2019-06-29

On What Grounds Would Anyone Argue That Luke’s Lord’s Prayer Post-Dates Matthew’s?

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

by Neil Godfrey

Statue of Jesus praying, from Pixabay

The following question arose in a Facebook forum a couple of weeks ago:

In comparing Matthew and Luke, we find that Matthew has a wider array of moral sayings (essentially a superset of the material in Luke). Also, Matthew has a more advanced rendering of the Lord’s Prayer, the Beattitudes, the Great Sermon and the Great Commission. It has a wider array of kingdom of God sayings, and a more evolved and expansive treatment of eschatalogical issues. From just about every perspective Matthew looks more ideologically evolved than Luke. On what grounds would anyone argue that Luke post-dates Matthew?

So why do many biblical scholars (most, I believe) say that Luke post-dates Matthew? Take the Lord’s Prayer. It certainly does appear to be “more advanced”, so why would Luke write a “cruder” form of it he was writing after the Matthean version was surely known?

From my earlier post “Jesus Did Not Compose the Lord’s Prayer”:

Matthew 6:9-13 Luke 11:2-4
9 “‘Our Father in heaven,
hallowed be your name,
2 “‘Father,
hallowed be your name,
10 your kingdom come, your kingdom come.
your will be done,
on earth as it is in heaven.
11 Give us today our daily bread. 3 Give us each day our daily bread.
12 And forgive us our debts,
as we also have forgiven our debtors.
4 Forgive us our sins,
for we also forgive everyone who sins against us.
13 And lead us not into temptation, And lead us not into temptation.’”
but deliver us from the evil one.’

I won’t repeat points from Michael Goulder’s article. Here I’ll set out how three other scholars subsequent to Goulder have made a case for Luke’s Lord’s Prayer being a revision of Matthew’s.

Luke’s Different View of Eschatology and the Church

Franklin earlier gave reasons for viewing Luke’s apparently “more primitive/less spiritual” beatitudes being a response to Matthew’s “more elegant and spiritual” list:

We have seen that even the beatitudes make good sense as vehicles of Lukan theology adapted from Matthew as their source and that they fit into a sermon which is itself an adequate expression of the Lukan purpose at this point. Again, the Lukan form of the Lord’s Prayer expresses Luke’s own beliefs and fits comfortably into its context of eschatologically motivated prayer (11.2-4). (Franklin, 350)

I posted my own take (probably inspired by Franklin or others with a similar view) on Luke’s beatitudes in The poor and Q — literary vs historical paradigms (2007).

Eric Franklin in a study comparing the Gospels of Matthew and Luke discerned the following thematic difference between them:

  • Matthew wrote of and for the Church, the assembly governed by rules and ordinances under Peter,  and that Church was a form of the Kingdom of God already here on earth even though at the same time it was waiting for the time when the Kingdom would come with the return of Jesus to extend it world-wide as foretold by the prophets. For Matthew, the Kingdom of God was already here in the church, and that meant the church was being judged now according to its adherence to the rule of Jesus. The final coming of the Judge would bring judgement on how those in “the kingdom” now treated one another.
    .
  • Luke did not think of the church in that way. For Luke (of course I am using shorthand when I speak of Luke and Matthew as the authors since we don’t know who those authors were, and other times I use the names to refer to the gospels themselves) the kingdom was not here on earth now in any form, not even partly, as in the church. No, for Luke the church consisted of people who were called upon to wait patiently and endure trials until the kingdom arrived with the coming of Jesus. What those Christians had until then was the Spirit of God or the Holy Spirit from Jesus and that spirit gave them power and strength to endure and hold fast, but it did not make the church a small advance part of the kingdom of God here and now. That was entirely future.

Again, all this means that Luke sees eschatology as being less realized in the present than does Matthew and he therefore accepts the parousia as having a positive role. It retains the aspect of hope in a way that Matthew’s emphasis upon its judgmental role does not. Luke is more ambivalent and thus more realistic about the realities of discipleship in the present. It is ‘through many tribulations’ that we enter the kingdom of God (Acts 14.22). His Jesus does not therefore indwell the church as he does in Matthew and the church is less directly related to the kingdom. (Franklin, p. 312)

See how that difference is reflected in the two prayers.

