Why the ESV is Errant in Matthew 1:7-10

          “All Scripture is breathed out by God.”  That statement is not only the introductory phrase of Second Timothy 3:16 in the English Standard Version; it is also an affirmation in the introduction of the ESV Reader’s Gospels  (in more traditional wording):  “All Scripture is inspired by God.”  At the ESVBible website, a brief essay teaches that “As the Bible is the inspired word of God, presenting us with God’s words as mediated through human language, it is likewise inerrant and infallible.”

          Evangelical theologians may therefore have good reason to wonder why the ESV New Testament promotes two errors on its first page.  I refer to the ESV’s erroneous claims that Asaph and Amos were among the kings of Judah in the ancestry of Christ.  The answer to this question involves textual variants.  

          The ESV’s preface was intended to give readers the impression that the ESV is a direct descendant of the KJV:  the ESV, the writer claims, “stands in the classic mainstream of English Bible translations,” and continues “the Tyndale-King James legacy,” and so forth.  However, those who read the section of the preface sub-titled Textual Basis and Resources will find a statement that the ESV New Testament is based on the fourth edition of the United Bible Societies’ Greek New Testament and on the 27^th edition of the Nestle-Aland Novum Testamentum Graece – which is another way of saying that the ESV New Testament was translated from a base-text that is very similar to the compilation produced by Westcott and Hort in 1881 – a compilation which thoroughly replaced the primarily Byzantine base-text of the KJV New Testament with primarily Alexandrian readings, resulting in over 5,000 changes.     

          In Matthew 1:7-10, there is a contest between Ασα (Asa) and Ασαφ (Asaph), and between Αμων (Amon) and Αμως (Amos).  The compilers of the UBS and NA-texts, like Hort, rejected the readings that are found in the vast majority of manuscripts (and in diverse early witnesses including Codex Washingtoniensis, Old Latin Codex Vercellensis, the Vulgate, the Sinaitic Syriac, and the Peshitta), and adopted the Alexandrian readings Ασαφ and Αμως, thus conveying errors, inasmuch as Asaph was a songwriter (the author of several psalms) and Amos was a prophet who prophesied in the time of Uzziah.  (Uzziah is mentioned in the genealogy in Matthew 1:8-9).  Neither Asaph nor Amos was an ancestor of Jesus.  

Codex K (Cyprianus) displays the Byzantine reading.
In 1:7, note the interesting proximity of Ασα
to the letters σαφ in the next line.

         It is for that very reason that Ασαφ and Αμως were preferred by the editors of the ESV’s base-text, on the premise that copyists would tend to replace difficult readings with non-problematic ones, instead of the other way around.  The preference for the more difficult reading – a text-critical canon sometimes expressed in Latin as lectio difficilior potior – initially seems to compel the adoption of Ασαφ and Αμως.  However, that impression may be reversed when additional factors are considered.

          The late Bruce Metzger, in his argument for Ασαφ, mentioned a statement from Lagrange (an earlier scholar) to the effect that inasmuch as anyone making this genealogy-list would have to consult the Old Testament, and anyone reading the Old Testament would see the kings’ correct names, “It is necessary, therefore, to suppose that Ασαφ is a very ancient [scribal] error.”  Metzger dismissed that line of reasoning via the supposition that “the evangelist may have derived material for the genealogy, not from the Old Testament directly, but from subsequent genealogical lists, in which the erroneous spelling occurred.”

          However, there is no evidence for the use of such a hypothetical genealogical list in the hands of the evangelist; meanwhile the evidence for Matthew’s familiarity with the Old Testament permeates his Gospel-account.  In addition, considering that Matthew knew the Old Testament and treated it as authoritative, which source is he more likely to have favored when they disagreed:  the Old Testament text, or some “subsequent genealogical list” (assuming that he ever had one)?  

          Metzger attempted to present Ασαφ and Αμως as if the evangelist merely had a strange way of spelling Ασα and Αμων.  Footnotes in the ESV make the same attempt.  However, on balance, the evidence that Metzger cited weakens his position.  In the Septuagint, out of the many occurrences of Asa’s name, he is almost always called Asa; the few intrusions of Ασαφατ and Ασαφ and Ασαβ are simply scribal mistakes.  As Jonathan Borland has pointed out:  “That only these few comparable examples exist out of 90 or so instances of the two names in the LXX demonstrates just what one should expect: while the vast consensus of manuscripts always distinguished the names, less than 10 percent of the time a single scribe (with the exception of 2 Chr 29:13 where 3 manuscripts vary) wrote one name for the other.” 

          Before I offer an explanation of the origin of the Alexandrian reading, it may be appropriate to point out the diverse name-spellings found in the flagship manuscripts of the Alexandrian Text in Matthew 1:1-13: 

1:2 – ﬡ (Sinaiticus) reads Ισακ instead of Ισαακ.

