Titles of Psalms (4)

Septuagint (LXX) and Qumran superscriptions

There are several instances where the LXX deviates significantly from the Masoretic Text in the titles and these should be noted. Five Psalms are attributed to David in the LXX but not in the MT (27, 71, 97, 143, 144), perhaps going some way to repairing the deficiency of the twenty four “orphan psalms”.[1] Several Psalms in English translations have the title “Of David” translating the Hebrew prefix (the letter ל lamed) as “of” or “by”. The attribution לדוד (le-David) is much discussed in the literature and is beyond the scope of this post, except to note Nahum Sarna’s useful observation:

If le-David indeed originally indicated authorship, then it is of interest that the form is unique to the psalms’ literature (cf. Hab. 3:1) for the ascription of no other biblical book to a historic personality ever involves the use of the lamed formula (cf. Song, Proverbs). Yet the Psalter is internally consistent in its employment of the same construction with other names such as the Korahites (Ps. 42, et al.), Asaph (Ps. 50, et al.), Solomon (Ps. 72), Heman (Ps. 88), Ethan (Ps. 89), and Moses (Ps 90).[2]

He asserts that “in Psalm 72 lamed must mean ‘about’ or ‘dedicated to’, and in Psalm 102 le-‘ani can only mean, ‘for [recitation by] the afflicted man’.”[3] If commentators expect to see consistency by the redactors of the Psalter in their use of lamed (not that this should be necessary) then לדוד could just as readily mean about David, or of a Davidic style or genre, as denoting authorship. The question is certainly not settled, and Childs asserts that, “whatever the expression לדוד may once have meant, the claim of authorship now seems most probable. This point is confirmed by the final clause in those titles which specify a particular historical incident in David’s life as providing the occasion for composition.”[4]

Adrian Curtis argues that the superscriptions provide evidence of the beginnings of a process of ongoing interpretation of the psalms, and that the presence of more such titles in the psalms from Qumran and in the Septuagint show that this process continued after the formation of the Psalter as a single collection.[5] Childs argues from the Hebrew version of Psalm 151 from Qumran (11QPsa) and the further expansion of titles in the Syriac Apocryphal Psalms, in the Targum and in the Peshitta, that this process continued for some time.[6] The difference in historical titles in the various sources (MT, Qumran, the Syriac and LXX) “suggests that titles were not fixed and that there was some fluidity”.[7]

Gerstenberger argues that most superscriptions betray later theological and liturgical interests, without heeding the original intentions of the psalm. For him, technical musical terms such as למנצח “to the choirmaster” or what he regards as obscure references to tunes (understanding על as being an indicator of a tune) would have been of interest only to the ritual expert or the leader of community worship. The frequent indications of authorship (David, Korah, Asaph, Solomon, and Moses), and the linkage with incidents in the life of David, made the psalms authoritative and edifying so that the people might expect that their prayers and songs in both private and communal worship would have the same powerful and beneficent effects as of old.[8] Norman Whybray finds confirmation in the historical headings that the editors intended the readers or worshippers to find encouragement and models for their own behavior in the life of David.[9] There is a hint here that these “historical” psalms may have been intended for personal use by the pious rather than for cultic purposes.

References in the titles to historical events provide some of the most convincing evidence that titles may not have been part of the original work but were added by later editors in a style similar to rabbinic midrash.[10] In the MT all the historical titles are in Books I and II (with one exception in Book V). The LXX has a further fourteen historical titles, mostly in Book V.

The evidence from Qumran

Psalms scroll from Qumran. Tehillim 11QPs

Thirty nine Psalms manuscripts have been found among the Dead Sea scrolls. The largest extant, and best preserved, Psalms scroll found at Qumran is 11QPsa . It contains forty nine (or fifty) compositions, or parts thereof, including thirty nine psalms found in Books IV and V of the Masoretic Psalter, and ten (or eleven) additional compositions including four which were previously unknown. It “diverges radically from the Masoretic Psalter, both in arrangement and by the inclusion of additional compositions.”[11] Differences in the order of psalms are also evident in seven manuscripts from cave 4 and a second Psalms scroll from cave 11.[12] Flint summarises the evidence and the opinions of several scholars and concludes that Psalms 1 to 89 (Books I to III) show a high degree of stabilisation during the Qumran period with no major deviation in content from the MT, and with only minor deviations with respect to the ordering of the psalms.[13] However, he finds abundant evidence of major deviations from the MT in Psalms 90 to 150 (Books IV and V), both in content and ordering. This supports the proposal that the compilers of 11QPsa may have regarded this collection as a “work in progress” and Flint concludes that the Book of Psalms was probably finalised in two stages: the first part (Psalms 1-89) was stabilised before the beginning of the Qumran period (which he puts at about 150 BCE); the second part (Psalms 90-150) remained fluid into the first century CE.[14]

