Friday, 29 March 2013

Was the Author of Matthew's Gospel Jewish? (Part 3)

I think I should give a résumé of the argument so far. As you read, I'm sorry for the formatting of this post - I can't get my Hebrew to line up with my translations given the right-to-left nature of Hebrew. Please bear with this.

A Summary of the Arguments So Far
Matthew's Gospel has, for a long time, been considered the most jewish of all four gospels considered canonical by christian churches. In my series of posts about Mark Goodacre's critique of Francesca Stavrakopoulou's views on the virginal conception, I pointed out that Goodacre seemed to take it for granted that Matthew was a devout Jew who knew his scriptures well. Goodacre is far from alone in his viewpoint, but actually the question of whether Matthew was a devout Jew, or even not a Jew at all, is a live one in modern biblical criticism.

I should mention that in contrast to Matthew's Gospel, Luke's Gospel has traditionally been considered as largely gentile.

Until the advent of biblical criticism, this "jewishness" view of Matthew's Gospel was largely based on "tradition" of who wrote the Gospel, rather than on a detailed analysis of the Gospel text itself. There are good reasons for thinking that this tradition may not contain reliable information handed down by those who actually knew who wrote the Gospel. This doesn't mean that the tradition is necessarily wrong, but it does lend importance to examining the text of the Gospel itself to see if there is anything there which might give a clue.

But how can you tell from a text if an author is jewish or not? And if Matthew was jewish, then how devout and knowledgable about judaism was he? In particular, are there signs which betray his intimate familiarity with the judaism(s) of his day? Alternatively, are there other signs which indicate that he considers himself separate from the religion, or lacking in knowledge about it?

These questions were addressed relatively briefly by Udo Schnelle in his The New Testament Writings, which has become a standard technical introduction to the New Testament (taking over this mantle from Werner Kümmel's Introduction to the New Testament). He set out what I have called "for and against lists" for Matthew's jewishness. Each list looks rather impressive when read in isolation from the other, although I spent some time in my last two posts criticizing the strength of some of the arguments on both sides.

A Further Argument
However, in his list of arguments against Matthew having been jewish, Schnelle has missed one which is very common these days. This argument concerns Matthew's complete misunderstanding of "Hebrew parallelism", a feature found in almost all poetic books of the Hebrew scriptures of his day and, albeit to a significantly lesser extent, in some prose texts . I have to confess I don't know from where this argument originates; like so many people, I first came across it thanks to Bart Ehrman's The New Testament, A Historical Introduction to the Early Christian Writings (p.118, fourth edition 2008), but it almost certainly didn't originate from Ehrman.

Before turning to the Matthean text, we need to take a look at this concept of "Hebrew parallelism".

What is Hebrew Parallelism?
Even if you have never come across this term, if you are familiar with the large percentage (exact figures are disputed - but it's often estimated at about 33%, sometimes significantly more) of the Hebrew bible that is poetry, you will undoubtedly have read a good deal of Hebrew parallelism, although perhaps you have read it in translation; it is probably a feature that you have kind of figured out yourself - because it's that widespread.

I've not been entirely satisfied with definitions of Hebrew parallelism that I've found, so let's start with an example. Psalm 103:10 says:
לֹא כַחֲטָאֵינוּ עָשָׂה לָנוּ
וְלֹא כַעֲוֹנֹתֵינוּ גָּמַל עָלֵינוּ׃
Not according to our sins has he done to us,
And not according to our iniquities has he dealt with us.
You can see here that the two lines of the verse treat one subject. The second line says pretty much exactly the same as the first, just in different words. Of course I've put the verse in two lines so as to show this (early manuscripts probably didn't line it up this way), but the principle still remains that there are two clauses outlining the same thing.

The majority of Hebrew parallelism occurs in this two-line couplet structure. But there are some examples of three-line, four-line and even greater number of lines to parallelism. I'll restrict myself to two-line parallelism because, not only is it by far the most common type, but it's the easiest to spot and decipher.

Psalm 103:10 given above is an example of synonymous parallelism, and there are indeed a number of different categories of parallelism; in fact there are a number of different ways to categorize parallelism, and I'll go through one of these here. Please do bear in mind that categorisations are loose and challengeable.

