fighting man,” respectively. The result: the co-inherence of the texts under consideration is lost in translation.
DE and middle-of-the-road translations, precisely because they invest more heavily in intelligibility and naturalness of language “one passage at a time,” rarely make concordance and co-inherence of language across passages a priority. To be sure, NRSV and NJPSV have “warrior” and “warriors” in all the above passages. In this instance – a rare event – KJV, ESV, and Alter have nothing on NRSV and NJPSV in terms of concordance.
Still, a type “A” male-gendered translation of the above passages per KJV, ESV, and Alter, as opposed to a “gender-neutral” translation like “warrior(s)” or “soldier(s),” has certain advantages. “Man of war” and “men of war” gender יהוה, Goliath, and the other fighting men of the passages under review in terms of what Cynthia Chapman, in her monograph on the gendered language of warfare, calls “the system of signs,” or set of “associated commonplaces,” associated with adult masculine performance in the ancient Near East. Even if it could be shown that איש and אנשי in these idioms particularize only, rather than ascribe gender, “man of war” and “men of war” are defensible translations insofar as they reflect the cultural grammar of the text at the discourse level. They are preferable to “warrior(s)” or “soldiers” on those grounds alone.
To be continued.