Supplement to The Great Expansion: The Night Trial, the Minor Agreements, and the Two Editions in Transmission

Abstract

This supplement to the second study of the series (Gentile, 2026b) examines a further candidate for the expansion of proto-Mark, prompted by correspondence: the night trial before the Sanhedrin (Mark 14:55–64). The evidence — an insertion seam with resumptive stitch, the emptied morning slot at 15:1, the shape of Luke's hearing, and a mimetic argument published by Emil Wendling in 1908 — supports the candidate in a specific form: not outright insertion but relocation and expansion of a morning hearing that proto-Mark already contained. Pursued to its consequences, the candidate forces one refinement of the series' model: Matthew's copy of Mark, canonical in structure, carried sporadic proto-Markan wording from a contaminated line of transmission. The refinement is additive rather than corrective — nothing in the previous studies is retracted — and it is productive: it derives Streeter's "fundamental fact," accounts for the texture of the minor agreements, reads the conflate variants of Mark's own manuscript tradition as the same contamination caught in the act, explains the comparative instability of Mark's text, and issues a pair of distributional predictions by which it can be tested or broken.

Keywords

Gospel of Mark; proto-Mark; Synoptic Problem; minor agreements; Sanhedrin trial (Mark 14:55–64); Marcion's Evangelion; textual criticism; Ur-Markus

Table of Contents

• 1. Preface: this supplement and the parent study
• 2. The candidate, and why it takes the relocation form
• 3. The seam and the emptied slot
• 4. The shape of Luke: the trial minus the expansion
• 5. Wendling's argument: the dressing cut from the trial next door
• 6. From deflection to declaration
• 7. The agreements of Matthew and Luke: a contaminated Mark in Matthew's hands
• 8. The mockery: a text that grew in one direction
• 9. The remaining fit
• 10. Grading, and what would count against this
• References

1. Preface: this supplement and the parent study

The second study of this series argued that canonical Mark is an expansion of an earlier gospel, proto-Mark, and identified the principal expansion materials by a battery of criteria: Lukan evidence, internal seams and doublets, the witness of Marcion's Evangelion for the proto-Lukan layer, and thematic fit (Gentile, 2026b). This supplement applies the same battery to a candidate raised in correspondence after that study was drafted: the trial before the Sanhedrin.¹

Two things should be said at the outset about what this supplement is and is not. It is not a correction: nothing argued here retracts or revises any finding of the previous studies, whose statistical and documentary arguments never depended on the assumptions this supplement examines. It is an extension, and of a particular kind: the candidate, pursued honestly, forces the model to say something it had not yet said about how the two editions of Mark travelled — and the thing it is forced to say turns out to explain several famous features of the synoptic data that the parent study never addressed. A model that meets a hard case and responds by issuing testable predictions is behaving as a model should; the reader is invited to hold this supplement to that standard, and its final section states what would count against it.

The argument runs as follows. Section 2 states the candidate in the form the evidence requires — relocation and expansion, not simple insertion — and shows why the simpler form fails on the series' own tree. Sections 3–5 make the case: the seam and the emptied slot, the shape of Luke, and Wendling's mimesis. Section 6 draws the consequence of greatest weight for the series: the expansion converted a deflection into the gospel's one open messianic declaration. Section 7 confronts the cluster of Matthew–Luke agreements the pericope contains and derives the refinement. Section 8 reads the mockery's textual history as corroboration. Sections 9–10 complete the fit and state the grading and the tests.

2. The candidate, and why it takes the relocation form

The proposal is this: proto-Mark's passion contained a hearing before the authorities — in the morning, at the position where 15:1 still stands — and it ran as Luke's hearing still runs: the Christ question put, and answered with deflection. Canonical Mark relocated that hearing into the night, setting it inside Peter's denial, and expanded it into the trial we read: false witnesses, the temple charge, the silence, the open ἐγώ εἰμι, the blasphemy verdict, the capping mockery.

