p. 43
THE PARCHMENT
It is not surprising that the skins most often used to make parchment
come from two of the most common domestic animals, namely
cattle and sheep. It will be remembered that parchment came into
regular use for making books in the course of the fourth century.
It is distinguished from leather by the fact that it is not tanned. The
purpose of the manufacturing process was to remove hair from the
outside of the skin and grease from the inside, and to produce a
p. 45
The parchment used for Codex Sinaiticus is of exceptionally high
quality. Most manuscripts contain a number of leaves with holes in
them. These arc either places where the animal had suffered trauma
or the result of damage in the preparation process. In Codex
Sinaiticus there arc only a few, all of small size, and usually in a
margin. (The parchment maker sometimes sewed them up when
the skins were wet, and sometimes they were repaired by somebody
at a later stage, by pasting a thin piece of parchment over the hole.)
The difficulty of identifying veins indicates that the blood was
expertly drained from the carcass after slaughter. The thinness and
quality of the parchment has led to speculation that the manufac-
turers used some otherwise unknown method, or even that they
worked out a lost technique of splitting calf skin into two sheets
(which is believed to be possible only with sheep skin). What is
certain is that the makers were exceptionally skilled, and that
they worked with excellent materials. It is clear that no expense was
spared. It would be too much to say that there was never another
such manuscript in the fourth century. We can only say that there is
no surviving manuscript to compare with it. Other manuscripts
whose parchment shows high manufacturing standards include
Codex Vaticanus and Codex Bezae (which was produced a little later,
in around the year 400).
p. 46
--------------------------- y----------0---J-----o----------------------
The skin of Codex Sinaiticus is very thin indeed. The Conser-
vation Assessment undertaken as part of the Project records the
thickness of each leaf as measured with a caliper at seven places. If
we take the example of the first leaf of Quire 35, we find that the
thickness varies between 100 and 150 micrometres,* with an average
of 116.2. By comparison, the pages of this book are 125 microns thick.
There are a few manuscripts of a similar age which are nearly as fine,
but the parchment of large books was soon to become markedly
thicker. It is remarkable that one of the oldest parchment manu-
scripts should also be of such fine quality. Was the method used to
produce it already in use for some other purpose, or was it especially
developed for parchment production? It is precisely the fineness
of the parchment that makes Codex Sinaiticus possible. Given the
opulent layout, which will be discussed below, one needed thin
material for the Codex to be of manageable size.
p. 48
THE INK
The ink used for writing the Codex was of two colours, brown and
red. The brown ink in which most of the text is written is tannin-
based, containing perhaps oak gall, perhaps bark. Sometimes the use
of an acidic ink has led to serious damage to the parchment of
ancient manuscripts. Here, however, the material has generally
remained stable, except in some of the New Finds and Leipzig leaves
which have been exposed to high moisture levels (water damage
is more likely than humidity as regards the New Finds, given the
desert setting of St Catherine's). The red ink was used for headings
in the Psalms and for chapter numbers and other numbering systems.
p. 53
absence of such evidence, palaeographers have worked out a
process of comparison, moving from the more to the less certain.
The kinds of evidence that help are the following:
first, and most obviously, the content of a manuscript has to post-
date that content's composition. This content may be the text
itself, or a particular edition or presentation of the text;
second, a roll may have had a documentary text with a date in
it written on the back, so it must have been written before that
date; or something may have been written in a mai^in or at the
beginning or end of a codex that serves the same purpose;
thirdly, where there are several scribes involved in the production
of a manuscript, or where there are distinctive corrections, the
evidence from several hands may lead to a more precise dating;
fourthly, any artwork in the manuscript can be dated by an art
historian;
fifthly, the copy may have physical characteristics which are
significant: a paper manuscript is unlikely to pre-date the twelfth
century; a parchment codex will not pre-date the fourth;
sixthly, there may be particular historical circumstances which
provide a possible context for a particular kind of production
From whatever fixed points emerge, and out of a careful compari-
son of every surviving document, a framework has slowly been
developed. Each manuscript contributes a part to the sum of our
knowledge.
Of the six kinds of evidence listed above, the first helps us to date
Codex Sinaiticus. The scribes included in the Gospels a system of
paragraph numbering devised by Eusebius of Caesarea. The precise
date of the system is unknown, but the probable chronology of
Eusebius' life indicates that it cannot have been earlier than about
320. Moreover, the incomplete and unconvincing way in which the
system is presented in Codex Sinaiticus suggests that it had not yet
become well known to scribes.
Dating a manuscript written as long ago as the fourth century
p. 54
cannot be done with the precision that wc would wish. We have to
allow for the general possibility that a scribe may have written in a
particular manner throughout a career which might have lasted, for
all we know, for fifty years, and we must acknowledge that changes
in scripts may not have been adopted and abandoned at the same
time in every place.
