Where the Witness Signs Off: John 19:35 and the End of the Fourth Gospel
The UPDV presents the Gospel of John as ending at 19:35. Everything after that verse — the fulfillment citations, the burial, the empty tomb, the resurrection appearances, the Sea of Tiberias epilogue — is excluded as editorial expansion. This is the largest single exclusion in the UPDV's treatment of John, and it requires explanation.
🔗The signed statement
Read John 19:35 as it stands:
"And he who has seen has borne witness, and his witness is true: and he knows that he says true, that you⁺ also may believe."
This is a sworn statement. The witness breaks frame — steps out of the narrative and addresses the reader directly — to say five things: I saw it. I have testified. My testimony is true. I know I am telling the truth. I am telling you so that you may believe. It reads like a deposition. The Peshitta renders this verse with ܐܣܗܕ (ʾashehd, Aphel of ܣܗܕ: "to bear witness") and ܣܗܕܘܬܐ (sahdūtā, "testimony") — the standard vocabulary of sworn testimony.1
After this signed statement, the text changes character. Before 19:35, the Gospel sustains a coherent Christology, a consistent narrative voice, and a clean theological framework. After 19:35, contradictions multiply, competing conclusions appear, and the theological direction fractures. The UPDV stops where the clean text stops.1a
🔗What goes wrong after 19:35
🔗Three competing endings
The text after 19:35 contains not one but three apparent conclusions:
1. 19:35 — the witness's own testimony formula ("that you⁺ also may believe")
2. 20:30–31 — a summary statement ("these are written that you⁺ may believe that Jesus is the Christ")
3. 21:24–25 — a third-party attestation ("this is the disciple who testifies... and we know that his testimony is true")
While the fulfillment citations (19:36–37) and the burial account (19:38–42) use earlier Johannine vocabulary, they serve as the structural bridge into the editorially disturbed chapters 20–21 and fall on the editorial side of the witness's sign-off.
The purpose clause ἵνα πιστεύσητε (hina pisteusēte, "that you may believe") appears in John's Gospel only twice: at 19:35 and 20:31.2 These are the only two places where the author breaks the fourth wall to address readers directly with the purpose of belief. The second looks like it imitates and expands the first.
🔗Christological contradiction
The Gospel of John maintains a sustained agency framework: Jesus acts on behalf of the Father who sent him. "The Father is greater than I" (14:28). "The only true God, and Jesus Christ whom you sent" (17:3). At 20:17, still within this framework, Jesus tells Mary Magdalene: "I am ascending to my Father and your⁺ Father, and my God and your⁺ God."
Then, just eleven verses later, Thomas addresses Jesus as "my Lord and my God" (20:28) — reversing the direction of the same phrase. Throughout every other post-resurrection reference, the risen Jesus is called κύριος (kyrios, "Lord"), the agency title. The single spike to θεός (theos, "God") as a title for Jesus at 20:28 stands alone. This abrupt reversal within the same chapter is difficult to attribute to a single author writing with consistent intent.
🔗The Beloved Disciple footrace
The passage 20:3–10 is widely recognized as an editorial insertion.3 Peter and the Beloved Disciple race to the tomb; the Beloved Disciple arrives first but waits for Peter; Peter enters first; the Beloved Disciple enters and "sees and believes" — but then verse 9 adds "for as yet they did not know the scripture, that he must rise from the dead," and verse 10 sends them home. The passage interrupts the Mary Magdalene narrative (20:1–2 flows naturally to 20:11) and serves primarily to establish the Beloved Disciple's priority over Peter — an editorial concern, not a narrative one.
🔗The Galilee problem
John 21 relocates the action to the Sea of Tiberias in Galilee — roughly eighty miles from Jerusalem — with no explanation for the journey. The disciples are fishing, as though the resurrection appearances of chapter 20 had not happened. Jesus appears, and "none of the disciples dared to ask him, 'Who are you?' — knowing it was the Lord" (21:12). This failure to recognize Jesus is inexplicable if the events of chapter 20 have already occurred.
The Galilee post-resurrection tradition itself is suspect. The UPDV has adopted the retrospective reading of the Aramaic qdm (קדם) fork in Mark's passion predictions: the verb means both "to go before" (spatial) and "previously" (temporal). Mark's Greek went prospective ("he goes before you into Galilee"), while Luke's went retrospective ("remember what he told you while he was still in Galilee"). The UPDV follows Luke's reading.4 Once the "go to Galilee" tradition is discredited as a mistranslation, any narrative that depends on the disciples traveling to Galilee after the resurrection takes a credibility hit — including John 21.
