Job probably dates from the seventh to fifth centuries before the common era. But the earliest Hebrew manuscript that survives was written some fifteen hundred years later. Through many centuries of oral and scribal transmission, corruptions are bound to occur even in the simplest text; and Job, because of its strange idiom and the extreme compression of its verse, must have seemed difficult even to the poet’s contemporaries. Difficult, and scandalous. In several places, it is obvious that some scribe has deliberately altered a word, out of a pious desire to suppress Job’s blasphemy. And there are numerous other errors that must be due to inadvertence or misunderstanding.
However these corruptions came into being, they are there, in plain view, and the translator can make only the most tenuous sense out of many verses, unless he emends them. Some emendations may require revision of every word in a line; but most are relatively simple, requiring the change of a letter or two. For example, in chapter 5, verses 6-7 in the traditional (Massoretic) text can be rendered:
For pain does not spring from the dust
or sorrow sprout from the soil:
man is born to sorrow
as surely as sparks fly upward.
Not only is the second verse a weak antithesis; it contradicts Eliphaz’s entire argument that misfortune is not the common lot of man, as Job thinks, and is not part of the natural order, but is rather the direct result of an individual’s wickedness or folly. Now the original written Hebrew consisted entirely of consonants; the vowel-points were added during the seventh century C.E. by rabbinic scholars called the Massoretes, following the then current tradition of pronunciation. A change in vowel-points here (yulad to yolid) yields a verse that fits well in the context:
man gives birth to sorrow
as surely as sparks fly upward.
Some emendations carry with them the authority of the Septuagint, the ancient Greek translation, which dates from the second century B.C.E. and follows a different, possibly older, textual tradition than the Massoretic text. The relation of the Septuagint to the Hebrew Bible is itself a highly complex matter; the translation is often paraphrastic, and there are many errors and lacunae. But sometimes the Greek will be crystal clear where the Hebrew is muddled or meaningless. A simple instance occurs in 23:13:
For he [God] is in one, and who can turn him back?
What he wishes to do, he does.
The first phrase has been interpreted, with varying degrees of strain, as meaning “he is of one mind” or “he is alone in power.” But the Septuagint, instead of b’hd, “in one,” reads as if it is translating bhr, “has chosen”:
For he has decided, and who can turn him back?
What he wishes to do, he does.
This makes perfect sense. And it is easy to account for the corruption: r was miswritten as d, which it closely resembles in Hebrew script; then some later scribe changed bhd, which has no meaning, to b’hd.
Emendation is one way of reaching a true text. Another is comparative philology. (There is no way to recover the true text; all interpretations are to some extent arbitrary; and in the end the only method is one’s own intuition.) Quite a few of the obscure expressions in Job can be elucidated by reference to cognates in other ancient Near Eastern languages. One example occurs in 39:21:
He [the horse] paws in the valley and rejoices;
in strength he goes out to meet the weapons.
The noun ‘mq, which usually means “valley” in Hebrew, means “strength, violence” in Akkadian and Ugaritic. There is an obvious parallel in the second line, and since parallelism is a structural principle of Hebrew poetry, the verse can with assurance be translated:
He paws violently and rejoices;
he runs furiously to meet the weapons.
Nevertheless, there are verses in Job, and even entire pas-sages, which are so obscure, or where the text is so corrupt, that no solution seemed acceptable. I have omitted such verses when the omission did no damage to the continuity of the translation. I have also deleted scribal glosses and verses which seemed redundant or out of place, and have occasionally changed the order of verses as they appear in the Massoretic text.
The Elihu interlude, which has long been recognized as an addition by some later, much inferior poet, and the added Hymn to Wisdom that constitutes chapter 28, with its uncharacteristic stanza-form and its unruffled piety, have been left out as well.
Literal translations of Job may convey its sense more or less accurately, but no literal translation can hope to embody the grandeur and pulsing urgency of its style. In trying to make Job into a living poem, my primary obligations have been to the spirit of the original and to the rhythms of the English language. I have translated closely when possible, freely when necessary; and have not hesitated to improvise, on those few occasions when less drastic methods seemed inadequate.