The standard work on biblical majuscule dates Codex Sinaiticus
to '360 or a little later'. A date much earlier is ruled out by the ’refined
elegance’ of the script and the decorative elements. A date much
later is ruled out by immediate changes in the script. According to
the same writer. Codex Vaticanus was made in about 350. It is a
theory which owes a lot to a schema according to which biblical
majuscule evolved, achieved a perfect canonical form, and then fell
into decline. This may be too rigid a concept. But the dating of
Codex Sinaiticus to the middle of the century is generally agreed.
Separating manuscripts by a decade at such a remove may be over-
confident. The best wc can say is that the evidence such as it is leads
us to believe that Codex Sinaiticus may have been written shortly
after the middle of the fourth century.
p. 63
SOURCES AND FURTHER READING
The description of the parchment on Q62-F2r was provided by Gavin Moorhead of the British Library Conservation Department, one of the
team who examined each page of the manuscript in intense detail. I also made extensive use of the report by Christopher Clarkson on the New Finds, which he kindly made available to me, and of the report by John
Mumford on the conservation of the London leaves in 'The Codex
Sinaiticus Project: 2. Conservation Work', in G. Fellows-Jensen and
P. Springborg (eds). Care and Conservation of Manuscripts 10: Proceedings of
the Tenth International Seminar held at the University of Copenhagen t9th-2oth
October 2006, Copenhagen: Museum Tusculanum Press. 2008. pp. 153-71.
For the fourth scribe, I made use of Amy Myshrall’s privately circulated
paper ‘The Presence of a Fourth Scribe in Codex Sinaiticus' (October 2006).
The authority on biblical majuscule is G. Cavallo, Ricerchesulla maiuscola
biblica (Studi e testi di papirologia 2), 2 vols. Florence: Le Monnier, 1967.
with 115 plates. The quotation is my translation from page 60.
The correspondence with Jerome is taken from his 1-etter 5.
The papyrus listing the star signs of different professions is quoted in
Peter Parsons' magnificent account of Oxyrhynchus as it is known from its
papyri: City of the Sharp Nosed Fish. Greek Lives in Roman Egypt. London:
Weidenfeld and Nicolson, 2007. p. 187.
For female scribes, sec K. Haines-Eitzen. Guardians of Letters. Literacy,
Power and the Transmitters of Early Chnstuin Literature. New York: Oxford
University Press. 2000, Chapter 2.
Cockerell's comments on the original binding may be found on page 82
of Milne and Skcat's Scribes and Correctors.
The details from Diocletian's prices edict are taken from S. Lauffcr (cd.),
Diokletians Preisedikt (Texte und Kommcntarc 5). Berlin: De Gruyter. 1971.
The calculations on which I have drawn arc found in R. S. Bagnall, Early
Christian Books in Egypt. Princeton and Oxford. 2009. 50-62. I am also
grateful to Professor Bagnall for a private communication on the subject.
p. 121
USE OF THE MANUSCRIPT IN THE EIGHTEENTH
AND EARLY NINETEENTH CENTURIES
At some point, Codex Sinaiticus ceased to be complete. It lost most
of its first thirty-four quires, and some also from the other end, along
with a block of seven halfway through. There is evidence that the
reason for at least some of this loss was the recycling of parchment
in bookbinding. Several of the fragments in St Petersburg were
taken from the bindings of manuscripts in the monastery library.
Some pieces among the New Finds were cut in a shape which
suggests that they were going to be used for the same purpose.
Another whole leaf has folds which indicate that it was used as a dust
p. 122
jacket. Tischendorf was to find a small fragment being used as a
bookmark (this was on his second visit, in 1853, an event to be
described in Chapter Nine). There is also evidence of bookbinding
having been carried on in the monastery. A number of tools for
stamping bindings have been found. Apart from the interest of the
tools themselves (they are very rare), it provides a way of identifying
books from St Catherine's. Unfortunately, it is harder to associate
these tools with binding operations involving pieces of Codex
Sinaiticus, since some of them date from the sixteenth century and
some from the mid-eighteenth.
The evidence from the New Finds sheds further light on these
events. A part of the leaf containing Genesis 23 and 24 was among
the pieces taken by Uspcnski to St Petersburg in the 1840s. A piece
of the previous leaf is among the New Finds. Likewise, a fragment
from Quire 11 is in St Petersburg, while parts of Quires 10 and 12
are among the New Finds. What Uspenski took to St Petersburg,
and what were subsequently found in bindings, were from quires, or
leaves close to quires, which were placed in a room in the monastery,
perhaps shortly after his visit.