🔗The scholarly tradition
The observation that the text of John went through multiple editorial stages is not new. It is, in fact, one of the most extensively documented findings in New Testament scholarship, supported by over eighty years of detailed analysis.
🔗Multiple editions
Pierson Parker argued in 1956 that the Gospel went through "two distinct editions," the first "devoted almost exclusively to Judea, the second expanded to include narratives about Samaria and Galilee." Parker showed that all of John's Galilean material — chapters 2:1–12, 4, 6, and 21 — shares distinctive features (sudden geographic shifts, re-introductions of characters already known, Synoptic-like parallels) and was added by the same author in a second pass. Chapter 21, he noted, "is plainly an addendum" attached after "the Gospel had been brought to a conclusion at 20:30–31."5
Urban C. von Wahlde extended this analysis to its fullest expression in his three-volume Eerdmans Critical Commentary (2010), identifying three distinct editions of the Gospel — the most detailed compositional study of the Johannine literature ever undertaken by an American scholar.6
🔗The Signs Gospel
Rudolf Bultmann, in his landmark 1941 commentary, identified a pre-Johannine "Signs Source" (Semeia-Quelle) underlying the miracle narratives, along with a separate Passion Source and the work of an "ecclesiastical redactor" who added material after the evangelist.7
Robert Fortna developed Bultmann's hypothesis into a full reconstruction of a pre-Johannine "Gospel of Signs" — a coherent narrative document combining the signs and a passion account, which ended at 20:30–31a: "Now Jesus did many other signs in the sight of his disciples, which are not written in this book. Yet these are written that you may come to believe that Jesus is the Christ, the Son of God." This, Fortna argued, was the original conclusion of the source, later augmented by the evangelist's addition of 20:31b ("and that believing you may have life in his name") and the Thomas episode (20:24–29).3
Significantly, Fortna attributed 19:35 itself to the evangelist rather than the pre-Johannine source: "Certainly the accompanying comment verifying the datum (v. 35) is Johannine."8 On this reading, 19:35 is the moment the evangelist breaks frame to personally vouch for what he witnessed — the author stepping out of the source material to add his own sworn testimony. Whether this verse belongs to the source or to the evangelist, it functions the same way: it is the last point at which an identifiable voice claims direct eyewitness authority.
🔗Two endings, two epilogues
Michael Martin demonstrated in 2006 that the two endings of John (20:30–31 and 21:24–25) conform to the conventions of ancient rhetorical epilogue as described by Aphthonius and other progymnasmata writers. Both endings use the "many other things could be said, but..." formula characteristic of encomiastic conclusion. The presence of two such epilogues, Martin argued, provides rhetorical-critical evidence — independent of redaction criticism — for "an original conclusion and an amended one."9
Harold Attridge of Yale, in a 2020 survey of Johannine scholarship, referred without qualification to "the first ending of the Fourth Gospel (John 20:31)," noting that the textual problem at that verse — whether the verb should be aorist (πιστεύσητε, "come to believe") or present (πιστεύητε, "continue to believe") — reflects the gospel's development through compositional layers.10
🔗Raymond Brown's five stages
Raymond Brown proposed in his Anchor Yale Bible commentary that the Gospel passed through five stages of composition, with a final editor from the Johannine community adding chapter 21 after the existing conclusion at 20:30–31.11 George Beasley-Murray confirmed that 21:24–25 "attest the origin of the Gospel in the Beloved Disciple" and "round off the epilogue and give the book a final conclusion (necessitated through the addition of chap. 21 after the conclusion of 20:30–31)."12
🔗The UPDV's position
The scholars surveyed above identify at least two endings (20:30–31 and 21:24–25), widely attribute chapter 21 to a later hand, and recognize extensive editorial layering throughout chapters 20–21. The UPDV draws the line one step earlier.
The standard view is that the "original" Gospel ended at 20:30–31. The UPDV observes that 20:30–31 itself imitates the witness's actual sign-off at 19:35 — using the same purpose clause, the same direct address to the reader, and the same testimony vocabulary — and concludes that 19:35 is where the recoverable eyewitness testimony ends. What follows is editorial expansion: some of it possibly built on primitive traditions, but too thoroughly reworked by later hands to extract as a coherent block.