This is important to note for an investigation of the superscripts and postscripts, as Thirtle’s theory relies on the ordering of the Masoretic Psalter. Interestingly, almost all the psalms containing the למנצח to the chief musician rubric in the superscript (or postscript, according to Thirtle), and therefore important for Thirtle’s thesis, are in Psalms 1 to 89 (fifty two psalms, with only three in Book V[15]). Thirtle argued that the “key” to the interpretation of the musical and liturgical notes in the superscripts was “lost early”. The concentration of these terms in Psalms 1 to 89 while almost completely absent in the remainder of the Psalter, together with the evidence from Qumran, supports his thesis and suggests that the understanding of these terms was lost before the compilation or stabilisation of Books IV and V.

To be continued …


[1] A Talmudic term for psalms without a superscription (Av. Zar. 24b)

[2] Sarna, “Psalms, Book of,” 669.

[3] Ibid.

[4] Childs, “Psalm Titles and Midrashic Exegesis,” 138.

[5] Adrian H.W. Curtis, “”A Psalm of David, when …”: Reflections on Some Psalm Titles in the Hebrew Bible,” in Interested Readers: Essays on the Hebrew Bible in Honor of David J.A. Clines (eds. Aitken, et al.; Atlanta: Society of Biblical Literature, 2013), 59.

[6] Childs, “Psalm Titles and Midrashic Exegesis,” 143.

[7] Curtis, “Reflections on Some Psalm Titles,” 55.

[8] Erhard S. Gerstenberger, Psalms, Part 1, With an Introduction to Cultic Poetry (ed. Rolf P. Knierim; Grand Rapids, Mich: Eerdmans, 1988), 30.

[9] Roger Norman Whybray, Reading the Psalms as a book (Sheffield: Sheffield Academic Press, 1996), 21.

[10] See Elieser Slomovic, “Toward an Understanding of the Formation of Historical Titles in the Book of Psalms,” Zeitschrift für die Alttestamentliche Wissenschaft 91, no. 3 (1979); Childs, “Psalm Titles and Midrashic Exegesis.”

[11] Peter W. Flint, The Dead Sea Psalms Scrolls and the Book of Psalms (Leiden: Brill, 1997), 39-40.

[12] Ibid., 49.

[13] Ibid., 141. The manuscripts for Psalms 1-89 are much more fragmentary than for Psalms 90-150, so caution must be exercised in drawing conclusions about these sections of the Psalter.

[14] Ibid., 146.

[15] In the MT appearing as superscripts to Psalms 109, 139 and 140.

Which biblical manuscripts are ‘right’: Qumran, the Septuagint, or the Masoretic Text?

“It is written” – Quotations from the Old Testament in the New Testament (7)

And again, when he brings the firstborn into the world, he says,

“Let all God’s angels worship him.” (ESV)

Commentaries usually propose two possible sources for this quotation in the New Testament from the Hebrew Bible:  Psalm 97:7 and Deuteronomy 32:43. Psalm 97 looks to be the closest match for this phrase, especially once we realise that the Septuagint occasionally translates the Hebrew word elohim with the Greek word angelos (the Hebrew MT of Psalm 97:7 reads השתחוו־לו כל־אלהים worship him all you gods [elohim] whereas the Septuagint reads προσκυνήσατε αὐτῷ πάντες οἱ ἄγγελοι αὐτοῦ worship him all his angels), and it is generally thought that the writer of the NT book of Hebrews usually quotes from the Septuagint (but more about this shortly).

But why Deuteronomy 32:43? If we are reading the King James Version (or one of many others) there doesn’t appear to be any connection. This is how the KJV translates the verse:

Rejoice, O ye nations, with his people: for he will avenge the blood of his servants, and will render vengeance to his adversaries, and will be merciful unto his land, and to his people.