1) Synonymous Parallelism
We have already seen Psalm 103:10 above where the two lines mean pretty much exactly the same thing. I haven't seen any statistics on this, but from my reading I would say that synonymous parallelism is the most common type. Just how close do the two lines need to be to each other? Well here's another example where the meaning is very close indeed. This is the second half of Micah 6:2:
כִּי רִיב לַיהוָה עִם־עַמּוֹ
וְעִם־יִשְׂרָאֵל יִתְוַכָּח׃
For Yahweh has a quarrel with his people,
With Israel he disputes.
I think you can see that the two lines mean the same thing; they are slightly further removed from each other than the previous example, as the second line relies on the implicit subject rather than repeating "Yahweh", and "his people" is paired with "Israel". If you read Hebrew you'll see the syntactical difference between the two lines as well (a possessive form changes to a verbal form).

Despite these differences, the two lines remain clearly synonymous - they mean the same thing.

Because synonymous parallelism is so common, it's worth giving a third example. Here's Proverbs 6:20:
נְצֹר בְּנִי מִצְוַת אָבִיךָ
וְאַל־תִּטֹּשׁ תּוֹרַת אִמֶּךָ׃
Keep, my son, the commandments of your father,
And do not forsake the instruction of your mother.
Now this one is also considered synonymous, even though we have a different parent in each line. Yes, there is this difference, but the overall idea is the same - obey your parents!

I wanted to include this example to show that synonymous parallelism can contain differences between the two (or more) lines; it's the overall idea expressed by the lines that needs to be the same. Indeed small differences between the lines are quite common, inevitably so - it would get rather boring if every line in synonymous parallelism were exactly the same!

2) Antithetical Parallelism
This is where one line gives the converse of the other line. Proverbs 10:1 gives another nice familial example:
בֵּן חָכָם יְשַׂמַּח־אָב
וּבֵן כְּסִיל תּוּגַת אִמּוֹ׃
A wise son brings gladness to a father,
But a foolish son is grief to his mother.
Here the two lines don't say the same thing, rather they say the converse which, taken together, give a unified overall idea. Note again how in the example here we have two different parents in the two lines - we don't need them to be identical for parallelism to work; I'm sure the writer would have considered a foolish son to be grief to his father just as much to his mother!

3) Chiastic Parallelism
In chiastic parallelism it is the order of elements that count. The first line will provide an order of elements (say, A then B then C) which will be reversed in the second line (C then B then A). Here's probably the most simple example - the opening of Jeremiah 4:5:
הַגִּידוּ בִיהוּדָה
וּבִירוּשָׁלִַם הַשְׁמִיעוּ
Announce [it] in-Judah,
In-Jerusalem make-[it]-heard
This chiastic structure is probably more difficult to spot if you don't read Hebrew - you're reliant on translators conveying it well, which is difficult to do. Even in this simple example, if you didn't know that "in" is a prefix to both "Judah" and to "Jerusalem" you might miss the chiastic structure in translation. I've resorted to using hyphens to indicate single words in Hebrew - I don't think many translations would go to these lengths!

Note how the two lines are synonymous, although Judah and Jerusalem stand for each other (as they do in many other places).

As chiastic parallelism is often lost in translation, I'll give another example to whet your appetite. Here's Psalm 89:35:
לֹא־אֲחַלֵּל בְּרִיתִי
וּמוֹצָא שְׂפָתַי לֹא אֲשַׁנֶּה׃
Nor will-I-break my-covenant
The-utterance of-my-lips not I-will-change
Hmm, my hyphens don't really help that much with this one (e.g. I've had to single out "not" in a peculiar place in the second line), but I hope you get the idea.

4) Staircase Parallelism
Staircase parallelism is also about structure. Whereas chiastic parallelism could be represented by two lines as follows:
staircase parallelism would be more like:
That is, you climb the stairs twice, but the second time you arrive at a different destination. Here's Judges 5:12:
עוּרִי עוּרִי דְּבוֹרָה
עוּרִי עוּרִי דַּבְּרִי־שִׁיר
Awake, awake Deborah!
Awake, awake, chant a song!
5) Janus Parallelism
In Roman mythology Janus was a two faced god who, thanks to his two faces, could look in two directions at once - right/left, front/back or more figuratively future/past. In Janus parallelism there is a word which has two meanings, and the parallelism hinges on this word - a sort of double entendre.

Janus parallelism is immensely difficult, if not impossible, to translate. So you'll have to bear with me for this example. Here's Canticles 2:12
הַנִּצָּנִים נִרְאוּ בָאָרֶץ
עֵת הַזָּמִיר הִגִּיעַ
וְקוֹל הַתּוֹר נִשְׁמַע בְּאַרְצֵנוּ׃
Okay, so how do we translate this? Well it depends how you take the word in red (if this has come out in red on the blog). This word can have two meanings; firstly it could be "the pruning" which fits in well with the first clause; secondly it could mean "the singing" which would fit in well with the second clause. So we can translate this as follows (I've numbered the clauses so you can see where the break is):
(1) The blossoms have appeared in the land,
the time of the pruning/singing has arrived
(2) and the voice of the turtledove has been heard in our land.
And why is this parallelism? Because the whole verse is about the signs of the arrival of spring.