🔗The editorial pattern
This conclusion is reinforced by a pattern that has emerged independently across every text the UPDV has examined. Developed theology — explicit Trinitarian formulas, Virgin Mary material, and transferable-entity pneumatology — consistently appears in editorial layers, never in primitive cores:
- The virgin birth narratives (Luke 1–2, Matthew 1–2) — in excluded editorial sections
- The Trinitarian baptismal formula at Matthew 28:19 — in the excluded section after 28:5
- The longer ending of Mark (16:9–20) — excluded
- The Johannine Comma (1 John 5:7–8) — universally recognized interpolation13
- The late chapters of Diognetus inserting Marian theology — excluded
- John 20:22 ("Receive the Holy Spirit" as a transferable entity) — in the disputed section after 19:35
This is an empirical observation, not a dogmatic filter. The UPDV did not set out to find this pattern. It emerged independently across different books, different methodologies, and different research sessions.
🔗Why the material cannot be extracted
There is a structural reason the post-19:35 material resists source-critical extraction. The earliest written Christian tradition — the passion narrative — ends at the empty tomb. Mark, the earliest Gospel, stops at 16:8 with the women fleeing and telling no one.14 There are no written resurrection appearances in the earliest stratum.
This is not because the early church lacked resurrection traditions. Paul, writing around 54 CE15 — decades before any Gospel — recites an oral list of appearances in 1 Corinthians 15:3–8. The Greek verbs he uses, παραλαμβάνω ("I received") and παραδίδωμι ("I delivered"), are the technical terms for transmitting oral tradition in first-century rabbinic schools.16 But Paul's list is just names: Cephas, the Twelve, the five hundred, James, all the apostles, and finally Paul himself. There is no setting, no dialogue, no physical description. The resurrection appearances existed as oral bullet points, not as written narratives.
When the later Gospel writers — Matthew, Luke, and John — reached the end of the written passion tradition, they had a story that stopped at an empty tomb and an oral tradition of who Jesus appeared to afterward. To finish their books, they had to take the bare oral tradition and flesh it out into scenes, supplying dialogue, geography, and theological framing themselves. This explains why the Synoptic resurrection accounts diverge so sharply — Matthew puts the appearance on a mountain in Galilee, Luke on the road to Emmaus and in Jerusalem — with no two agreeing on where or to whom Jesus appeared.
The Johannine community had oral memories of Mary Magdalene at the tomb, of Peter investigating, of a Galilean appearance by the sea. When later redactors composed these into the Gospel, they fused the primitive memory with their own theological concerns: the giving of the Spirit, the Thomas confession, the rivalry between Peter and the Beloved Disciple. The primitive tradition and the theological elaboration were fused at the point of composition, and no source-critical scalpel can separate them.
Siegfried Bergler and Folker Siegert proposed a two-track model of pre-Johannine tradition that accounts for this pattern: a written signs source (Zeichenquelle) and an oral passion narrative (Passionsbericht) that was never committed to writing before the evangelist composed from it.17 On this model, the post-19:35 material is oral tradition being written down for the first time, not a badly redacted written source. What looks like editorial disturbance is actually compositional fusion — the inevitable result of turning oral memory into narrative prose.
The available source-critical models confirm the impasse. Von Wahlde's first-edition reconstruction loses the death scene while retaining the elaborative hundred-pound burial — criteria-driven extraction that produces a narrative hollow.6 Siegert's reconstructed VNT correctly identifies the Thomas story and the upper room commissioning as redactional, yet still retains the hundred-pound burial as source material — showing that even a discriminating source critic cannot reliably separate primitive tradition from theological elaboration in this block.17 Fortna prefaces his own reconstruction of the resurrection section with "I can claim no certainty for the following analysis."3 None of these models produces a cleanly extractable primitive layer, which is exactly what the oral-fusion hypothesis predicts: there is no written source to extract from.
🔗The Acts precedent
The UPDV has excluded the entire book of Acts — twenty-eight chapters, 1,007 verses — on internal evidence: forensic stylometry demonstrated that the Acts author is the same editorial hand that reworked Luke's Gospel.18 The precedent is noted here only to acknowledge that the project has previously relied on internal evidence where manuscript attestation is universal.