The simpler hypothesis — that the trial is outright insertion, absent from proto-Mark altogether — fails a test internal to the series' own tree. Proto-Luke has a hearing (22:66–71), and the Evangelion attests it for that layer (BeDuhn, 2013, p. 189); its core — "Are you the Christ?", the Son of Man seated at the right hand — stands in near-Markan wording. If proto-Mark had no hearing at all, that material has no source: not proto-Mark, and not Matthew, whom the Evangelion's attestation places out of reach for this layer. Relocation dissolves the problem: proto-Luke's hearing has a source because proto-Mark had a hearing — in the morning, where Luke still keeps it. On this model 15:1 stops being an accidental doublet and becomes the crater of the move: the morning time-marker and the convening retained, because the handover to Pilate needed them; the content gone, because it had been carried forward into the night.

3. The seam and the emptied slot

The marks in canonical Mark's own text are what such a move would leave. The trial interrupts a scene in progress: at 14:54 Peter has followed into the high priest's courtyard (ἐν τῇ αὐλῇ) and sits warming himself; the trial occupies vv. 55–64; and at 14:66 the narrative must re-establish him — "and while Peter was below in the courtyard" (κάτω ἐν τῇ αὐλῇ) — the resumptive repetition that is the classic stitch of an insertion.² And after the night trial has run to its verdict, 15:1 opens: "and immediately in the morning the chief priests held a consultation (συμβούλιον) with the elders and scribes and the whole council" — convened as though for a proceeding, with no proceeding left to hold. The awkwardness of 15:1 after 14:55–64 is among the oldest observations in the criticism of the passion narrative: Bultmann took the night trial as secondary within an older passion report, and Winter pressed the case at length; Donahue argued the scene as Markan composition throughout.³ On the relocation model the verse is neither error nor mere ratification: it is the hearing's original slot, kept because the morning was structurally necessary, emptied because its contents had moved.

4. The shape of Luke: the trial minus the expansion

Luke's passion sequence at this point is not a rearrangement of canonical Mark's; it is the shape canonical Mark's takes when the proposed operation is undone. Luke has the arrest; Jesus led to the high priest's house and Peter following (22:54); the denial, continuous and uninterrupted (22:55–62); a mockery of Jesus by "the men who were holding him" (22:63–65); and a single hearing, in the morning (22:66–71), from which Jesus goes to Pilate. The reconstructed proto-Markan sequence — 14:53a, 54, 66–72, the mockery, the morning hearing at 15:1, the handover — reproduces Luke's order member for member.

What the Mark-Luke comparison establishes is the decomposition claim: that the difference divides as shared core plus Markan surplus, the core proto-Markan and the surplus canonical — against the standing alternative, that Luke abbreviated the trial he read in Mark. Three independent considerations decide between them. First, co-location: the insertion marks in Mark's own text — the resumptive stitch at 14:66, the emptied slot at 15:1 — were identified without reference to Luke, and the boundary of the Mark–Luke difference falls exactly upon them; on the abbreviation account, Mark's text has no reason to bear an insertion's scars at precisely the line of Luke's omissions. Second, the surplus possesses a unifying property of its own, discoverable without Luke and set out in the next section: its members duplicate, item for item, the Roman trial that follows — including the clouds of 14:62, cut not from new cloth but from the gospel's own apocalypse (13:26, whose Lukan parallel at 21:27 places it in the base). Third, economy: abbreviation must find a separate motive for every removal — and can find none for the clouds, which Luke retains without embarrassment at 21:27 — where relocation explains the whole surplus in one operation. And the decomposition faces one witness that did not help draw the line and so can genuinely test it: the Evangelion, whose attested hearing could have contained the witnesses, the verdict, or the clouds for the proto-Lukan layer, and contains none of them (BeDuhn, 2013, p. 189).

The grading, stated plainly: this evidence is of a weaker class than the Great Omission's, for Luke does not lack the material but carries a divergent form of it — similar to what we discussed in the call of the twelve. (Gentile, 2026b, §4.2.2) What converts divergence into evidence here is that the divergence falls along the seam canonical Mark's own text marks.