In Matthew, we have a prayer for the coming of the kingdom but that plea is coupled directly with an admonition on the church to live within that kingdom here and now. Luke, on the other hand, has no focus on intra-church relationships as the kingdom in operation now but urges his readers to entirely focus on the coming of Christ.

In the passages in Luke following directly on from the Lord’s Prayer we see this theme directly and strongly emphasized.

In Matthew, the Lord’s Prayer is part of general instruction to the disciples which inculcates a relationship to the Law and a life in the community that are both centred in ethics arising out of a right approach to God (5.1-10; 6.1-18). Its petitions to God are to be related to, and the effectiveness defined by, the petitioner’s relationship to other people (6.14-15). Luke’s setting on the other hand is not one of general instruction in a discipleship measured largely in ethical terms, but is rather one about the nature of the kingdom and of the need of perseverance before its coming. In contrast to Matthew, the element of conditionality is not provided by dealings with others, but by trust in God. It is persistence in prayer (11.5-13) which looks for the presence of the Holy Spirit which is then seen as a gift to the community to enable perseverance in the light of disappointments and persistence in the response to the word of God (11.28). The presence of the Spirit enables disciples to discern the reality of the kingdom of God out of which they live and to continue steadfast in the continuing struggle with the hostile powers (11.24-27; Acts 4.23-31). Times of peirasmos bring a falling away from that constancy (8.13) resulting in a falsification of the Spirit’s significance and an exclusion from the community (Acts 5.4-5). Blasphemy against the Holy Spirit is the refusal to respond to the Spirit’s empowering of witness to Jesus ( 12.8- 12).

The Lukan Lord’s Prayer is therefore a prayer of the community living out of the presence of the kingdom but also in the face of earthly trials and disappointments. It becomes a much more specific prayer than its Matthaean counterpart and its significance is wholly eschatological. The kingdom is not realized on earth in such a way that its presence can be seen as an anticipation of its final state, and so his version does not relate the petition for the coming of the kingdom to the extension of the doing of God’s will on earth. Luke therefore emphasizes the reality of daily discipleship. The situation is such that the need of a daily taking up of the cross (9.23) can be met only by a petition for a daily living out of the power of the eschatological kingdom. Peirasmos in Luke is more awful than it is in Matthew. He does not mellow its starkness with the first Gospel’s ‘but deliver us from evil’ (Mt. 6.13). Matthew (like Mark) in the Gethsemane episode sees the disciples’ sleeping as their falling into temptation (Mt. 26.41). Luke gives a greater significance to peirasmos. Jesus’ first statement, ‘Pray that you may not enter into temptation’ is not occasioned by their falling asleep (22.40) and it is repeated after the sleep. The sleep becomes only a pointer to something much greater and, in the light of his saying of Jesus at the supper about Satan’s request to ‘sift’ the disciples (22.31-34), suggests a falling into Satan’s clutches and a falling away from Jesus. It is therefore that apostasy which expresses the unforgivable sin and which is seen as an anticipation of the final release of evil which will embrace so many before the manifestation of final salvation (8.13).

Life within the Christian community does not for Luke actually embody the life of the kingdom. The community is indeed the eschatological community as it lives its life out of the power of the kingdom and Luke’s picture of it in Acts is at pains to make this clear (Acts 2.43-47). But it has not ‘arrived’ in the way that Matthew, for all his emphasis upon the judgment of the parousia, understands it. Luke understands the church as living out of the kingdom rather than as anticipating it in the Matthaean manner. It is not a community boasting in the direct presence of its Lord in its midst (Mt. 28.20) but is conscious that just as its Lord was constrained on earth (12.50), so its goal is beyond the present with which it cannot rest secure. It is by way of many trials that the kingdom is to be entered (Acts 14.22) for the present remains in some way a contrast to, rather than an anticipation of, the kingdom. (Franklin, 272f)

Franklin thus appeals to thematic or theological differences between the two gospels to justify seeing Luke’s Lord’s Prayer as a revision of Matthew’s.

Luke’s Hates Repeating Himself

Ken Olson brings in a stylistic argument to suggest that Luke’s Lord’s Prayer deliberately abridged Matthew’s version.

