It would be the same evidence that Luke knew Matthew.

And there are some scholars who have tried to legitimately argue that (the Wilke hypothesis still has defenders).

So the question becomes: what evidence is there of the direction of borrowing?

And there are three categories of that evidence: authorial fatigue shows it was Luke using Matthew not the other way around (per Goodacre); Luke-Acts used the Antiquities of Josephus but Matthew never uses any such material and seems unaware of it (so more likely came earlier); and redaction analysis indicates it was Luke transforming Matthew not the other way around (i.e. we can explain well the content of Luke only by supposing it is altering Matthew, but we can’t explain Matthew that way; e.g. Matthew’s Nativity makes sense as a transformation of Moses literature, in and out of the Bible, such as The Biblical Antiquities, but Luke’s Nativity only makes sense as a transformation of what Matthew thus constructed).

Other evidence gets tangled up in debates over Markan priority. But those debates are in part solved by the sort of evidence I just mentioned.

It’s great to see a detailed well written take down. Thanks again.

Any time i need to discus this with someone in future i’ll direct them here

There is some evidence of that.

For example, the oft-noted conundrum of the intersection of words in the “spit-prophecy” passage in the crucifixion narrative can be explained by “Q had a crucifixion narrative” or by “Luke picked-and-chose between Mark and Matthew” (the usual hypotheses). But it can also be explained by “Mark originally contained the whole passage, and Matthew copied it, and then it became corrupted with a dropped word or two, and Luke copied the version he had.”

Another example: Matthew and Luke pretty consistently use Nazorian everywhere Mark has various other formulations (exceptions are rare); the best explanation of this is that Mark originally said Nazorian everywhere also and this came to be “emended” to sound more like a word for “denizen of Nazareth” (like Nazarenos) in various ways (producing inconsistent corruptions in our copy). Matthew and Luke also show corruptions in the manuscripts over time in the same direction. This would mean our copy of Mark has become corrupted even in relation to Luke’s, not just Matthew’s (unless Luke’s copy had those corruptions, and Luke relied on Matthew; or Luke’s only had some of the corruptions; etc.).

Another example: the supposed geographical mistake in Mark that Matthew supposedly corrected regarding the position of Tyre can be explained by a simple scribal corruption; i.e. Mark probably had it right, and it became corrupted after Matthew, and Matthew was just replicating Mark and not correcting him after all.

Another example: many manuscripts of Matthew say “Jesus Barabbas,” twice. It’s very unlikely this was added. It’s more likely it was deleted (and Origen even suggests motives for it, in his commentary on Matthew, where he is aware of the alternate readings and suspects them to be original). It exactly matches the literary intent of the story in Mark (see OHJ, pp. 402-08), so is more likely to have been in Mark, and then copied over by Matthew, but then deleted from Mark (and maybe Matthew, though not necessarily) by the time Luke got ahold of it.

I have not done any survey of the literature and variants to see if more evidence exists or if anyone else expert has checked and argued for or against this in books or journals. But I’ve been interested to find out.

-:-

Note:

This “could” explain the Great Omission, but I am not confident it does. That probably can only have been put there by the original constructor of Mark (see my discussion in OHJ, pp. 412-17), but there are reasons someone who was an anti-Marcionite editor would want to get rid of the second and third cycle of it. But IMO, that editor is most likely Luke themselves, and not a prior redactor of Mark (otherwise Luke would have used the material as replicated in Matthew, unaware that it was ever in Mark; its thorough expungement suggests awareness of what was being left out as it stands in our Mark, and so it was probably in his—and Luke has an additional motive, which is to cut a large chunk of material to make room for things he wanted to add). Meanwhile, it can’t have been added to Mark because its structural connection to the first cycle is too elegant, and that can’t have been replicated by “reverse engineering” it out of the corresponding material in Matthew, who broke up that structure. So the Great Omission must have been in Matthew’s copy of Mark.

So it’s well-accepted that Luke was not only a plagiarist, but a lazy one. Which demonstrates that he actually was not slavishly committed to accuracy when he put forward sacred texts.

Which, as you note, means that even if there is a Q, we could never confidently reconstruct it.

That feels like a colossal own goal.

Has anyone in the Q literature even tried to make some kind of cases that we can see a difference between Luke’s own additions to Mark and Matthew and other materials he may be citing, the way you think he could be citing court transcripts for Acts?