5. Wendling's argument: the dressing cut from the trial next door

Emil Wendling saw where the added members came from. In Die Entstehung des Marcus-Evangeliums (1908) he argued that the Jewish trial is a mimesis of the Roman trial — replicating its template member for member: the assembled authority, the many accusations, the silence of Jesus, the decisive question ("Are you the Christ?" / "Are you the King of the Jews?"), the answer, the condemnation, the mockery that caps it.⁴ The two-trial symmetry of Mark 14–15 has often been observed; Wendling's point was its direction — the Jewish trial is built upon the Roman one, as an imitation upon its model.

On the relocation model Wendling's observation finds its exact scope: the mimesis is the tailor, not the architect. The relocated hearing did not need creating; it needed dressing for its new climactic position, and the dressing was cut from the trial next door — the accusers, the silence motif (14:60–61a duplicating 15:4–5; Luke's hearing, tellingly, has no silence), the condemnation, the mockery cap. The one substantial addition of content, the temple charge, is the expander's own: a prophecy of the temple's destruction and rebuilding set in the mouths of witnesses he brands false — and planted a second time at the cross (15:29, itself without Lukan parallel), taunt echoing charge, twin deposits of one layer. Even here he composes in his manner, building from what the gospel already holds: the temple's destruction foretold stands in the base at 13:2; the charge garbles the prophecy into a boast, and the narrator disowns the garbling as false witness.

The parent study met this compositional signature at its centerpiece: the expansion contains the second feeding, a mimetic double of the first (Gentile, 2026b, §4.1.2); the expander's demonstrated method is to grow the gospel by doubling its own scenes. Wendling identified the fingerprint in the passion in 1908, without a model to place it in. The present model supplies the hand.

6. From deflection to declaration

What the expansion changed in the hearing's center is the point of greatest consequence for the series. Proto-Mark's Jesus — reconstructing the base form from the Lukan shape, as the relocation argument requires — answers the Christ question by declining it: "If I tell you, you will not believe, and if I ask you, you will not answer" (22:67–68), and then ὑμεῖς λέγετε ὅτι ἐγώ εἰμι — "it is you who say that I am" (22:70). That is the deflecting form, the same register as the σὺ λέγεις that canonical Mark himself retains before Pilate at 15:2. In proto-Mark, then, Jesus answered both tribunals alike, with the reserve — and 22:67–68 is the reserve's own principle spoken aloud at the climax: I will not say it, because you cannot receive it — the logic of the parable theory (4:11–12), in the defendant's mouth, under the direct question. The base gospel did not merely happen to keep the secret through the passion; it staged the question and had Jesus decline it on the reserve's stated grounds.

Note where the two words stand in that answer. The base form contains ἐγώ εἰμι — subordinated, inside the refusal, on the accusers' lips by attribution: you say that I am. The expansion's edit consists in stripping the frame and letting the two words stand alone. Canonical Mark, relocating the hearing to serve as the story's night climax, did not compose a declaration; he released one — deleting the deflection around the words the base had already set down, and attaching Daniel's clouds to what remained. The one open messianic self-affirmation in the gospel was created by the same operation that created the night trial, and created by subtraction.

The finished gospel keeps the seam of that operation on its surface. Mark's Jesus answers the Sanhedrin with ἐγώ εἰμι and Pilate, one scene later, with σὺ λέγεις — a directness and an evasion side by side that commentators have long had to harmonize, by a change of audience, of hour, or of legal register. On the present account there is nothing to harmonize: the two tribunals answered alike in the base, and only one was rewritten. The inconsistency is not an infelicity to be explained but a join to be read — the mark of the layer that changed one answer and left the other standing.

The reserved character of the base gospel is the argument of the third study of this series (Gentile, 2026c); the present candidate, if it stands, is that argument's passion-narrative corollary in its strongest form: the expansion did not merely break the base gospel's silence — it overwrote the base gospel's explicit refusal to break it.