The KJV follows the Hebrew Masoretic Text. Initially it looks like it has no connection to the quotation in Hebrews. However, several modern translations (such as the ESV below, with footnotes) include the additional words which I have underlined:

“Rejoice with him, O heavens;[a]
bow down to him, all gods,[b]
for he avenges the blood of his children[c]
and takes vengeance on his adversaries.
He repays those who hate him[d]
and cleanses[e] his people’s land.”[f]


  1. Dead Sea Scroll, Septuagint; Masoretic Text Rejoice his people, O nations
  2. Masoretic Text lacks bow down to him, all gods
  3. Dead Sea Scroll, Septuagint; Masoretic Text servants
  4. Dead Sea Scroll, Septuagint; Masoretic Text lacks He repays those who hate him
  5. Or atones for
  6. Septuagint, Vulgate; Hebrew his land his people

It is evident from the footnotes in the ESV that there are several differences between the Masoretic Text and other ancient translations such as the Septuagint, Dead Sea Scrolls and the Latin Vulgate. Just to complicate things further, Romans 15:10 may also be quoting this verse in Deuteronomy:

And again it is said, “Rejoice, O Gentiles, with his people.”

If Paul is quoting Deuteronomy 32:43 in Romans then he is following the Masoretic Text with Gentiles or nations,  where  the Septuagint has “rejoice O heavens“. There could be a clue here that the writer of Romans is not the same person as the writer of Hebrews, but having said that, one Qumran version (think ‘Dead Sea Scrolls’) of this text reads “Praise, heavens, his people” (1QDtb) while another reads “Praise, nations …” (4QDtq) so we have two different Qumran-Hebrew texts of this verse and the Masoretic Text represents one Hebrew text while the Septuagint corresponds to another. Romans follows one version, while Hebrews appears to follow the other. The ancient Aramaic version known as Targum Onkelos has an Aramaic equivalent to the Masoretic Text with “Praise, Gentiles, his people”. The Samaritan Pentateuch has the same reading as the Masoretic Text.

Interestingly, in an article by George Howard published as early as 1968 he argued that Hebrews may very well have been following a Hebrew text which was different to the Masoretic text, rather than following the Greek Septuagint, and that Hebrews 1:6 is closer to Qumran Deuteronomy than to the Septuagint. He found that some quotes are actually closer to the Aramaic versions (Targum Onkelos and the Peshitta) than to either the Hebrew or Greek.

“It has been popular in the past to begin a commentary or an introduction to the Epistle by stating that the writer always uses the Septuagint version of the OT (sometimes in the form of Codex Vaticanus, but more often in the form of Codex Alexandrinus) and never shows acquaintance with the Hebrew). Since the discovery of the Qumran Literature and the impetus given by it to the study of the pre-Masoretic text, it is now probable that the text used by the author of Hebrews is, on occasion, closer to a Hebrew recension more ancient than the Masoretic Text.”[1]

But what about Psalm 97:7? If the writer of Hebrews is quoting from this psalm then we can forget the difficulties with Deuteronomy. There are some problems here too, although possibly not as much as with a Deuteronomy source. As I mentioned in my opening paragraph, the main difference between the two texts is that the Hebrew Masoretic Text of Psalm 97:7 reads השתחוו־לו כל־אלהים worship him all you gods [elohim] whereas the Septuagint reads προσκυνήσατε αὐτῷ πάντες οἱ ἄγγελοι αὐτοῦ worship him all his angels). To many scholars this isn’t much of a difficulty because we know that the Septuagint translators sometimes used ἄγγελοι angels in place of אלהים gods. However, Hebrews has “angels of God” rather than simply angels and this suggests the writer was tranlating from Hebrew בני אלהים sons of God rather than simply “gods” [elohim]. Interestingly, 4QDtq from Qumran has “sons of God” in Deuteronomy 32:43, so this may steer us back to a Deuteronomy source and away from Psalms.

It looks like a bit of a mess! At least two versions of one biblical text, with the New Testament writers quoting from both versions. How can they both be ‘right’? The problem is actually a modern one. Timothy Law, in an interview with  Peter Enns, has concluded rather well: ‘We know now that there were many other variant forms of the Hebrew scriptures circulating before the time of Jesus … the existence of multiple forms of scripture (Greek and Hebrew) in antiquity, both before, during, and after the time of Christ, did not bother early Christians. The search for an “original text” on which to ground one’s faith is a distinctively modern worry’ (his emphasis). It seems to me that the New Testament writers reflected current and earlier scribal practices where it was not necessary to copy or translate the exact form of words, but rather to faithfully transmit the ideas and the essential message.