They have, and sometimes with a little success, but it’s a limited tool. To make statistical predictions usually requires longer strings of material for high confidence, otherwise seeming variations can just be anomalous. Authors, after all, were specifically taught in school to vary their style. Conversely, it has been shown that Luke was very fond of emulating the style of the Septuagint (just as Joseph Smith did the KJV), which makes it harder to isolate what is Lukan, when much of what is Lukan is intelligently designed to look Septuagintal.

But for an example, Goodacre’s studies of “authorial fatigue” contain examples of what you have in mind, where he shows that as Luke progressed in copying material from Mark and Matthew, he got tired of being innovative and increasingly just started slavishly copying. This can be seen pericope by pericope, where a borrowed story Luke will change up until a paragraph in he gets tired and makes fewer changes. There is no fixed law of physics here though. How lazy he got and how quickly and what effect this had in each case varies a great deal. But it’s possible to look across all of these cases and collect some statistically significant patterns. Where things are changed more, tells us something of how Luke himself preferred to write.

However, this can become a Catch-22. Since it means half of Luke isn’t Lukan, due to his lazy copying pattern. So often some scholar will go in and see all the instances of Luke using some construction, and then declare that Lukan style, when in fact those are all the lazy bits where Luke is just borrowing Mark or Matthew’s style. So when one gets to Acts, it’s difficult to figure what could be more or less Lukan, because we don’t have any of his source texts there to compare (if even there were any).

So, for example, I’m quite sure Luke’s summary of the letter of Claudius Lysias has been simplified and paraphrased, not copied verbatim (there are elements of such letters we have for real that it lacks), though in principle “authorial fatigue” might predict that it is more verbatim towards the end than at the beginning. But that doesn’t really help us much. It’s too short to get anything definitive about a difference in style from top to bottom. And we don’t have any original (or any other derivation even) to compare it to. So there is no way to prove Luke used a real letter with this technique.

Yeah, as Richard notes, that doesn’t mean the method is illogical. It means they are using it illogically because they have a woobie to want to get at other theories. When someone tells you something embarrassing, you don’t know what motivated them to do that without more evidence. Maybe they are the kind of person who likes to be embarrassed? Maybe they are using the fact that you will trust them after such an admission to set up a con, a very common tactic that con artists and grifters use? (For example, the “I was such a bad man before Jesus” trope is such a cliche, and you would only trust them that they were in fact such a bad man if you hadn’t yet heard their sales pitch, because then you have an answer for why they had that embarrassing material). Maybe they’re incompetent? (I’m reminded of the utterly creepy way that David Wood went off about how because he was an evolutionist he was basically a sociopath, talking in such a way that confirmed that, actually, no, man, you still are a sociopath, you’ve just sublimated it with God talk). And that’s assuming that it was embarrassing to either them or their target audience. You may be importing your anachronistic assumptions “cringing” on their behalf when that was never on the table.

In particular, every statement against interest argument proceeds from the assumption that “This information they gave is embarrassing, therefore they must value telling the truth more than the embarrassment”. But that only works if you have good reason to think the source is trustworthy. Because the only thing you know is that the person who said the embarrassing thing was more motivated to have you believe X than not believe the embarrassing fact Y. They may want you to believe X for some deeper reason. This happens all the time in politics: Someone apparently making an argument out of their normal worldview is very often them setting up for something else. Like when a crypto bro tells you that, no, despite their market libertarianism they actually don’t believe in intellectual property rights (a remarkable claim apparently against interest)… because they want to mint their new coin based off of a copyright-protected IP (which makes the previous admission no longer very remarkable).

But one would never say that a statement against interest is somehow an irrelevant piece of information. it just doesn’t mean what one might naively think it does. I find it funny how Richard pointed out that legal scholars would take Biblical studies to task for basically having a pre-scientific rube’s idea of how embarrassing statements work. It shows the utterly infantile state of the field in basic methodology, which is your point.

So… it sounds, hilariously, that the only theory even worth considering would be L. Because the only thing that matches everything that isn’t in evidence that isn’t just “Luke made it all up” which I think you’ve shown is a perfectly reasonable explanation would be that he used Mark and Matthew, and then introduced his own theological ideas by also referring to his own church’s traditions and ideas, and the latter would clearly be the only source he ever had any reverence for because it was “his team”.

That sounds like what is usually promoted as the hypothetical source L.

Speculation is idle. It has no place in assertions of what’s likely. Anything is possible. That doesn’t make it probable, much less knowable, even less usable as a premise in any further step of reasoning.