7. The agreements of Matthew and Luke: a contaminated Mark in Matthew's hands

The pericope-pair now yields a cluster of data that any account must face together, and facing them forces the promised refinement. Three times in these adjacent scenes Matthew agrees with Luke against canonical Mark. The taunt-question of the mockery — τίς ἐστιν ὁ παίσας σε, "Who is it that struck you?" — stands verbatim in Matthew (26:68) and Luke (22:64), five words identical, on a verb the New Testament scarcely uses, and is absent from Mark alone. The answer to the Christ question is deflected in Matthew (σὺ εἶπας, 26:64) as in Luke (ὑμεῖς λέγετε, 22:70), against Mark's ἐγώ εἰμι. And Matthew's ἀπ' ἄρτι, "from now on" (26:64), answers Luke's ἀπὸ τοῦ νῦν (22:69), where Mark's text has nothing. One such agreement might be reflex or accident; three across two adjacent scenes want a channel. And the cluster's texture is itself a datum: verbatim at the fixed formula, substance at the answer, synonym at the time-phrase — the gradient of transmission by ear, not the uniformity of transmission by copying, as will matter below.

The channels on offer, and their prices: To lodge the mockery in the sayings source is to give that source a passion narrative — to convert, by fiat and for one datum, the sayings collection characterized statistically in the first study (Gentile, 2026a) into a gospel; the price is declined. To run a line from proto-Luke to Matthew is the proposal of Klinghardt, argued precisely from this class of agreement;⁵ it is compatible with the chronology of this model, but it posits a cross-line dependency the present data will be seen not to require. The third channel costs nothing the model does not already own, once one question is asked: what did Matthew's copy of Mark actually look like?

For the two editions of Mark did not travel in sealed containers. They circulated in the same communities, within a generation of each other, read aloud in the same assemblies — and a copyist of the new edition who knew the old one by ear will let the familiar wording bleed into the new text: harmonization toward the familiar, the best-documented behavior in the whole history of the gospels' transmission, operating in its earliest window, when the familiar text was proto-Mark. Copies of canonical Mark carrying sporadic proto-Markan wording were not a possibility but an inevitability. The refinement, then: Matthew had one copy of Mark — and it was such a copy. Canonical in structure (he follows the night trial, the intercalation, the whole architecture of the new edition), sporadically proto-Markan in phrasing, where his exemplar's line of descent had kept the older wording at the very points the expander had altered: the deleted question still standing, the sharpened answer still soft, the "from now on" still in place. This answers the objection that a two-edition Matthew invites — why not simply use the older, if he had both? — because he had neither choice nor knowledge: one book, already mixed, its mixture invisible to him.

The refinement explains more than it was built for, and the excess is the argument. First, it accounts for the minor agreements' most famous structural property, Streeter's "fundamental fact": Matthew and Luke agree against Mark in wording, never to speak of in order. On this model that is a theorem — order descends from the exemplar (canonical for Matthew, proto for Luke), and contamination, which works by the copyist's ear, moves phrases, not architecture — and moves them along just the gradient the cluster shows: formulas exactly, sense freely. Second, it predicts the agreements' equally famous texture: sporadic, uneven, trivial here and striking there, systematic nowhere — the fingerprint of patchy contamination, and the very character that has made the agreements indigestible to every clean-channel hypothesis. Third, it predicts that the same interference should be visible inside Mark's own manuscript tradition — for Matthew's exemplar was only one member of a whole population of mixed copies — and it is. The witnesses of Mark 14:62 that read σὺ εἶπας ὅτι ἐγώ εἰμι (Θ 565 700 f¹³ 1071) have usually been taken as harmonizing toward Matthew; but read the reading: it is the two editions welded — proto-Mark's deflection and canonical Mark's declaration standing side by side in a single clause, a textbook conflation, the contamination this refinement posits caught in the act within Mark's own tradition.⁶ And with that, a fourth thing falls into place that has stood unexplained in the background of all synoptic study: the text of Mark is the least stable of the three — the gospel of the disputed ending, of the freest early copies, of the one text-type scholars had to invent a name for largely on Mark's evidence alone.⁷ A gospel that existed in two editions, whose copies cross-contaminated for a generation, is a gospel whose transmission would look exactly so. The instability is not noise around the model; it is the model's residue.