1:3 – B (Vaticanus) reads Ζαρε instead of Ζαρα.

1:4 – ﬡ reads Αμιναδαβ correctly the first time the name is written, but Αμιναδαμ the second time.

1:5 – B, ﬡ, and P1 read Βοες against diverse opposition favoring Βοοζ.  (Nevertheless the UBS-compilers adopted Βοες).

1:5 – B and ﬡ and some Alexandrian allies read Ιωβηδ instead of Ωβηδ.  (33: Ιωβηλ.)

1:6 – ﬡ* reads Σαλομων instead of Σολομωνα.

1:6 – B reads Ουρειου instead of Ουριου.

1:7 – ﬡ reads Αβια, Αβιας instead of Αβια, Αβια.

1:8 – B and ﬡ read Οζειαν instead of Οζιαν. 

1:9 – ﬡ reads Αχας, Αχας instead of Αχαζ, Αχαζ.

1:10-11 – B and ﬡ read Ιωσειαν, Ιωσειας instead of Ιωσιαν, Ιωσιας.

1:12-13 – B reads Σελαθιηλ instead of Σαλαθιηλ, in addition to reading γεννα instead of εγεννησεν three times.

1:13 – ﬡ* reads Αβιουτ instead of Αβιουδ.

          (Except for the readings in 1:5, these readings disagree with both the UBS/NA compilation and with the RP2005 Byzantine Text.  This shows a high level of variation in the spelling of proper names in the Alexandrian text-stream.)  

          Several Old Latin manuscripts agree with the Alexandrian text’s readings for Asaph and Amos.  While, on one hand, this gives the reading some diversity, on the other hand it may indicate that at these points the primary Alexandrian witnesses ﬡ, B, and P1 reflect an early Western intrusion. 

          In 1885, J. Rendel Harris proposed that the reading Ασαφ, Ασαφ originated as the result of a “ghastly line-errors,” that is, Ασαφ was accidentally written when a copyist’s line of sight drifted to the letters σαφ in the nearby word Ιωσαφατ.  He suggested that the same phenomenon can account for the origin of the reading Αμως, Αμως – the copyist’s line of sight straying, in this case, to the letters ωσ in the nearby word Ιωσειαν.  Harris concluded, “It can hardly be accidental that this coincidence of letters is found in the proper names.  And this simple paleographic explanation being given, is not to be shaken by an array of excellent MSS in which the archaic error may be preserved.”  (The same sort of syllable-interchange may account for ﬡ’s reading Σαλομων in verse 6, echoing the Σαλ from Σαλμων’s name in verse 5.)

          I am not persuaded by Harris’ theory; the occurrence of two such mistakes so close together seems unlikely.  However, I am also not persuaded by proponents of the idea that Matthew would risk confusing his readers by listing Asaph and Amos as kings of Judah, knowing that his readers would recognize Asaph as a well-known psalm-writer, and Amos as a well-known prophet.   

         What has happened, I suspect, is that an early Western scribe, unfamiliar with Old Testament chronology, introduced the names of Asaph and Amos as a primitive attempt to pad Jesus’ Messianic résumé, so to speak, by adding prophets among his ancestry.  The tampering of this scribe influenced the Western transmission-line represented by some Old Latin copies.  When these Western readings intersected with the Alexandrian transmission-line, they blended into a crowd of orthographic variations – that is, in some Western Old Latin copies, and in Egypt, the names of Asaph and Amos were assumed to be variant-spellings referring to Asa and Amon, and for that reason, they were not corrected.  Elsewhere, though, these readings were either never encountered, or were almost always rejected as variants which Matthew had not written and which he had been highly motivated not to write. 

          Among the passages in the ESV New Testament which its editors should revisit when preparing the next edition, Matthew 1:7-10 is near the top of the list.  Ask yourselves, ESV editors:  where is the evidence for Metzger’s theory that Matthew used a “subsequent genealogical list” instead of simply consulting the Old Testament text?  And how realistic is the theory that Matthew would take for granted that his readers would identify Asaph and Amos as kings of Judah?  Why wouldn’t Matthew – especially if one affirms that Matthew was writing under the inspiration of God –write the usual names?  The rationales which some commentators, advocating the Alexandrian readings, have attributed to Matthew, may more readily be assigned to scribes.

          Lectio difficilior potior has its limits.  However difficult it may be to picture a scribe introducing the names of Asaph and Amos into the text of Matthew 1:7-10, whether accidentally or deliberately, it is much more difficult to picture Matthew (or any first-century author familiar with the contents of the Old Testament) doing so. 

The English Standard Version (ESV) is Copyright ©2001 by Crossway, a publishing ministry of Good News Publishers.  All rights reserved.