[1] George Howard, “Hebrews and the Old Testament Quotations”, Novum Testamentum, Vol. 10, Fasc. 2/3 (Apr. – Jul., 1968), pp. 208-216

Does Matthew quote from the Hebrew Bible or the Septuagint?

“It is written” – Quotations from the Old Testament in the New Testament (5)


Francois-Joseph Navez, “The Massacre of the Innocents,” 1824

Jeremiah 31:15 is quoted in Matthew 2:17-18 using the introductory formula “Then was fulfilled what was spoken by the prophet Jeremiah …” The background to the quotation is the “massacre of the innocents” – Herod’s murder of children born around the time of Jesus’ birth in an attempt to eliminate a claimant to the title of “King of Israel”. Matthew presumably quotes Jeremiah 31:15 because the context of Jeremiah 31 is the grief experienced by Jewish mothers who watched their sons go off into battle or exile. [1]

Then was fulfilled what was spoken by the prophet Jeremiah: “A voice was heard in Ramah weeping and loud lamentation, Rachel weeping for her children; she refused to be comforted, because they are no more.”

If we compare the Hebrew and Greek texts of Jeremiah 31 with Matthew’s quotation we see that Matthew is closer to the Hebrew than to the Septuagint in a few ways and follows the order of words in the Hebrew more closely than the Septuagint. For example, Matthew’s ὀδυρμὸς πολὺς “much grieving” (or “loud lamentation” in the ESV) is possibly a smoother rendition of the Hebrew construct בְּכִי תַמְרוּרִים “bitter weeping” than the Septuagint θρήνου καὶ κλαυθμοῦ καὶ ὀδυρμοῦ “of lamentation, and of weeping, and wailing”. The Septuagint omits Jeremiah’s first use of the phrase “for her children”, which Matthew retains, although he omits the second use of the term which is present in the Septuagint (it is not unusual for the Septuagint to omit repetitive phrases, or to reorder them, but the fact that Matthew’s order is different to the Septuagint suggests that he was using a similar transaction technique but not copying directly from the Septuagint). I would argue that Matthew was making his own independent translation from the Hebrew of Jeremiah rather than quoting from a Greek (‘Septuagint’) translation. Richard Longnecker has pointed out that in the Gospel of Matthew the evangelist’s own quotations of the Old Testament usually follow the Hebrew reading, whereas the citations by Jesus “are strongly Septuagintal”.[2] This raises the interesting question of why Matthew would make his own translation from the Hebrew at times while using the Septuagint Greek translation when quoting the words of Jesus, especially since it is hardly likely that Jesus himself taught in Greek.

The most likely explanation in my view is that Matthew was using several sources when writing his gospel. We can be confident that one of his sources was the Gospel of Mark as Matthew quotes verbatim almost all of Mark’s gospel (approximately 90% of Mark is in Matthew). Luke also used Mark extensively. There is also a considerable number of sayings of Jesus which are in Matthew and Luke, but not in Mark. This has led many scholars to accept the theory that Matthew and Luke used at least two sources to write their gospels: Mark, and another gospel, or, more likely, a collection of sayings. This hypothetical second source is usually called ‘Q’ (from the German Quelle, meaning “source”) and many scholars believe that  Q was written in Greek. Jesus’ quotations from the Old Testament in Q appear to be from the Septuagint. If this theory is correct then it would explain why Matthew’s quotations of Jesus’ sayings follow the Septuagint (he was simply copying directly from his Q source) while his own quotations from the Hebrew Bible were his own translation. This argument presupposes that Matthew wrote his gospel in Greek, which is the view of many New Testament scholars, although there are some who believe that Matthew wrote originally in Hebrew or Aramaic and his gospel was later translated into Greek. Either way, there is strong evidence that Matthew used at least two sources for his quotations from the Old Testament: when Jesus was quoting the Bible the quotations came to Matthew via a Greek source which drew on the Septuagint, and when Matthew was quoting Scripture directly he drew on his own knowledge of the Hebrew Bible. Our text from Jeremiah 31 is in the second category and is quoted from the Hebrew Bible rather than the Septuagint.