The refinement has a pedigree, and it is the oldest in the field: explaining the minor agreements of Matthew and Luke by their access to differing forms of Mark was Holtzmann's original Ur-Markus argument — the founding solution, abandoned when Ur-Markus was, recoverable now that the two editions have been established on grounds independent of the agreements.⁸ And it generates a test, recorded here and left for future work: on this model, the intractable minor agreements should cluster at canonical Mark's editorial interventions, and Mark's own manuscript tradition should show elevated variation at those same loci — twin distributional predictions the machinery of the first study is built to examine. The alternative channel remains discriminable: agreements of Matthew with demonstrably proto-Lukan redaction would require Klinghardt's line; agreements confined, as this cluster is, to plausibly proto-Markan wording do not. The question is left to that test.

8. The mockery: a text that grew in one direction

The mockery attached to the trial (14:65) preserves, in its text and parallels, a small stratigraphy of its own, and every documented motion in it runs the same way.

The scene's fixed element, across the whole tradition, is the taunt-question. It stands in Matthew, in Luke, and — decisively for its antiquity — in the Evangelion, attested for the proto-Lukan layer (BeDuhn, 2013, p.189);⁹ on the argument of the previous section it stood in proto-Mark itself. It is the caption of a known game — the blind-guessing game in which the struck player must name his striker¹⁰ — traditional, proverbial, and everywhere: except canonical Mark, the sole text in the tradition that lacks it.

That absence is deletion, and the relocation supplies the motive. In the custody scene the words are a game; capping the relocated trial, "Prophesy!" absolute becomes a verdict-taunt aimed at the trial's own content — the temple prediction (14:58), the enthronement oracle (14:62) — and it detonates the irony the intercalation was built to stage: at the moment the council mocks him as prophet, his prophecy is being fulfilled by Peter in the courtyard below (14:30; 14:66–72). The game's caption would re-shrink "Prophesy" to the game; the expander cut it, and the trimmed scene — spitting and blows — takes the shape of the Servant of Isaiah 50:6, the composition conformed to scripture in his documented manner.

The blindfold, by contrast, is the scene's unstable element, and the instability is instructive. In Mark the covering clause (περικαλύπτειν αὐτοῦ τὸ πρόσωπον) is omitted by D (Swanson, 1995); in Luke the corresponding clause (περικαλύψαντες αὐτόν) is absent from the Evangelion and from Old Latin witnesses including b (BeDuhn, 2013, p. 189). Where the two canonical traditions do carry the covering, they carry it in disagreeing forms — Mark covering the face, Luke the man — and disagreement of form is the recognized signature of independent supplementation. The most economical account is that no authorial text had the blindfold: the game was proverbial, its prop the sort of thing that gets added, and scribes completed the scene independently in each gospel — as other scribes completed Mark by restoring the deleted question to it,¹¹ as the Byzantine tradition completed Luke by importing the face-clause from Matthew,¹² and as the mixed copies of Mark carried the conflate answer at 14:62. The alternative — that canonical Mark wrote the blindfold and lost it in the West, and in Matthew’s copy, while Luke's tradition lost it separately — requires at least two independent deletions against the current of everything the manuscripts actually show. For every documented scribal motion in this scene, in every gospel, is addition toward the composite. The tradition completes, conflates, and harmonizes at every stage; the only hand ever caught subtracting anything is canonical Mark's — and his subtractions are the ones with redactional motives, made in the course of the operation this supplement argues for.

