Synoptic Gospels

The first three gospels in our NT are called the ‘synoptic’ gospels. They share a lot of material, and it seems very clear they were written with a great deal of mutual influence. Unpicking the particular pattern of influence is a tricky puzzle, commonly called the “Synoptic Problem”. Currently the most widespread view is that Mark was first (known as “Markan priority”), then both Luke and Matthew used Mark as a source, along with another shared source (known as “Q”, short for the German “quelle” which just means “source”) that Mark did not have access to. I’m not so sure, I tend towards a more complex picture that is somewhere between this and the “Farrer” hypothesis, which says that Mark was first, then Matthew used Mark, then Luke used Matthew*.

These debates are played out based on the patterns of sharing between the gospels. Obviously the same stories are found in multiple gospels (although, surprisingly very few are shared between the Synoptics and John – other than a couple of key events, John seems to have a conspicuously different life of Jesus). But beyond the same stories, the exact same words are often used, in idiomatic ways that are very unlikely to be coincidence. There are other clues: signs that one writer has copied another. The Fatigue argument** shows how writers can copy, intending to make a change to the copied text, but then get bored and revert to a straight copy. For example in the well known parable of the talents, Luke (19:17) has the servants given “cities” initially, but is obviously copying the same text Matthew uses (Matt 25:21 doesn’t say specifically what they are given), because by the end of the parable Luke has reverted to calling them talents as Matthew does.

When you learn about the synoptic problem, your homework is to colour in synoptic parallels: versions of the gospels printed with the same story from each gospel printed side by side. In this way you can see the material and the phraseology that is unique to each gospel, that is shared between each pair, and that is common across all three.

Now in the modern age, we can program computers to do our homework for us. And behold: a complete colour map of the synoptic gospels.

In this map the three columns represent the three gospels, Matt first, then Mark, then Luke. Each small square (2×2 pixels) in the diagram is one Greek word (from the Nestle-Aland 27 critical edition), so the size of the column tells you the relative size of the gospel (Mark is by far the smallest). The words are from left to right and top to bottom, just as you’d expect.

The color of the pixels tells you whether the word is shared in that story between the gospels (so there’s some complicated matching behind the scenes to compare equivalent stories, no matter which order they appear, and then map them back out into the proper order for diagramming). The color key is:

Red – Matthew’s gospel only.

Green – Mark’s gospel only.

Blue – Luke’s gospel only.

The other colors are combinations of Red, Green and Blue light:

Red+Green = Yellow – Matthew and Mark only.

Red+Blue = Magenta – Matthew and Luke only (so called “Double tradition” material, normally associated with the Q source).

Green+Blue – Cyan – Mark and Luke only.

And finally all three gospels agreeing is shown in black (I know, it should be white, but white is difficult to distinguish from the Yellow and Cyan).

So the diagram tells you an awful lot about the gospels: it shows that virtually none of Mark is unique to Mark. It shows the big block of magenta material near the start of Matthew and Luke which is the sermon on the mount and other so called “sayings” material of Jesus – the stuff that is thought to come from Q. It shows there is a lot more Yellow than Cyan, so Matthew sticks to Mark far closer than Luke does. And it shows there is more Magenta than Cyan, so Luke sticks far closer to Matthew than to Mark (giving further credence to the idea that Luke used Matthew who used Mark). You can also see that the birth narratives at the start of Matthew and Luke are very different, and that Matthew and Luke also have a number of other large blocks of original material.

So the colouring scheme used here is based on words in stories. there are other ways of dividing this up. In particular you can divide it just by which stories are shared. This doesn’t tell you as much historically, since we can’t tell if the stories were copied from one another or just in common circulation. But if you colour that way you get very different amounts of each color. You get much more black, for example – which is to say that the gospel writers often all write the same story, but in quite different words.

The final thing to say is about John. I said John is very different. How different? Well I couldn’t put John in the same diagram as above, because I didn’t have enough distinct colours to clearly show the 15 different patterns of sharing between 4 gospels. But here is John on its own. In this diagram the brown color is the stuff that is only in John. The other colors are as before: the stuff that John shares with Matthew (red), Mark (green), Luke (blue), Matthew and Mark (yellow), Matthew and Luke (magenta), Mark and Luke (cyan), and all three of them (black). Clearly John is a totally different beast.

