Re: Thiede on Magdalen papyrus fragment

From: Carl W. Conrad (
Date: Mon Sep 11 1995 - 11:25:55 EDT

Perhaps this has lost interest by this point, but the question was raised
earlier and I sought permission from Sigrid Peterson and Stuart Pickering
to reproduce their Ioudaios-L posts on the subject to b-greek. I have only
just gotten permission from Stuart Pickering, has been out of touch from
his e-mail, and so I'll go ahead and post this item of his from last March.

He adds:
"Some points in the note will require some modification, but the essential
arguments remain valid.

"I understand that an important note by T.C. Skeat on the connection between
P4 and P64+67 will soon appear, or it may have appeared recently."

From: Stuart Pickering <>
To: Multiple recipients of list <>
Subject: Magdalen College Matthew
X-Comment: First Century Judaism Discussion Forum
Date: Thu, 2 Mar 1995 20:54:26 EST


This note is drawn up in the hope of dispelling some of the
confusion caused by a recent article which proposes a first-century
dating for P64.


1) P64+67.
[van Haelst, Catalogue, no. 336]
Literature includes:
P64. Ed. C.H. Roberts, 'An Early Papyrus of the First Gospel',
Harvard Theological Review 46, 1953, 233-237, pl. (between pp.234-
P67. R. Roca-Puig, Un papiro griego del Evangelio de San Mateo, con
una Note de Colin Roberts, Barcelona, 1962, 3 pll. (showing the
Barcelona fragments).
Recently: C.P. Thiede, 'Papyrus Magdalen Greek 17 (Gregory-Aland
P64). A Reappraisal', ZPE 105, 1995, 13-20, pl. IX.

2) P4.
[van Haelst, Catalogue, no. 403]
This may be from the same papyrus. It is similar in script and
format (two-columned; surviving page dimensions close to
reconstructed dimensions for P64+67). It is from the binding of a
codex of Philo from Coptus.

3) Note that fragments from the binding include:
Unidentified fragments and fragment with heading of Gospel of

4) Thiede gives the impression that the connection of P4 with P64+67
has been universally abandoned. This is not so.

B. TRANSCRIPTION (remarks based on plate in HThR 46, 1953; plate in
ZPE 105, 1995, pl. IX less clear, and photos are enlarged).

1) Verse 32. Spelling of 'Galilee'. Transcribe galeilaian. Supposed
gamma before second lambda (Roberts, Thiede) is an iota followed by
superfluous stroke too low for bar of gamma. The correct reading is
suggested in the apparatus of Roberts' 'Complementary Note' (1962)
(but dot not needed for iota).

2) Verse 22. Transcribe mu not nu, and a punctuation point before
it. Plate supports Roberts' au]tw. m[hti, rules out Thiede's

3) Verse 14. Roberts: [o] legomenos. Thiede: no room for restoring
[o]. A reasonable observation. Omicron is almost always large in
these fragments. But a small omicron in fragment (c) back line 2 (if
I correctly read it), so that the matter is perhaps not quite beyond

4) Verses 22-23. Fragment (c) back line 2. Neither Roberts nor
Thiede can be completely correct. Plate does not support a delta.
Check original here and earlier in line, where the letters are
abraded and unclear from plate.

C. DATING (suggestion: try late [i.e. second half] third century)

1) Roberts correctly associated the script with the style called
biblical uncial or biblical majuscule. The essential question is to
analyse the development of this style and locate P64 within the
Roberts regarded the hand as 'an early predecessor' of the biblical
uncial style (HThR, p.235). We may question whether it was an early
predecessor or whether it belongs within the fuller development of
the style.

2) In his approach to handwriting analysis, Thiede relies on
individual letter forms, without identifying and analysing the
handwriting styles of P64 and of examples he adduces for comparison.
The same mistake of method was made by Y.K. Kim, 'Palaeographical
Dating of P46 to the Later First Century', Biblica 69, 1988, 248-
257, figg. 1-3 (between pp.256 and 257), who failed to assess the
Chester Beatty - Michigan hand in relation to the style called by
Cavallo the Alexandrian majuscule. Both Kim and Thiede have gone
badly astray because of this error of method.