In short, Luke hated repetition. Whenever he saw an opportunity to shorten a narrative or saying he found in his source he did so. Some examples:

Luke and Matthew often shorten Mark (i.e., they contain shorter readings than Mark in parallel material in the triple tradition), including the words of Jesus. For example, Luke shortens Mark’s Parable of the Sower a good deal for his gospel, cutting Mark’s nine verses down to five (Mk 4.1-9 // Lk. 8.4-8), and reducing Jesus’ speech from 105 words to 72. Similarly, he cuts Mark’s Gethsemane scene down from eleven verses to five, reducing Mark’s 193 words to 89, and has Jesus finding the disciples asleep only once rather than Mark’s three times (Mk 14.32- 42 // Lk. 22.40-46).

This abbreviation may be at least partly due to a general tendency on Luke’s part to avoid repetition. Many Lucan commentators have noted Luke’s omission of doublets from his sources. This has come to be known as his ‘fear of doublets’. In his analysis of Luke’s treatment of his sources, ‘avoidance of repetition’ is one of the 17 features of Lucan style examined by Henry J. Cadbury. In particular, Cadbury contends: ‘Luke’s Greek instinct would lead him to avoid distinctly Semitic parallelism’. Further, he notes six cases in which Luke abbreviates Mark’s questions, especially where they are in the form of double questions or a question immediately followed by an answer. One example of such Lucan abbreviation is found in the introduction to the parable of the sower in Mk 4.13 // Lk. 8.11, where Mark writes, ‘Do you not understand the parable? Then how will you understand all the parables?’ and Luke has, ‘Now the parable is this’.

Cadbury also adduces eight cases from ‘Q’ (double-tradition) contexts where Matthew has paired parallel or antithetical clauses and Luke does not. One example is the saying from Mt. 13.16 // Lk. 10.23, where Matthew has, ‘But blessed are your eyes, for they see, and your ears, for they hear’, and Luke has: ‘Blessed are the eyes that see what you see’. An example of a similar phenomenon, not listed by Cadbury as it occurs in a triple tradition passage, is in Luke’s version of the cry at the triumphal entry. Mark’s gospel has two parallel blessings, while Luke has one (Mk 11.9-10 // Lk. 19.38). Cadbury also lists examples of Luke abbreviating or avoiding repetition by omission or the substitution of pronouns or synonyms for words previously used. In short, Luke is in many cases less repetitive than the other two evangelists and has a tendency to eliminate repetition and unnecessary verbiage from his sources, though it is not an invariable tendency. (Olson, 106f)

Similarly, in relating the story of Jairus, “ruler of the synagogue”, Mark repeats the identifier three times, three times readers are told Jairus was a ruler of the synagogue. Luke cut out the repetition and told readers just once.

But isn’t the general tendency for the church to have expanded liturgical material rather than shorten it over time? The Didache, for example, is thought to have expanded the Lord’s Prayer in Matthew by adding the line “for yours is the power and the glory for ever” at the end. Some manuscripts of Luke do in fact show a Lucan text that expands the Lord’s Prayer. Can we really expect Luke to have shortened the Lord’s Prayer?

Olson refers to James Charlesworth’s observation that the common notion that expansion of such texts over time is not supported by consistent evidence.

He briefly discusses the Astronomical Books of Enoch (1 Enoch 72–82), 2 Enoch, the Testaments of the Twelve Patriarchs, and (possibly) Jeremiah, in which the LXX is about one eighth shorter than the MT, as examples of texts that became shorter in the course of transmission. Charlesworth anticipates the objection that these examples do not come from liturgical texts and so produces an additional example from the apocryphal Psalms. The additional example is Psalm 151, which has been found in Hebrew at Qumran and is also known in Syriac, and presumably comes from a Greek intermediary. In translation, the Hebrew version of Psalm 151 reads:

Psalm 151.1-2 Hebrew (11QPsa 151) Psalm 151.1-2 Syriac
I was the smallest of my brothers,
and the youngest among the sons of my father;
I was the smallest of my brothers,
And a child of my father’s house.
and he made me shepherd over his flocks.
And the ruler over his kids.
I was tending my father’s flocks.
My hands made a flute,
And my fingers a lyre;
My hands made instruments,
And my fingers fashioned a lyre.
And I shall render glory to the Lord,
I thought within myself.

The Syriac has clearly truncated the Hebrew psalm.