Janus parallelism is quite rare in the Hebrew bible, which is perhaps fortunate considering how difficult it is to translate!

Hebrew Parallelism - a Summary
There are other categories of parallelism that I could outline. Alternatively, parallelism could be divided differently into different categories.

The point that I'm trying to make from all the above is that parallelism is very widespread in its different forms in the Hebrew bible. If you are someone who is familiar with the Hebrew bible I hope that, as you have read the few examples I've given above, you will think to yourself, "Oh yes, I see this all the time". Perhaps, if you read in translation, you'll miss particularly chiastic and Janus parallelism, because these ones are very hard to convey in translation but, overall, parallelism should be very familiar to you indeed.

Similarly, if the author of Matthew's Gospel was familiar with the Hebrew bible, he should also have been familiar with this form of poetry. Undoubtedly, he would not have divided parallelisms into categories as I have done; but he would recognize what was going on when he met an example of parallelism.

Matthew's Unfamiliarity with Scripture
We have already seen in previous posts some examples where the author of Matthew's Gospel shows dubious-at-best familiarity with jewish scriptures of his day. I'll now turn to a passage in Matthew which shows just how unfamiliar he was with the very widespread feature of Hebrew poetry that is Hebrew parallelism. In fact he was so ignorant of this feature that he created a little absurdity in his Gospel.

If you remember, it is widely accepted in modern scholarship that Matthew's Gospel took Mark's Gospel as one of his sources. Matthew has the story of Jesus' arrival in Jerusalem in 21:1-9, which he has taken largely from the equivalent story in Mark 11:1-10. The stories share, amongst other things, the following elements:
  1. Jesus is on his way to Jerusalem and comes near to the village of Bethphage.
  2. Jesus instructs two disciples to go and find mounted transport, with specific instructions as to what to say to anyone who questions this action.
  3. The two disciples carry out this instruction and bring mounted transport to Jesus.
  4. They lay cloaks (ἱμάτιον) on the mounted transport, and Jesus mounts on top.
  5. Proceeding towards Jerusalem, they are met by crowds who lay branches in their path and cry "Hosanna".

There are two major differences between the two versions, both of which are linked. In the earlier text (Mark's) Jesus, quite reasonably, is sitting on one animal - a horse (πῶλος). This is sometimes translated as "colt" or even "foal", but BDAG (a standard Greek lexicon) makes it clear that this term can refer to a horse of any age, and clearly it was old enough to ride.

In Matthew's derived version, bizarrely Jesus sits astride two steeds (καὶ ἐπεκάθισεν ἐπάνω αὐτῶν - notice the use of the aorist and how this describes the completed action); one of these is a horse (πῶλος) just as in Mark's Gospel; the other steed is an ass (ὄνος).

Why does Matthew introduce this extraordinary change to Mark's version, and render his account so unrealistic? Well the reason for this is the other difference in Matthew's version: Matthew has gone searching for a scripture to provide a fulfillment for this episode. He has hit upon Zechariah 9:9 - a quite long verse which has two (arguably three) examples of parallelism in it - and cited it as a fulfillment quotation in Matt 21:5.

Matthew doesn't get the first example of parallelism in this verse at all - a shame really, because it is a very clear example of synonymous parallelism:
גִּילִי מְאֹד בַּת־צִיּוֹן
הָרִיעִי בַּת יְרוּשָׁלִַם
Rejoice greatly daughter of Zion,
Cry out daughter of Jerusalem
A more exact synonymity would be hard to find! Matthew, strangely, has rendered this as (21:5):
Εἴπατε τῇ θυγατρὶ Σιών
Say to the daughter of Zion
So not only does he miss the parallelism here, he has a mistranslation of the text, which again is a shame, because the Hebrew text here would have fitted his context quite nicely. Instead, perhaps he has confused the opening of Zechariah 9:9 with a clause from Isaiah 62:11, which unfortunately really doesn't fit his context at all.

But Matthew's biggest problem is missing the parallelism at the end of Zechariah 9:9:
עָנִי וְרֹכֵב עַל־חֲמוֹר
וְעַל־עַיִר בֶּן־אֲתֹנוֹת׃
Lowly and riding on an ass,
and on a male-ass, son of a she-ass.
How many animals are we talking about here? Just the one, evidently as we're dealing with synonymous parallelism (albeit the one ass is an offspring of a second, its mother), but Matthew just doesn't get it! Matthew fails to recognize that we have parallelism in the text, and so thinks he has two animals at work here. So he takes a perfectly reasonable story from Mark and renders it absurd by having Jesus mount two animals at once.