3) Thiede has not adequately re-evaluated the palaeographical basis
for dating P64 to the second, third or fourth century. He cannot
successfully challenge Roberts and others without doing this.

4) The evidence of P4 needs to be taken into account.
Note E.G. Turner's different datings:
P64+67. Typology, p.149: second century.
P4. Typology, p.145: third or third/fourth century.
Palaeographical judgment is affected differently by large and small
amounts of handwriting. There is more of P4 to see.

5) I have not yet sighted all the examples for comparison used by
Thiede. When time and resources permit, it will be useful to discuss
some details of letter forms in relation to the wider development of
Greek handwritings. If Thiede wishes to claim that any of his
examples are precursors of the biblical majuscule, the debate will
enter a new phase.


1) The main conclusion, in relation to Thiede's article, is that a
first-century dating is definitively ruled out by the association of
the hand of P64 with the biblical majuscule style.

2) Roberts' first edition and revised edition were evidently hurried
and left room for improvements (palaeographical, codicological,

3) Thiede shows up some (not all) of the faults in the earlier
editions, but also continues or introduces some errors.

4) In relation to the question of dating, Thiede makes two main
errors of method:
(i) He makes comparisons based on individual letter forms, without
assessing the overall style of the hands in which the letters occur.
(ii) He does not draw relevant evidence into consideration, in
(a) the association of the hand of P64 with the biblical majuscule
(b) the evidence of P4.


1) Check original at various points to verify readings.

2) Clarify the connections between P64+67 and P4 (and associated
fragments). Consider implications of the archaeological context of

3) Analyse the development of the biblical majuscule style and
locate P64 within the development. Draw the hand of P4 into
consideration. Consider codicological and text-critical aspects.
Identify examples of biblical and other papyri in the biblical
majuscule style and establish so far as possible scribal connections
and relative dates.
On the biblical majuscule style cf. G. Cavallo, Ricerche sulla
maiuscola biblica, 2 vols. (Studi e testi di papirologia, 2),
Florence, Le Monnier, 1967 (2nd vol.: 1 case containing 115 plates);
G. Cavallo, and H. Maehler, Greek Bookhands of the Early Byzantine
Period, A.D. 300 - 800 (Institute of Classical Studies, Bulletin
Supplement 47), London, 1987.
The biblical majuscule style - not limited to biblical texts -
begins to emerge in the second century, reaches its definitive form
by the third century, and appears to have been the most common
formal script from the fourth to the eighth centuries. (Cf. Cavallo
and Maehler, Greek Bookhands, introd.)

4) Assess the relevance of the styles of the Herculaneum and Judaean
Desert scripts which have been referred to by Thiede.
For help with Herculaneum hands (including mention of the question
of 'regional particularism') cf. E.G. Turner, Greek Manuscripts of
the Ancient World, 2nd ed., rev. by P.J. Parsons (Institute of
Classical Studies, Bulletin Supplement 46), London, 1987, no. 78

5) A full text-critical analysis of P64+67 and P4 is needed.
On P4 cf. T. Hirunuma, The Papyri Bearing the New Testament Text,
vol. I, Osaka, 1994, pp.18-43, including 7 plates for P4 (pp.20, 24,
26, 28, 32, 36, 40) (commentary in Japanese).

This note is based on an article in preparation, in which
acknowledgments will be made.

Corrections of any errors in this note are welcome. I could amend
this note as necessary and circulate an amended version or - if it
is too long - amended sections.

I thank the Ioudaios discussion list for providing a forum for this

Stuart R. Pickering
School of History, Philosophy and Politics
Macquarie University
Sydney NSW 2109
e-mail: or
end of cited post

Carl W. Conrad
Department of Classics, Washington University
One Brookings Drive, St. Louis, MO, USA 63130
(314) 935-4018 OR

This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:37:26 EDT