Similarly with another Jewish historian (Luke is arguably a historian), Josephus, abbreviates and in other ways changes prayers he finds in his source 1 Maccabees:

‘A reader who compares 1 Maccabees with the paraphrase in Antiquities 12–13 discovers interesting discrepancies. While the events as reported by Josephus normally correspond to those in 1 Maccabees, the speeches are embellished by him. The prayers, on the other hand, are reduced to short passages almost negligible in the flow of the narrative.’ (Enermalm-Ogawa quoted by Olson, p. 110)

To bring the point back to Luke’s Lord’s Prayer compared with Matthew’s:

1. Luke just 22 verses earlier wrote that the Father to whom he prayed was “in heaven” — Luke 10:21. Why repeat his location? It is surely clear that it is the heavenly father being addressed.

Compare the other repetitions in Matthew’s prayer: “thy will be done on earth as it is in heaven”. That can be understood (departing from Franklin’s interpretation here) as simply a Semitic parallelism or elaboration of “thy kingdom come”. “Deliver us from evil” …. is not that simply another way of expressing the same thought above, “lead us not into temptation (or trial)”?

So on stylistic grounds we can understand Luke being frustrated with Matthew’s wordiness and repetitiveness and therefore producing a crisper version of the Lord’s Prayer.

Our View of Aesthetics Denies Luke His Own Creativity

Thomas Mosbo expands somewhat on Olson’s explanation by stressing the need to allow for Luke’s creativity as an author and accepting that aesthetic judgements are not reliable objective foundations for determining priority.

One of the responses to this line of reasoning by advocates of the Q theory is that in at least one case, the Lord’s Prayer, Matthew’s version is clearly superior to Luke’s, and therefore if Luke had had access to Matthew’s version, he would neces sarily have used it instead of a more “primitive” version from a different source. Therefore, Luke could not have been familiar with Matthew’s version at all. But here, yet again, we have the same aesthetic argument that has been made already in several different contexts: Luke could not have reordered Matthew’s Gospel because Matthew’s order is superior. Luke could not have chosen an alternate birth narrative because Matthew’s version is superior. Luke could not have preferred a different version of the Lord’s Prayer because Matthew’s version is superior.

But did Luke think so? One suggestion made by some Farrer scholars is that he preferred a shorter, less-wordy version of the prayer, perhaps one he had learned before having read Matthew’s Gospel, Or perhaps he felt that the shorter version made the point better of making simple prayers to God. There are any number of possible reasons why Luke may have preferred his own version over Matthew’s. Here is yet another case where the expectation is that Luke must necessarily have had the same aesthetic sense as Matthew. But this cannot be a valid argument since it denies Luke the ability to make his own choices about the contents of his own Gospel (and, of course, the analysis already carried out in this book has shown that Luke’s aesthetic sense was quite different from Matthew’s), nor can it prove that Luke could not have read Matthew’s version of the Lord’s Prayer and still have chosen to use a version from another source (as we have also seen in other cases). The assumption that Luke should have treated his material in the same way Matthew treated his simply cannot stand.

Mosbo then summarizes some of the key points — ones I have covered above — in Olson’s argument as another possibility that

the version of the Lord’s Prayer in Luke 11 may actually be an intentional abridgment by Luke of Matthew’s version. Luke often abbreviates speeches and parables of Jesus from his other sources and especially tends to eliminate repetitious material. . . . . Olson’s argument therefore is that, if Luke has himself trimmed down Matthew’s version of the Lord’s Prayer, this is entirely consistent with procedures that we can identify Luke using elsewhere in his treatment of his written sources. Hence, it is impossible for us to make a determination that Luke’s version of the prayer is more “primitive” than Matthew’s, and indeed it may well be that its brevity is simply due to Luke’s own well-established editing techniques. Again, the argument from mutual primitivity simply cannot prove that Luke could not have used Matthew as one of his sources. (Mosbo, from an unpaged digital copy on Scribd)

There remain some differences between Matthew’s and Luke’s Prayers that the above discussions do not explain:

  • the difference between debts in one prayer and trespasses or sins in the other;
  • the slight change in wording petitioning for our daily bread.

I’ll return to Michael Goulder’s article for his explanation of those two differences in the next post.


Franklin, Eric. 1994. Luke: Interpreter of Paul, Critic of Matthew. Sheffield: Sheffield Academic Press.