Is this really the kind of mistake that someone familiar with Hebrew scriptures could make? If he was such an educated Jew as is frequently maintained (as Mark Goodacre did in his podcast), then he had a singularly brainless moment here! Personally, I think this beggars credibility. Rather, as we have already seen, Matthew's knowledge of jewish scripture of his time was rather limited.

Finding A Path Through The Data
So what are we to do with all this? We have strong evidence both that:
  • the author of Matthew's Gospel was very familiar with judaism and had a strong connection with it, yet
  • that he felt distant from judaism and had complete failures to understand its texts.

I think the first thing to do is rule out the extremes. It simply isn't possible (pace Goodacre) to say that Matthew was a very devout and educated Jew, steeped in the scriptures and judaic customs of his day. Nor is it possible to hold that Matthew was entirely disconnected with judaism and uninfluenced by its ideas, traditions and scriptures.

Of course this leaves scope for a very wide range of options in between these two extremes, and a degree of judgement must be exercised when weighing up the evidence we have spent 3 posts examining (and perhaps others will have more evidence).

In terms of Matthew's familiarity with scriptures, we have seen that his knowledge is incredibly shaky; he seems to have skimped on his research into this area (if he did any); he is reliant on the Greek translation of his day, but can't get that right; he is unfamiliar with basic principles of Hebrew scriptural writings. If he was a Jew, he certainly wasn't one who had the opportunity or inclination to familiarize himself with Hebrew scriptures. But this still leaves plenty of room for him to have considered himself a Jew.

Matthew, if he was jewish, has also abandoned the idea that Israel has a unique place in god's salvific history. This is a considerable shift for a Jew to make. It is certainly very possible the later the Gospel is dated. If the Gospel were dated, as many scholars do, to the 80s, I find it hard to believe that a Jew would have moved this far in their thinking.

One Proposal Among Many
If Matthew was not jewish, then the strong jewish features of his Gospel have to be accounted for somehow. I think it is inevitable that Matthew's religious community at the time of the writing of his Gospel must have contained significant numbers of Jews; if this weren't the case it would hardly be possible to explain the numerous jewish references in his Gospel.

Is it possible the Matthew was not only writing for a jewish readership but, being himself a Gentile, was assisted by jewish members of his religious group? Perhaps in this scenario, Matthew was very keen to become educated in jewish beliefs, customs and scriptures; after all the founder of his religious movement, Jesus, had himself been a Jew and the movement had started within judaism and was probably still strongly associated with judaism in Matthew's day. Perhaps this keenness was not accompanied by the time and opportunity to get more than a basic understanding of the jewish religion; he himself would be dependent on Greek translations of jewish scriptures, whether those already in circulation (the Old Greek, aka LXX) or those provided on spec by jewish members of his religious group.

I think the above would represent one possibility that fits the data; there are, however, numerous others, including some where Matthew is jewish himself, but operates with limited knowledge of scriptures in Hebrew and a significant degree of detachment from the religion of his birth.

In the final examination, other than eliminating the extremes as outlined above, we simply don't have enough data to explain the tensions in the Gospel pertaining to Matthew's jewishness or lack thereof.


  1. I’ve enjoyed this series regarding Matthew’s Jewishness. I agree his failure to capture this Hebrew idiom would strike against his being Jewish. Additionally, I struggle with his wrestling Mark’s “Jesus the Nazarene” to making Jesus “from Nazareth.”

    Just a lurker posting a thanks.

  2. It seems to me that the easiest explanation of Matthew's background is that he is one of the many gentiles who were seeking access to Judaism, a "god-fearer," as they were called. A gentile reading the Jewish scripture in translation, who desperately wanted to join the Jewish club. He didn't really understand the rules, or the literature. He was one who did think, unlike Paul, that you had to keep the rules to get right with god. Matthew's early text, without the birth narrative, was the gospel of the Ebionites (per Irenaeus), and they too were extreme rule-keepers: most likely new converts to Judaism, alight with the fire of fresh conversion (and probably circumcision).

  3. Winking-demon, I think your suggested scenario is one that fits very nicely within the boundaries of the extremes I’ve laid out, and is entirely possible. As for the Gospel of the Ebionites, I think everyone would love to resolve the conundrum of exactly what it looked like, and how it related to the Gospel of Matthew.