Anyway, I’ve written enough on this. I love diagrams like this – diagrams that are incredibly complicated and specific, but that the overall patterns can be seen from those details.

* If you know anything about this topic, then it might not surprise you to know that I was a student of Mark Goodacre, who is one of the most longsuffering proponents of the Farrer hypothesis. I’m not sure I totally agree with Farrer, however, the Q arguments still sway me somewhat.

** The fatigue argument, one of the most elegant synoptic arguments, in my opinion, is also due to Mark Goodacre.


Filed under Uncategorized

20 responses to “Synoptic Gospels

  1. Hi Ian – is this your own work? Impressive! Have you looked at using bioinformatics approaches, such as those we use in comparing genomes? It strikes me that this could be fruitful 🙂 FWIW I subscribe to Marcan priority with a fair degree of later scribal fiddling…

  2. Really interesting and informative stuff, thanks for putting it all together. Also appreciate the Farrer hypothesis explanation. had heard Goodacre mention he does not prescribe to ‘Q’ in his podcasts, but didn’t know what his alternative was.

  3. Ian

    Shane, yes absolutely. I think there’s probably a huge opportunity to apply bioinformatics algorithms to this question. It was a topic I’d wanted to investigate myself before. I got a little way, but nowhere near anywhere useful. The visualizations on this page are my own yes, the source material was a tagged version of NA27, but all the analysis is original.

    attr – I have some diagrams somewhere on the priority theories I might try to make another post out of them, if they would be interesting.

  4. Hi Ian, Good stuff. Initial thoughts are that the architecture of the large-scale elements (i.e. the order of the stories) is important, nay critical, to the visualisation of these things. I’ll have to look up the various hypotheses still in serious contention (thanks for the pointers). I’ve also dedicated a wee post on my blog to this post, and flagged it up on Facebook. For the grand total of 5 people who ever come near my blog :-). What I think is fascinating about all this is that back at some point in time we actually have real people writing these things down, with real thoughts going through their heads, yet we have the slimmest of evidence to try to see things through their eyes, and dissect the various layers of edition and redaction that have occurred since the most of the material was generated. It would be great to stumble across a truly ancient corpus of texts, like a Christian Qumran, or Nag Hammadi II…

  5. I certainly would be interested if you find them, thanks.

  6. Ian

    Thanks Shane, yes a Qumran for early Christian texts would be very interesting. We have a number of good papyrus stashes, however. And one of the big themes of recent biblical scholarship is to re-evaluate the texts that have come to us through Constantinople in the light of the earlier Egyptian texts.

    The large scale elements can be swapped around, yes. Although it doesn’t give you as much of a sense of context in the diagram, I found. In particular it is difficult to order them well, since each gospel has its own ordering, so any order is artificial.

    attr – I’ll have a dig, probably not until the w/e now.

  7. Ian

    “recent biblical scholarship” = past 100 years….

  8. @ Ian
    I think you showed me these diagrams on your earlier site about 2 years ago. Have you touched them up since then?

    I remember this fun graphic you did on the mix of sources for the jumbled Christmas story.

    The colors are great at showing the lack of illusory homogeneity of the Christian scriptures.

    If we could do the same with our sense of self, I think we’d all see how illusory it is too.

  9. Boz

    I really like the way you have put this in to a diagram.

    It helps me understand the situation.

  10. Ian

    Sabio – Yes, it is similar. There’s some tweaks in the output, but otherwise the same. I think I’ll revisit the Christmas story later in the year, because I have some refinements to make to that. It would be interesting to see a color map of the self, yes. Although data might be hard to come by!

  11. The color maps of the Self come by introspective data — very bad research material but it is all we’ve got for now.

  12. Afternoon, chaps! I ran across an interesting paper by a chap called John Lee from MIT, where he has applied dot-plot techniques to the cross-analysis of Mark and Luke. Dot-plots are pretty common bioinformatics graphs that allow us to look at regions of synteny in genomes. I’ve blogged a brief take on it here:

  13. Ian

    Thanks for that link, Shane, its a good’un.

    I might have a go at tweaking my software to generate his kind of 2D plots.