🔗What we acknowledge
The UPDV accepts the following costs of the 19:35 position:
1. John had a resurrection narrative. The internal prophecies demand it — "Destroy this temple and in three days I will raise it up" (2:19); "I lay down my life that I may take it again" (10:17–18). The original Gospel almost certainly contained resurrection material. What survives in chapters 20–21, however, is too editorially disturbed to reconstruct as a coherent block.
2. The Mary Magdalene encounter (20:11–18) has genuine primitive markers. Raymond Brown identified three independent forms of this tradition (John 20:14–18, Matthew 28:9–10, Mark 16:9–11), suggesting a tradition older than any single witness.11 The raw Aramaic "Rabboni," the criterion of embarrassment (a woman as first witness),19 and the agency Christology of 20:17 ("my God and your God") all point to early tradition. Brown also judged that the core of 20:1–2 preserves "the earliest form of the empty tomb story found in any Gospel." For the post-resurrection appearance sequence in 20:19–29, Brown adopted Lindars's analysis that the Synoptic disbelief/upbraiding pattern was relocated to the end and reworked as the Thomas episode (20:24–29), with the Spirit-giving at 20:22 filling the vacated position. These are noted as evidence in support of Johannine agency Christology — but not as grounds for surgical extraction from a chapter the UPDV considers unreliable as a whole.
3. John 20:17 confirms agency Christology in the Johannine tradition. The fact that even the editorial material preserves "my God and your God" on Jesus' lips shows that this theology was embedded in the tradition the editors inherited. It validates the framework found throughout the Gospel's core — but it does so from within a passage the UPDV cannot include.
These are real costs. The UPDV accepts them as less damaging than including text that cannot be verified as a coherent eyewitness block. The witness signed off at 19:35. What follows bears the marks of other hands.
🔗Notes
1 The Peshitta uses the Aphel stem of ܣܗܕ (s-h-d): ܐܣܗܕ (ʾashehd, Aphel perfect 3ms), "to testify, bear witness." The noun ܣܗܕܘܬܐ (sahdūtā, "testimony," here with 3ms suffix: ܣܗܕܘܬܗ sāhdūṯeh, "his testimony") is the standard Syriac term for testimony — it renders Greek μαρτυρία across the NT and appears in covenant, legal, and martyrological contexts. The Peshitta makes the straightforward translational choice of the s-h-d root for the Greek μαρτυρ- vocabulary of sworn testimony. See CAL, s.vv. shd (vb.) and shdw; Payne Smith, Thesaurus Syriacus, s.v. ܣܗܕ; Sokoloff, A Syriac Lexicon, s.v.
1a This assessment --- that the Gospel's coherence fractures after 19:35 --- is the article's thesis and is supported by the scholarly tradition surveyed in the following sections.
2 The subjunctive ἵνα ... πιστεύσητε / πιστεύητε ("that you may believe") addressed directly to the reader occurs at 19:35 and 20:31. No other verse in John's Gospel uses this construction to break frame and address the audience with a purpose of belief. See Carson, D. A. "The Purpose of the Fourth Gospel: John 20:31 Reconsidered." Journal of Biblical Literature 106 (1987): 639--51.
3 Fortna, Robert T. The Fourth Gospel and Its Predecessor: From Narrative Source to Present Gospel. Philadelphia: Fortress, 1988. On the resurrection section (20:1--22): "Because of the many difficulties in the chapter, I can claim no certainty for the following analysis" (Section 19, Analysis). Fortna argues that the Johannine evangelist reworked a pre-Johannine source, and his analysis of chapter 20 identifies multiple editorial insertions including the Beloved Disciple material. On the source ending: "For the first time we hear the pre-Johannine writer consciously addressing readers; he or she has written a book, with a clear purpose relative to them" (p. 197). See also Fortna, "Signs/Semeia Source," in The Anchor Yale Bible Dictionary, ed. David Noel Freedman (New York: Doubleday, 1992).
4 See The Gospel of Luke, discussion of Luke 24:6b and the qdm Aramaic fork; also the variant exception entry for Mark 16:7.
5 Parker, Pierson. "Two Editions of John." Journal of Biblical Literature 75, no. 4 (1956): 303--14. Parker's key observation: "The Fourth Gospel thus went through a second edition and that is what produced the structural break at 21:1. ... They were caused by later additions which the author made to his own first draft at the same time that he added chap. 21" (p. 306).
