The page pattern

The pattern in displacement hypotheses

The apparent inconsistencies in order in the text have led a variety of scholars to propose that a few passages have been displaced from their original positions.
Many have commented on the fact that the supposedly displaced passages are mostly (if not all) close to multiples of 800 Greek letters in length.
Let's consider the five supposedly displaced passages mentioned (but not taken as such) by Barrett. [1] These passages are (with the number of Greek letters [2] in parentheses) 3:22-30 (740), 6:1-71 (5663), 7:15-24 (765), 10:19-29 (776) and 15-16 (4784).
As is evident from these examples, the unit is approximately 800 Greek letters.

A possible mechanism for such displacements

If the current order of the text is illogical, it is difficult to see why anyone would have composed it this way, so it is reasonable to suppose that it was once in a more logical order. But it is equally difficult to see why anyone would have deliberately dislocated it.
Could the displacements have come about accidentally?
Clearly a completed scroll does not lend itself to accidental displacements. Nor does a codex, for every sheet except the central one contains material separated by at least four pages. [3] Therefore the episodes must have been written on separate sheets, probably (as on a scroll) using only the recto of each sheet. [4]
Furthermore if the displaced sheets are to make grammatical sense in their new context, each episode must have been wholly contained on its set of sheets. This may have been a somewhat unusual procedure, [5] but it did have the benefit that the author would have been able to write the episodes in any order.

The pattern in the pericopes


The same pattern is clearly evident in many of the pericopes.
For example, the table below indicates logical sections defined by the UBS editors which are also close to whole multiples of 800 letters, which unit is here taken to represent a page.

Pericope Length Pages Topic
1:43-51 802 1 The calling of Philip and Nathanael
2:13-22 818 1 The cleansing of the Temple
4:1-42 3255 4 Jesus and the woman of Samaria
8:12-20 769 1 Jesus the light of the world
8:21-30 811 1 Where I am going you cannot come
13:1-20 1621 2 Washing the disciples' feet
13:21-30 727 1 Jesus foretells his betrayal
20:1-10 781 1 The resurrection of Jesus
20:11-18 784 1 The appearance of Jesus to Mary Magdalene

Summary of page pattern evidence

The evidence from the sizes of proposed dislocated passages can be combined with the pericope evidence. We can also add those resulting gaps which are close to multiples of 800 letters.

Pericope Length Pages Topic
1:43-51 802 1 The calling of Philip and Nathanael
2:13-22 818 1 The cleansing of the Temple
3:22-30 740 1 ** dislocated passage ? **
4:1-42 3255 4 Jesus and the woman of Samaria
6:1-71 5663 7 ** dislocated passage ? **
7:15-24 765 1 ** dislocated passage ? **
8:12-20 769 1 Jesus the light of the world
8:21-30 811 1 Where I am going you cannot come
8:31-10:18 7200 9 ** gap **
10:19-29 776 1 ** dislocated passage ? **
13:1-20 1621 2 Washing the disciples' feet
13:21-30 727 1 Jesus foretells his betrayal
13:31-14:31 3147 4 ** gap **
15-16 4784 6 ** dislocated passage ? **
20:1-10 781 1 The resurrection of Jesus
20:11-18 784 1 The appearance of Jesus to Mary Magdalene

Are these observations significant?

At this stage a "section close to 800 letters" has been taken to be any section which is within 100 letters of a whole multiple of 800. In a random structure, 1 in 4 sections would be expected to satisfy this criterion. The table shows 16 sections, representing about half of what the majority of commentators take to have been the original gospel, i.e. chapters 1-20.

Bear in mind that UBS may not have correctly identified all the pericopes which the Evangelist or the Redactor had in mind. Consider also that there may have been interpolations by a later editor who had no interest in the standardization of pericope sizes. (Evidence for this is presented on the next page of this site.) In the light of these considerations the pattern observed does appear to be significant. But this can only be confirmed if we can make sense of the whole gospel on the page hypothesis.

Is there any evidence of this pattern in the extant papyri ?

Not to my knowledge.
But the absence of papyrus evidence is not surprising. For the very earliest extant gospel fragment, p52, which happens to be from John's gospel (containing 18:31-33,37-38), had been part of a codex. So the use of separate sheets for copies of the gospel had already been abandoned by that time (ca. 175 CE). Indeed it had probably already been abandoned by the time our current text, the Third Edition, was defined. For as we shall see on the next page of this site, some of the later interpolations do not conform to the earlier page size convention. [6]

Notes for this page

1. C.K.Barrett, The Gospel according to St. John(London; SPCK, 2nd. edn., 1978) p. 23
2. All letter counts are based on the UBS text: K.Aland et al.,   The Greek New Testament (New York; United Bible Societies, 3rd. edn., 1975), ignoring 7:53-8:11 which is within double square brackets and was recognized by the editors not to have been part of the original text.
3. At such an early stage in the history of codices, it is fairly safe to assume that we would be talking about single quire codices.
4. According to Haenchen, such separate sheets could have been written with a view to making them up into a scroll: E.Haenchen, A Commentary on the Gospel of John, 1 : Chapters 1-6 (ET: Hermeneia; Philadelphia; Fortress, 1984) p.50
5. Such a procedure was suggested in Feine and Behm's Einleitung in das Neue Testament, quoted in: R.Schnackenburg, The Gospel According to St. John, Vol. ONE (ET: London; Burns & Oates, 1968) p.54
6. The change of format from separate sheets to a codex was almost certainly influenced by the format of the synoptic gospels which, I have reason to believe, existed from their creation in codex form.