Does the Old Testament have variant readings? 8 examples

The standard Hebrew text is called the Masoretic Text, after the Masoretes, the scholars who between the seventh and tenth centuries added vowel signs, accents and notes to the consonantal text. Their notes include a feature called ketiv and qere: “what is written” and “what is read”. In each such place the consonants of the text are left as they were, and a note in the margin tells the reader to pronounce a different word.^[1]

There are somewhere between eight hundred and fifteen hundred of these notes, depending on how they are counted and which manuscript is used. Many are matters of spelling or decency, where a coarse word in the text is replaced in reading by a polite one. Some change the sense. In Isaiah 9:3 (9:2 in English Bibles) the written text says: “You have multiplied the nation; you have not increased the joy.” The reading in the margin, which differs by one letter of the same sound, says: “You have multiplied the nation; you have increased its joy.” Modern translations follow the margin.^[2]

A copy of scripture whose own margins tell the reader to read something other than what is written cannot be described as having no variant readings. The Masoretes preserved the text they received and recorded, beside it, the readings they judged to be correct.

The rabbinic tradition goes further. Early rabbinic works and the Masoretic notes list passages called tiqqune soferim, “corrections of the scribes”, usually given as eighteen. In each, the tradition says that the scribes changed a word that seemed disrespectful to God.^[3]

The best-known example is Genesis 18:22. The text now reads: “Abraham عليه السلام remained standing before the Lord.” The tradition says that the original read: “The Lord remained standing before Abraham”, which the scribes thought unfitting, since standing before someone implies waiting on him. Another is 1 Samuel 3:13, where the sons of Eli are said to have “brought a curse on themselves”; the tradition records that the original read that they cursed God, and the Septuagint indeed reads “cursing God”.^[4] A third is Job 7:20, where Job asks why God has made him a burden “to myself”, which the list says was originally “to you”.

Medieval and modern Jewish scholars disagree about whether these were real alterations or only places where the rabbis taught that the text spoke with restraint. The Midrash on Genesis calls Genesis 18:22 a correction of the scribes, and some modern scholars accept a number of the cases as genuine.^[5] Whatever their status, a tradition that lists eighteen places where its own scribes are said to have changed the wording of scripture out of reverence has already conceded that the text was touched.

Rabbinic literature preserves two further admissions about the state of the text in antiquity.

The first is a story told in the Jerusalem Talmud and the tractate Soferim. Three Torah scrolls were found in the court of the Temple, and they differed from one another in three places. In each place, the sages followed the reading of two scrolls and set aside the reading of the third.^[6] The story is told to show how carefully the text was guarded, and it does show care. It also shows that the scrolls kept in the Temple itself did not agree, and that the standard text was settled by comparing witnesses and choosing among them.

The second admission concerns spelling. Hebrew words can be written with or without certain letters that mark vowels, and the scribes counted letters to guard against error. The Babylonian Talmud records that the scribes of old counted every letter of the Torah and found that the vav of the word gahon in Leviticus 11:42 was the middle letter. The rabbis then ask whether that letter belongs to the first half or the second, and reply that they cannot say, since “we are not expert in full and defective spellings”.^[7] In the Masoretic Text printed today, the vav of gahon is not the middle letter of the Torah by count. Either the spelling of the text has changed since the count was made, or the count was wrong. The Talmud itself admits that the rabbis of its day did not know the exact spelling of their own scripture.

Neither admission touches the meaning of the text in most places. Both answer the claim that the Old Testament has no variant readings, from the mouths of the people who transmitted it.

The Samaritans, a community that traces its descent from the northern tribes of Israel, preserve their own Hebrew text of the five books of Moses عليه السلام. It differs from the Masoretic Text in about six thousand places. Most are matters of spelling and grammar, and about nineteen hundred of the differences agree with the Septuagint against the Masoretic Text.^[8]

A few differences concern the central dispute between Jews and Samaritans: where God is to be worshipped. In Deuteronomy 27:4 the Masoretic Text commands Israel to set up an altar on Mount Ebal; the Samaritan text says Mount Gerizim, the Samaritans’ holy mountain. Many textual critics now think that Gerizim is the older reading, since it is also found in an Old Latin manuscript and in a small Hebrew fragment of Deuteronomy, and that “Ebal” was introduced in Judea in opposition to the Samaritans.^[9] The Samaritan text also adds to the Ten Commandments a command to build an altar on Gerizim, and throughout Deuteronomy it reads “the place that the Lord has chosen”, in the past tense, where the Masoretic Text has “will choose”, leaving room for Jerusalem.