LXX (Jer 38:15)


כֹּה אָמַר יְהוָה

קֹול בְּרָמָה נִשְׁמָע נְהִי בְּכִי תַמְרוּרִים רָחֵל מְבַכָּה עַל־בָּנֶיהָ מֵאֲנָה לְהִנָּחֵם עַל־בָּנֶיהָ כִּי אֵינֶֽנּוּ׃

οὕτως εἶπεν κύριοςφωνὴ ἐν Ραμα ἠκούσθη θρήνου καὶ κλαυθμοῦ καὶ ὀδυρμοῦ Ραχηλ ἀποκλαιομένη οὐκ ἤθελεν παύσασθαι ἐπὶ τοῖς υἱοῖς αὐτῆς ὅτι οὐκ εἰσίν Τότε ἐπληρώθη τὸ ῥηθὲν ὑπὸ Ἰερεμίου τοῦ προφήτου, λέγοντος,Φωνὴ ἐν ῥαμᾶ ἠκούσθη, θρῆνος καὶ κλαυθμὸς καὶ ὀδυρμὸς πολὺς, Ῥαχὴλ κλαίουσα τὰ τέκνα αὑτῆς καὶ οὐκ ἤθελεν παρακληθῆναι, ὅτι οὐκ εἰσίν

[1] Ramah was 10 kilometres north of Jerusalem on the road captives would have travelled as they were taken into exile. Rachel was said to have been buried in the vicinity of Ramah which was equidistant with Bethlehem from Jerusalem.

[2] Richard N. Longnecker, Biblical Exegesis in the Apostolic Period, 2nd. ed. (Grand Rapids: Eerdmans, 1999), p. 48

Misquoting the Old Testament (in the New)

“It is written” – Quotations from the Old Testament in the New Testament (4)

Jeremiah 18:2-3 cited in Matthew 27:9-10

ESV (OT Sources)

ESV (Matthew)

Jeremiah 18:2-3“Arise, and go down to the potter’s house, and there I will let you hear my words.” So I went down to the potter’s house, and there he was working at his wheel. Matthew 27:9-10Then was fulfilled what had been spoken by the prophet Jeremiah, saying, “And they took the thirty pieces of silver, the price of him on whom a price had been set by some of the sons of Israel, and they gave them for the potter’s field, as the Lord directed me.”
Zechariah 11:12-13Then I said to them, “If it seems good to you, give me my wages; but if not, keep them.” And they weighed out as my wages thirty pieces of silver. Then the Lord said to me, “Throw it to the potter” – the lordly price at which I was priced by them. So I took the thirty pieces of silver and threw them into the house of the Lord, to the potter.
Jeremiah lamenting the destruction of Jerusalem. Rembrandt (1630)

Jeremiah lamenting the destruction of Jerusalem. Rembrandt (1630)

Matthew 27:9-10 is the most puzzling citation of Jeremiah in the New Testament. In fact, it is possibly one of the most puzzling citations of any Old Testament text. Matthew introduces this ‘quotation’ with the formulaic “Then was fulfilled what had been spoken by the prophet Jeremiah, saying …” The most remarkable thing about this is that nowhere in any of our manuscripts of Jeremiah do the quoted words appear. The wording which is most similar is Jeremiah 18:2-3 which refers to Jeremiah being directed to go to the potter’s house. Both texts refer to a potter but there the similarity ends. In fact, Matthew’s quotation parallels Zechariah 11:12-13 more closely than any text in Jeremiah. Zechariah refers to thirty pieces of silver as well as a potter, but not to a field. Elsewhere in Jeremiah (32:6-9) the prophet bought his cousin’s field in Anathoth, but there the price is seventeen shekels. Matthew’s ‘quotation’ appears to be a composite of Jeremiah 18:2-3 and Zechariah 11:12-13 with a possible allusion to Jeremiah 32:6-9 and it is difficult to see how it is a ‘fulfilment’ of any specific prophecy. Craig Blomberg has argued that “Rabbis would sometimes create a composite quotation of more than one Scripture but refer to only one of their sources by name, often the more obscure one (though sometimes also the more important one) to ensure that others would pick up the reference”.[1]

Matthew’s quotation of Zechariah preserves a clause found in the Masoretic Text but missing from the Septuagint: “the lordly price at which I was priced by them”. His concluding clause (“as the Lord directed me”) is not in either source text, but could be alluding to Jeremiah 13:15 “So I went and hid it by the Euphrates, as the Lord commanded me”. Neither source text refers to a potter’s field, although this is a key item in the fulfilment of the prophecy to which he is referring. Later in the story about the potter in Jeremiah (19:2) the prophet is instructed to “buy a potter’s earthenware flask, and take some of the elders of the people and some of the elders of the priests, and go out to the Valley of the Son of Hinnom at the entry of the Potsherd Gate”. We may be tempted to see a connection between the Valley of the Son of Hinnom (known as Gehenna in the Gospels) and the traitor Judas Iscariot but Matthew makes no such connection, nor does Luke when writing about Iscariot’s fate in Acts 1:18-20.