9. The remaining fit

Two further features of the expanded trial fit the expansion's profile. The move manufactures a verdict: proto-Mark's hearing, like Luke's, condemns no one — the authorities hand Jesus to Pilate without sentence, which is also the procedurally coherent shape.¹³ The night trial gives the Sanhedrin a capital condemnation (κατέκριναν, 14:64), reached at night, on witnesses the narrator brands false and disagreeing, for blasphemy on words that are not blasphemy — the institution convicted out of its own mouth, the expansion's polemic against the Jerusalem authorities at full voltage. And two adjacent riders may belong to the same operation, each separable from the candidate and from the other: "but let the scriptures be fulfilled" with "and they all deserted him and fled" (14:49b–50), extending the anti-circle polemic; and the fleeing young man (14:51–52), whose νεανίσκος stands in noticed relation to the young man at the tomb (16:5). Each should be weighed as its own decision, not carried by the trial's evidence; and the second is so weighed in the third study of this series (Gentile, 2026c), which argues that the fleeing young man belongs to proto-Mark.

10. Grading, and what would count against this

The parent study grades its own criteria (Gentile, 2026b, §2.6): size and the absence of Mark–Luke agreement in wording are each applied with caveats, thematic fit is the most subjective, and the load-bearing pillars are an internal seam in Mark coinciding with Luke either placing the material in a non-Markan location or lacking it altogether. Since the present proposal asks to stand beside that study's identifications, it should be measured against that grading, and its failures stated with its successes.

The two pillars: Both are met. The seam is visible in Mark's own text at 15:1, where a morning consultation follows a night session that has already returned a verdict and the trial interrupts the courtyard scene. And Luke's line places the material in a non-Markan location: where Mark has a night session, proto-Luke's branch carries a morning hearing.

Wording: A caveat is required, as the parent study entered one for the Beelzebul block. Luke appears to know part of the text, but not the expanded text, thus we can not claim there is no agreement in wording, only that there is no agreement in wording in the specific text that is in question. The distinction drawn there applies: agreement at points belonging to material proto-Luke inherited is not agreement with the expansion's framing of it. Here is a sampling of the core agreement: The high priest, Peter followed at a distance, the courtyard, the chief priests, the council, the Christ question, and the Son of Man seated at the right hand of power. The rest, as we have noted, is expansion material.

Size: The candidate fails this criterion, and the parent study has already provided for candidates that cannot satisfy it: when it turned from large blocks to smaller items, it observed that the size criterion could no longer apply, and pressed the pro-Gentile character established for the larger blocks into service as an additional criterion in its place (§2.3). The present candidate stands where those items stand.

Theme: Pressed into service, per §2.3, in place of the size criterion rather than alongside it. All the material in the expansion was found to be pro-Gentile and/or anti-Temple cult, either on its surface or in its reserve. This candidate fits that criterion well.

Where this leaves the candidate: It meets both pillars. It enters a caveat on wording where the parent study entered one before. It fails on size, and succeeds on theme as a substitute criterion. That places it in the company of the Tertiary Fragments — identified by the load-bearing marks, held apart from the blocks that satisfy every criterion, and offered on that footing.

The candidate would be materially weakened if the resumptive κάτω ἐν τῇ αὐλῇ could be shown characteristic of Markan scene-management outside insertion contexts; or if 15:1 could be read naturally as ratification of a verdict already reached — the standard harmonization, which the wording ("held a consultation," not "confirmed their sentence") resists but does not exclude. The refinement it carries — the contaminated exemplar — is exposed to its own twin tests, stated above: it would be strained by hard agreements of Matthew with demonstrably proto-Lukan redaction, which would argue Klinghardt's channel, and it predicts that minor agreements and Markan textual instability should co-locate at the canonical edits.

The candidate is offered, accordingly, as a well-marked tertiary member of the expansion inventory: required in its relocation form by the source-logic of the tree itself; carried by the seam, the emptied slot, the Lukan shape, and Wendling's mimesis; corroborated by the mockery's stratigraphy; productive of one refinement with a Holtzmannian pedigree, a mechanism from documented scribal behavior, and a stated pair of tests; and graded below the centerpiece of the parent study, whose Lukan evidence is of a stronger kind.