Mosbo, Thomas J. 2017. Luke the Composer: Exploring the Evangelist’s Use of Matthew. Minneapolis: Fortress Press.

Olson, Ken. 2015. “Luke 11.2-4: The Lord’s Prayer (Abridged Edition).” In Marcan Priority Without Q: Explorations in the Farrer Hypothesis, edited by John C. Poirier and Jeffrey Peterson, Chapter 5. The Library of New Testament Studies. London: T&T Clark.


The following two tabs change content below.

Neil Godfrey

Neil is the author of this post. To read more about Neil, see our About page.

If you enjoyed this post, please consider donating to Vridar. Thanks!

15 Comments

  • Steve Ruis
    2019-06-29 15:58:45 GMT+0000 - 15:58 | Permalink

    When you say (in the second bulleted paragraph) that “Luke did think of the church in that way.” Did you mean “Luke did not think of the church in that way.”? It makes more sense that way.

    • Neil Godfrey
      2019-06-29 16:07:46 GMT+0000 - 16:07 | Permalink

      details, details, picky picky. as if a little three letter word “not” is going to change the meaning of anything! 🙂

  • Robert Jase
    2019-06-29 17:04:44 GMT+0000 - 17:04 | Permalink

    But believers tell me that Jesus wrote the Lord’s Prayer there for there cannot be different versions, it’s all an illusion.

  • JBeers
    2019-06-29 20:05:01 GMT+0000 - 20:05 | Permalink

    A request: would it be possible to have an open discussion sometime on the Lord’s Prayer? Perhaps line by line (1 or 2 versions) or the whole thing (including Didache)

    Or maybe too big a mess of comments would appear too fast.

    It could be all too easy for me and possibly others to divert this particular discussion with overly tangential questions or comments.

    Don’t bother to respond now or ever–just for consideration.

  • 2019-06-29 20:29:37 GMT+0000 - 20:29 | Permalink

    What would be accomplished by such a Bible study project?

    After spending quite a bit of time studying Luke and just finishing translating his entire Gospel and Acts I have noticed that not only does Luke shorten many prior texts but deliberately leaves out whole narratives and sayings (including sayings from prayers) that are not conducive to his apologetic agendas.

    eg. Luke leaves out Jesus calling Peter “satanas”… why? that is a very serious issue in Mark and Matt. He can’t have Peter looking scandalous because he needs him to be the new priest of the new Temple.(Luke is big time into Temple theology in his gospel and in Acts)…and he removes any scandal regarding Peter…. there is more to it… but that is one of the things that I consider problematic, among many others, re Luke’s so-called reliability as a good historian..

  • Danila Oder
    2019-06-30 00:33:39 GMT+0000 - 00:33 | Permalink

    Another explanation is that Luke’s project (Luke-Acts) was a response to Marcion. GLuke was a catholicization of GMarcion. Matthew learned about GLuke and said, “Good idea, but it doesn’t go far enough in reclaiming Jesus for Judean-ethnic people. My Jesus figure is pro-Law, and I’m going to rewrite the story to make that clear.” Matthew went back to the source and got a copy of GMark. Matthew then used GMark as the basis for GMatthew, but incorporated material from GLuke, and texts and ideology from Matthew’s own congregation. All of this in the 130s-150s.

    • 2019-06-30 04:47:04 GMT+0000 - 04:47 | Permalink

      Yes,

      Joseph B. Tyson is an ardent advocate for the Marcionite controversy and so this affects texts and shows how polemic and apologetic the NT really is … even in different genres… it is polemic from the very beginning…

      It is impossible in my view to actually get back to any “real” or “the true Christianity or “the Jesus stuff or things” …a common phrase in Luke-Acts…

      I don’t think this prayer goes back to Jesus… again,, I said earlier in my own bloggings that he may well have got it from the character in the story…John the Baptizer…who taught Jesus to pray!!!!

      I don’t think the Bible was written to me….(or anyone today for that matter)…(sometimes I get into a lot of caveats) but to someone else, singular or group,,, and this frees me up to examine these texts for myself without feeling any condemnation for reading the texts differently but with the use of highly tuned tools so we can not just accept claims of any theology….