What is Matthew doing here with a composite quotation which he attributes to Jeremiah and how can he say that Iscariot’s actions ‘fulfil’ Scripture when there is no such prophecy? Joseph Fitzmyer has noted that “the use of well-known introductory formulae to cite a passage which is not found in the Old Testament (or at least which is not found in any of the known texts or versions)” is a phenomenon found both in the NT and in the Qumran literature” (think “Dead Sea Scrolls”).[2] He put this in the category of “modernized texts” rather than as a literal fulfilment of prophecy.[3] This could also be a case of what he later describes as an “accommodated text”, that is, one which is “wrested from its original context or modified somehow to suit the new situation”[4].

Archer and Chirichigno put this text in the category of quotations which give the impression that unwarranted liberties were taken with the Old Testament text in the light of its context. It would probably be even better, in my view, to categorise the Matthew quotation as a composite allusion rather than a quotation.

[1] Craig Blomberg, [“Matthew” in G.K.Beale and D.A.Carson (eds.) Commentary on the New Testament Use of the Old Testament, (Grand Rapids: Baker Academic, 2007), p. 95]. Unfortunately Blomberg doesn’t provide any sources for or examples of this Rabbinic practice. Archer and Chirichigno also claim that in combining elements from both Jeremiah and Zechariah Matthew is “simply conforming to contemporary literary custom when he cites the name of the more famous of the two” [Archer, Gleason L. and G. C. Chirichigno Old Testament Quotations in the New Testament: A Complete Survey (Chicago: Moody, 1983) p. 163]. but they don’t provide references for their claim either.

[2] Joseph A. Fitzmyer, “The Use of Explicit Old Testament Quotations in Qumran Literature and in the New Testament”, New Testament Studies 1961;7(04):297-333, p. 304. Another example of this in the Gospels is Matthew 2:23 “he shall be called a Nazarene” while ‘As for that which it said, “Your own hand shall not avenge you”‘ (CD ix 8-9) is an example of a Qumran text quoting an unknown source.

[3] Ibid, p.315

[4] Ibid, p.316. Fitzmyer finds twelve examples of accommodation in Qumran texts.

Does the New Testament always quote from the Septuagint?

“It is written” – Quotations from the Old Testament in the New Testament (3)

I wrote earlier that ‘there is a popular misconception that the earliest Christians used the Greek translation of the Hebrew Bible as their Scriptures, and that all the quotations from the ‘Old Testament’ in the New Testament are from this Greek translation, commonly known as “The Septuagint”.’ Before moving on to look at some New Testament quotes that are likely to be direct translations from a Hebrew manuscript I’d like to comment further about the use of the “Septuagint”. I put “Septuagint” in quotation marks because this term is somewhat of a misnomer because there are in fact several quite different Greek translations of the Hebrew Bible which are all identified as “Septuagint”. It would be more accurate to label these texts as “Septuagints” (plural), as many scholars do, rather than identifying any one text as “the Septuagint”. This multiplicity of Greek translations may account for why the New Testament quotations differ quite markedly from the popular Septuagint texts (such as the translation by Sir Lancelot Brenton: The Septuagint version of the Old Testament, according to the Vatican text: translated into English; with the principal various readings of the Alexandrine copy, and a table of comparative chronology, London: Samuel Bagster and Sons, 1844.)

But the differences between the New Testament quotations and the Septuagints could be explained on other grounds as well. The NT writer may have been making his own translation of a Hebrew text (or an Aramaic translation – a targum – for that matter), quoting or paraphrasing from memory, or making a deliberate change for his own theological reasons. I’d like to explore these possibilities with a few examples.

Mark 7:6-7 and Matthew 15:8-9 are parallel accounts which include a quotation from Isaiah 29:13.

 This people honours me with their lips,
but their heart is far from me;
   in vain do they worship me,
teaching as doctrines the commandments of men.