6 Von Wahlde, Urban C. The Gospel and Letters of John. 3 vols. Eerdmans Critical Commentary. Grand Rapids: Eerdmans, 2010.
7 Bultmann, Rudolf. Das Evangelium des Johannes. Kritisch-exegetischer Kommentar über das Neue Testament. Göttingen: Vandenhoeck & Ruprecht, 1941. English translation: The Gospel of John. Philadelphia: Westminster, 1971.
8 Fortna, The Fourth Gospel and Its Predecessor, commentary on 19:31--42 (Section 18, Analysis).
9 Martin, Michael. "A Note on the Two Endings of John." Biblica 87, no. 4 (2006): 523--25. Martin's conclusion: "The obvious implication of the presence of these two Fourth Gospel epilogues is that we have, as Johannine scholars have long argued, two conclusions to the Gospel, the original (which echoes in its claim that the Christ is Jesus the prologue) ... and one that was added later. Hence rhetorical criticism offers new evidence for an old thesis."
10 Attridge, Harold W. "A Textual Problem and a Gospel's Purpose: A Reflection on Current Johannine Studies." Toronto Journal of Theology 36, no. 2 (2020): 205--14. For the compositional layers he cites: Bultmann (1941), Fortna (1970, 1988), von Wahlde (2010), and Siegert (2004).
11 Brown, Raymond E. The Gospel According to John. 2 vols. Anchor Yale Bible 29--29A. Garden City, NY: Doubleday, 1966--1970. For the five-stage composition theory, see vol. 1, pp. xxxiv--xxxix; for the analysis of the empty tomb traditions, the Magdalene christophany, and the Lindars-dependent post-resurrection sequence analysis, see vol. 2.
12 Beasley-Murray, George R. John. 2nd ed. Word Biblical Commentary 36. Nashville: Thomas Nelson, 1999. Commentary on 21:24--25.
13 Metzger, Bruce M. A Textual Commentary on the Greek New Testament. 2nd ed. Stuttgart: Deutsche Bibelgesellschaft, 1994. On 1 John 5:7--8: "the passage is absent from every known Greek manuscript except eight, and these contain the passage in what appears to be a translation from a late recension of the Latin Vulgate."
14 R. T. France, The Gospel of Mark, NIGTC (Grand Rapids: Eerdmans, 2002), commentary on 16:1--8: "in the view of all but a tiny minority of scholars" the authentic text of Mark ends at 16:8.
15 Robertson and Plummer, 1 Corinthians, ICC (Edinburgh: T&T Clark, 1914), "Table of Pauline Chronology" (INTRO.4.6), which places 1 Corinthians at 54--57 CE.
16 TDNT IV, s.v. παραδίδωμι (Delling), §6: the verb is "used as a technical term" for transmission of teaching, with 1 Cor 15:3 explicitly cited and the Hebrew equivalent מסר noted. TDNT IV, s.v. παραλαμβάνω (Delling), §2.b: "As a tt. ... 'to receive in fixed form, in the chain of Christian tradition'" at 1 Cor 15:1, 3, with the Hebrew equivalent קִבֵּל. See also Birger Gerhardsson, Memory and Manuscript: Oral Tradition and Written Transmission in Rabbinic Judaism and Early Christianity (ASNU 22; Uppsala: Almqvist & Wiksells, 1961), 262--323.
17 Siegfried Bergler and Folker Siegert, as discussed in G. Van Belle, "The Signs Source in the Fourth Gospel: A Critical Evaluation of the Semeia Hypothesis in Recent Research (1994--2013)," SNTS Johannine Seminar paper, Szeged 2014. See also Folker Siegert, Das Evangelium des Johannes in seiner ursprünglichen Gestalt: Wiederherstellung und Kommentar (SIJD 7; Göttingen: Vandenhoeck & Ruprecht, 2008).
18 See the forensic stylometry studies in PRs #172 and #174. The methodology and findings are summarized in The Gospel of Luke. The conclusion that the Acts author is the same editorial hand that reworked Luke's Gospel is the UPDV's own stylometric finding, not an external scholarly claim.
19 Meier, John P. A Marginal Jew: Rethinking the Historical Jesus. Vol. 1. New York: Doubleday, 1991. Meier discusses the criterion of embarrassment on pp. 168--71: events or sayings that would have embarrassed or created difficulty for the early church are unlikely to have been invented.