      I also share from time to time that I don’t really understand why they want to preach at me given that I am an agnostic-atheist on a continuum no doubt 🙂 and that the term “unbeliever” in the NT does not mean “atheist”… There are no “atheists” in the NT!! It is someone attacking another person or doctrine of a different “faith”.. I am astounded how the average thinking reader or historian of the NT can’t see this, again and again!

      Many outside this blog would criticize us for even attempting an analysis of an alleged spontaneous prayer, and Jesus didn’t know Gk so what are we to make of the poetic structure of the text as well .. Jesus was allegedly illiterate…. yet the NT presents him as such…. I smell something fishy…

      Anyway,, keep having fun folks….

  • 2019-06-30 01:31:59 GMT+0000 - 01:31 | Permalink

    I think there’s too much sophistry going on in those arguments. I understand the effort, since Matthean posteriority is apparently so distasteful to most scholars.

    It’s not impossible that Luke shortened Matthew, with no apparent regard for the only prayer that the Saviour Himself ever taught. But the simplest answer is more likely to be correct: Luke’s prayer was earlier.

    • db
      2019-06-30 01:55:09 GMT+0000 - 01:55 | Permalink

      The simplest answer is that “Lectio brevior” is not applicable to Gospel According to Luke and it is demonstrable that “Luke frequently, in fact usually, abbreviated Mark”.

      Per Carrier (10 December 2018). “Adventures at the Society of Biblical Literature Conference, Part 3: Closing Out”. Richard Carrier Blogs.

      Alan Taylor Farnes of Brigham Young University gave a masterful talk thoroughly refuting the all-to-common principle in textual and literary analysis of the Bible called , “the shorter version is more likely the original version.” You’ll hear this said in different ways, like, “Luke’s use of Q is more primitive, therefore Luke’s text is closer to Q than Matthew’s” (therefore “Luke didn’t use Matthew,” and “therefore there really was a Q,” as the argument then continues to its target conclusion). The alternative requires Luke to have frequently abbreviated Matthew. But the Q theorists say, “No way would anyone abbreviate, so surely Luke isn’t abbreviating Matthew, therefore there must have been a Q!”

      Farnes so thoroughly refuted this argument it was beautiful to behold. He showed with numerous examples and extensive comparative charts of shared pericopes and passages, that Luke frequently, in fact usually, abbreviated Mark. Thus already refuting the major premise (that authors didn’t abbreviate their sources, or that Luke specifically didn’t abbreviate his sources).

    • Neil Godfrey
      2019-06-30 03:16:37 GMT+0000 - 03:16 | Permalink

      But now you know why some scholars do indeed argue that Luke post-dates Matthew even if you don’t like their arguments.

      As for the idea that there was a tradition of a prayer Jesus had prayed, that falls apart under Goulder’s analysis in the earlier post: https://vridar.org/2019/06/16/jesus-did-not-compose-the-lords-prayer/ — there was no oral tradition of a Jesus prayer.

      As for scholars finding Matthew’s coming after Luke to be “distasteful”, there are solid arguments grounded in a range of data, evidence, and not simply the one subjective view of aesthetics. They work with data-based analysis, not what they find “distasteful”.

  • Luke Burrage
    2019-06-30 03:04:18 GMT+0000 - 03:04 | Permalink

    Great post! Thanks for putting all this together.

  • JBeers
    2019-06-30 07:33:54 GMT+0000 - 07:33 | Permalink

    In my own case questions or comments would be anthropological or historical–such as ? of links to elements from other religions or ? allusions to infighting within Christian or Jewish sects of the time. But others might have far different points to bring up. In the routine and most familiar can be the most overlooked and ignored. I do not know.

    • JBeers
      2019-06-30 07:36:13 GMT+0000 - 07:36 | Permalink

      Oops. I was trying to reply to my own comment above wondering about having a general discussion sometime on the Lord’s Prayer.

      • Neil Godfrey
        2019-07-01 07:01:59 GMT+0000 - 07:01 | Permalink

        Maybe my latest post on the Lord’s Prayer — How Matthew Invented … — would be the best place for such a discussion but it looks like the comments on this topic have run dry. Maybe a discussion could be taken up in one of the forums like the Biblical Criticism and History Forum.

  • Pingback: How Matthew Invented the Lord’s Prayer (A Goulder View) |

  • Leave a Reply

    Your email address will not be published. Required fields are marked *

    This site uses Akismet to reduce spam. Learn how your comment data is processed.