So each community possessed a text of the Torah that supported its own sanctuary, and each charged the other with having altered it. That dispute is older than Christianity.

The discovery of the Dead Sea Scrolls from 1947 onward gave scholars Hebrew manuscripts a thousand years older than the oldest complete Masoretic copies. Their evidence needs to be stated in both directions.

The Great Isaiah Scroll, 1QIsa-a, copied around 125 BCE, is close to the Masoretic Text of Isaiah. Most of its differences are in spelling, and its agreement with the medieval text is strong evidence that the scribes who transmitted Isaiah did their work with great care.^[10] Defenders of the Bible are right to cite it.

Other scrolls show a different picture. A manuscript of Samuel from Cave 4, 4QSam-a, contains a paragraph at the beginning of 1 Samuel 11 that is missing from the Masoretic Text. It tells how Nahash, king of the Ammonites, had been oppressing the tribes of Gad and Reuben, gouging out the right eye of every man, and how seven thousand men escaped to Jabesh-Gilead. Without the paragraph, the siege of Jabesh that follows comes without explanation. The historian Josephus, in the first century, knew the same account.^[11] The New Revised Standard Version added the paragraph to its text in 1989. A passage that had dropped out of the standard Hebrew Bible, probably through a copyist’s eye skipping a line, was restored from a manuscript found in a cave.

In Deuteronomy 32:8, the Song of Moses says that when the Most High divided the nations, he fixed their boundaries “according to the number of the sons of Israel”. That is the Masoretic reading. A manuscript of Deuteronomy from Qumran, 4QDeut-j, reads “according to the number of the sons of God”, and the Septuagint reads “angels of God”.^[12]

Most textual critics hold that “sons of God” is the original. The verse pictures the nations allotted to heavenly beings, with Israel kept as the Lord’s own portion in the next verse, an idea also found in Daniel 10 and in Psalm 82. The change to “sons of Israel” removed a phrase that could be read as admitting other divine beings. Several modern translations, including the English Standard Version and the New Revised Standard Version, now print “sons of God” in their text.^[13]

This case matters because the change, if the critics are right, was made deliberately, to bring a verse into line with a stricter reading of monotheism. A scribe who altered scripture for reverence’s sake, as the tradition of the eighteen corrections also says scribes did, still altered it.

Two examples from the Psalms show the kind of variant that matters to readers.

Psalm 145 is an alphabetical acrostic: each verse begins with the next letter of the Hebrew alphabet. In the Masoretic Text, the verse for the letter nun is missing, so that the poem jumps from mem to samekh. The Psalms Scroll from Cave 11, 11QPs-a, has the missing line: “God is faithful in his words and gracious in all his deeds.” The Septuagint has it too. Modern translations insert it as verse 13b.^[14]

Psalm 22:16 (22:17 in Hebrew) is the most argued case. Christian Bibles, following the Septuagint, read: “They have pierced my hands and my feet”, and read the psalm as a prophecy of the crucifixion. The Masoretic Text reads: “Like a lion, my hands and my feet.” The difference lies in one letter, a yod or a waw at the end of a short word. A fragment of the Psalms from Nahal Hever reads with the waw, which some scholars take to support a verb, “they dug” or “they bored”, though its meaning is disputed.^[15] Jews accused Christians of changing the text, and Christians accused Jews. Both readings have ancient support.

The claim that the Old Testament has no variant readings usually rests on the Masoretic Text as it is printed today. That text has a history of its own.

The consonantal text that the Masoretes received, often called the proto-Masoretic text, was already dominant among the Jews of Judea in the first century, and some of the scrolls show it in an early form.^[16] That much is common ground. The vowels, accents and notes that give the text its fixed reading were added between the seventh and tenth centuries. The oldest complete manuscript of the whole Hebrew Bible, the Leningrad Codex, is dated 1008 or 1009. The Aleppo Codex, written about 930, was the more authoritative copy, and about a third of it, including most of the Torah, was lost after riots in Aleppo in 1947.^[17]

So the oldest complete copy of the standard Hebrew Bible is about three and a half centuries later than the Quran. The Dead Sea Scrolls show that the consonantal text behind it is much older, and they also show that other forms of the text circulated alongside it until the first century. A careful account of the Old Testament’s transmission recognises both facts. The claim that it has no variant readings recognises neither.