In Beale and Carson’s “Commentary on the New Testament Use of the Old Testament” on Mark 7:6-7  the author (R.E. Watts) notes that Mark’s quotation “generally follows the tradition in the LXX” but is actually closer to the Masoretic Text than the Septuagint (p. 163). This suggests that Mark either used a different Greek manuscript which was closer to the MT, or he was translating directly from the MT. To illustrate this, if we compare the NT quotation with the LXX we see that the NT has the same words for the latter part of the verse but in a different order and omitting καὶ, demonstrating that Mark and Matthew were not dependent on the LXX.

διδάσκοντες ἐντάλματα ἀνθρώπων καὶ διδασκαλίας (LXX)

διδάσκοντες διδασκαλίας ἐντάλματα ἀνθρώπων (NT)

A careful analysis of the NT quotations of the OT reveals that practically every quotation has  at least minor variants from the Septuagints (or major ones) and is never verbatim. That is significant. Either the NT writers were using different Greek manuscripts to the extant versions of the Septuagints or something else was happening. If the Greek Jewish Scriptures were regarded so highly by the NT writers why do they appear to be so careless in quoting it (if they were indeed quoting it) so as to have so many variants? There isn’t a single quote in the entire New Testament which quotes verbatim from any Septuagint manuscripts that we have. I think the current scholarly consensus is that for at least the first two centuries of Christianity the church used a variety of Greek translations as well as Hebrew manuscripts. Some New Testament quotations of the Old Testament appear to be translations directly from a Hebrew text, while others are paraphrases, possibly from memory.

From this one example I think we could conclude that the NT writers were either using a different Greek text to our Septuagints, they were making their own translation from the Hebrew, or they were using a Septuagint but changing it or improving it as they went, but more examples will follow.

Testimonia – help from Qumran

“It is written” – Quotations from the Old Testament in the New Testament (2)

(Probably) Valentin de Boulogne (ca 1594-1632), Saint Paul Writing His Epistles (c. 1618-20)

(Probably) Valentin de Boulogne (ca 1594-1632), Saint Paul Writing His Epistles (c. 1618-20)

When you come, bring the cloak that I left with Carpus at Troas, also the books, and above all the parchments.

2 Timothy 4:13

In the period in which the New Testament was written the Jewish Scriptures were written on several large scrolls usually made of parchment, rather than in a single ‘book’ as we know it. These texts were not put into book form, or codices (singular codex), until considerably later. Codices/books were used in this period, but these were mainly made of papyrus and used for taking notes. Parchment was more expensive than papyrus and therefore reserved for more important texts. Notes might be written on papyrus, but valuable texts which were intended to last longer would be written on parchment.


Biblical scrolls were large and it would not have been possible to carry around the whole ‘Bible’ as we know it. There is evidence from the dead sea scrolls and later Christian literature that several biblical quotations were sometimes grouped together thematically and copied onto a single parchment. These are known as ‘testimonia’ or ‘florilegia’ and the dead sea scrolls includes two Jewish examples including one (4Q175 or 4QFlor, the Florilegium) which is a series of five quotations from the Hebrew Bible about a messianic or prophetic figure, with a pesher or commentary on the last quotation. There are good reasons to believe that some of the writers of the New Testament used similar testimonia as the source of their quotations rather than copying directly from a biblical scroll. For example, some scholars suspect that Romans 3:10-18, which appears to be a lengthy quotation from a single source (in fact, the lengthiest quotation in Paul’s letters), was actually making use of a testimonia. This ‘quotation’ is actually a chain of quotations from several biblical texts (a catena): Psalm 14:1-3, 5:9, 140:3, 10:7, Isaiah 59:7-8 and Psalm 36:1 (references are to the English versions as chapter and verse numbers differ in the LXX and MT). There are two possibilities for the source of Paul’s ‘quotation’: (1) this chain of texts was Paul’s own composition, or (2) Paul was quoting from a Testimonia which had been produced earlier by another writer. It is possible that among the “books and parchments” which Paul asked Timothy to bring with him were such testimonia which Paul had collected or arranged himself, or notes which he had taken at those times when he had access to actual biblical scrolls. Either way, this text in Romans 3 illustrates how that New Testament writers felt no sense of obligation to quote Scripture precisely, and it was even possible to ‘rework’ Scripture by bringing together several texts thematically to create something new. Remarkably, Paul could introduce his chain of texts with the words “it is written” even though that particular arrangement of biblical texts was not actually written anywhere in Scripture.