  14. hi! im a student at St. Stephens university in New Brunswick Canada. I have to present a gospel “diagram” in a couple weeks, and iv chosen to look at Luke. I loved looking at you work here because we had to do something very similar. I emailed your site to my proff and told him to check it out. (i think i would like to have your “art” up on my wall in my living room) 🙂
    well i was just wondering if you could give me some ideas for presenting creatively the themes and major ideas of Luke. Examples of what other have done is pop-up books, mobiles, charts, things like that.
    Thank you for your time!

  15. Ian

    Thanks for commenting Rachelle!

    I don’t know really what they’re after, if they’re happy to accept popup-box and mobiles. Are you wanting to do something a bit more rigorous, or are you looking for something also that just gives your impression?

    I can think of lots of different data you might want to diagram. You could certainly do it thematically – you could go through and mark each story or verse with the main themes, and show how the themes recur through a time-line view of the book. Or you could map how popular different verses in the book are, or like this page, where the literary sources are. Both of those I can give you the raw data for, if you want to turn them into something better looking than I have.

    How about mapping Luke on a copy of a map of Israel, showing the trail of where Jesus is at each point in the story, so you can see the flow of the gospel. You can use a zoomed in inset map for the passion week.

    Or how about a big venn diagram showing the different themes in the 20 miracles in Luke (nature miracle, healing, Jesus speaks, disciples witness, etc, as many categories as you can think of). That will show interesting groups of miracle narratives.

    Those are just off the top of my head.

    It is important, I think to distinguish the content from the form. A popup book is not a diagram, though you could produce a diagram in the form of a popup book. The key thing is the question or data you want to show: the chronology, literary dependencies, use, presence of themes, categories of story, etc, etc. Then you can decide the best form to present that data to make it as clear as possible.

  16. Ed Gentry

    I love your diagram. Though pedagogically I may want to easily show different things.

    So since I affirm Marcan priority I’d be intersted in illustrating

    Matthew usage of Mark (irregardless of what Lukes also uses)
    Luke’s usage of Mark (irregardless of what Matthew also uses)

    Luke and Matthew or Q.

    Hmmm being both a Bible geek and a computer geek, I was wondering if an interactive HTML app could be made. It shouldn’t be too hard. Would you be willing to share the data? I could build a simple web site that would interactively color the gospels.

  17. Ian

    Thanks Ed. I appreciate the appreciation!

    Because I tried to stick to the color channels (except black), you should be able to knock out the color component you aren’t interested in. If you’re interested only in Matt-Mark, then knock out blue (any graphics editor will be able to do this). You’ll still see triple tradition (because I used black not white), but you won’t see Matt-Luke and Mark-Luke. If you really wanted to be accurate, you could turn black into white first then you wouldn’t even see triple tradition material.

    As for the raw data. Yes I guess so, although in some ways the images will be as easy to work from as anything. What format did you want? I can give you a list of quads if you like (in matt, in mark, in luke, in john) for each of the four gospels.

    Having an interactive visualiser for this stuff I’ve thought about before, but don’t have the time. Have you seen the Glo digital bible? It doesn’t do synoptic visualization, but does some very nice other visualizations.

    What I’d actually love to do is a visualization based on texts – so synoptic would be one example, but also being able to see within one book the text-divergences between text families. Where is the Western tradition a significant departure from the Alexandrine, etc? My sense is that most folks don’t have a great handle on those patterns of difference.

  18. Pingback: The Books of the Bible | Irreducible Complexity

  19. Thanks Ian, for sharing your diagram. I translated the NT into Portuguese. During the translation process from Greek into Portuguese there were found to be many parallel and similar passages between the synoptic gospels with formulations of two or more words (up to entire verses) with exactly the same words in Greek or with minimal differences (nothing new about that!). To translate the words of these passages equally into Portuguese, I felt the need to make a close comparison of the three gospels. So I compared all words and gave a color to each gospel. This was done by hand. The material became very useful to do some very interesting research. In this process I found your diagram and placed it side by side to a very similar diagram with my results. The research resulted in many very interesting charts and diagrams (a total of over 2000 pages of research material). My presentation now has 183 slides and isn’t finished yet. It is being translated from Portuguese into English, and will be published shortly. This now is objective material and will seriously question many of the synoptic theories.
    My desire: would you give me the permission to include your computerized diagram to be placed side by side to my hand made diagram, in the publication of my book/presentation? If this would be possible, I would be thankful to know your full name, in order to give you the credit. Thank you so much!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s