====
The strength and
suppleness of
parchment makes
it ideal for use in
reinforcing spines
The binding
The battered state of the last leaves of Hermas is an indication that
the manuscript had been for some time without its covers, and this
theory may be supported by the evidence of the stitching. The
manuscript appears to have had two bindings in its history. The first
was at its creation. The second was in comparatively recent limes.
Separately, the five leaves from Judges among the New Finds have at
some point been 'overcast', stitched together in the way that today
one would join a few sheets with staples down the left edge.
To make sense of all this, wc must leap ahead in the story to 1975.
THE NEW FINDS
The discovery of ‘new’ manuscripts is always exciting, and the
p. 123
or the unsuitable storeroom we do not know). Most significant is the
fact that Tischendorf and Uspenski. on the visits to be chronicled in
the next chapter, were to see materials adjacent to some of the leaves
in the New Finds. This undoubtedly suggests that the materials
found in 1975 and those studied since the nineteenth century were
not completely separated even in the 1840s. It is probably safest to
conclude that the room of the New Finds was not completely
forgotten, but that there was some flow of materials in and out in
the early nineteenth century.
p. 124
SOURCES AND FURTHER READING
The quotation from Milne and Skeat's Scribes and Correctors may be found
on page 81.
I am indebted for information about the Arabic glosses to the report by
Dr Nikolaj Serikoff, Asian Collections Librarian at the Wellcome Library,
and to the comments of my colleague Professor David Thomas.
Donati's diary was edited by G. Lumbroso, Ricerche Alessandrine, in Atti
della Reale Accademia dei Lincei, scrie III. Memorie della dassedi scienze morali.
Vol. 4.1879-1 have not been able to find a copy of this volume. The diary
itself is in the British Library (Ac. 102/3). I have been no more successful
in tracking down Dr Ricci's article, which was published in Revue archeo-
logique 14 (1909). p. 159. My source of information is H. J. M. Milne and
T. C. Skeat, The Codex Sinaiticus and the Codex Alexandrinus, London: British
Museum, 2nd edition. 1955. p. 5. note 1. and Professor Christfried Bttttrich's
account of the history of the manuscript written for the Project. It is to the
latter that I owe the information about Major Macdonald, which is to be
found in a note by S. P. Tregelles, who made an important edition of the
Greek New Testament, in his Additions to T. H. Horne. An Introduction to
the Critical Study and Knowledge of the Holy Scriptures, Vol. 4, 11th edition,
London. 1863, p. 775. It is to Professor Bdttrich also that I owe the quotation
from the letter to Borrow, which may be found in W. I. Knapp, Life, Writings
and Correspondence of George Borrow (1802-1881), Vol. 2, New York and
London: John Murray, 1899. pp. 57-59.
For bookbinding at St Catherine's, see Nicholas Sarris, 'The Discovery
of Original Bookbinding Finishing Tools at the Monastery of Saint
p. 132
We can say, then, that Tischendorf’s account has been taken at
face value by western scholarship for too long, and that there are
serious difficulties with it. Parchment will not burn. Even if it
did, the burning of Codex Sinaiticus is disproved by the discovery
of fragments used in bookbinding and above all by the New Finds
of 1975! Finally, it is important to recognise that no other record of
the events of 1844 has yet been identified, so that his report lacks
corroboration.
p. 175
IMAGING
The technical standards were the highest possible. There was also a
greater challenge, namely that of imposing the same standards in
four different locations so as to achieve a uniform result.
For the technically minded, the selected equipment used at the
British Library consisted of:
PowerPhase and PowerPhase FX digital scanning backs
TGi Filter/ARi Filter
DeVere 480 rostrum camera
Rodenstock Apo-Sironar f9o/ 135mm lens
The PowerPhase FX digital camera back employs a 10,500 x 12,600
pixel CCD chip that enables image captures of 380 MB (24 bit RGB
uninterpolated). Sironar digital lenses were chosen for their greater
depth of field, giving better focus for bound materials, as well as
conservation benefits because no mounting with glass plates or clips
is required. Standard fluorescent cold lighting was used, again with
the welfare of the manuscript in mind. A machine creating a slight
vacuum was used to keep the pages flat while they were being
scanned.
p. 176
CODEX SINAITICUS
The first stage was to make various trials and to assess the results.
We met a number of times in the spring of 2006. The first images
were taken using very strong light, with the result that the writing
from the other side of the page being imaged (or even from the level
beyond that) was almost as visible as the page being photographed.
Lowering the light level and using backing paper solved that
problem. But the type of backing paper affected the colour of the
parchment. It was either too creamy, or too grey, or too white. After
a number of attempts, wc selected a colour which kept something
of the warmth of the manuscript’s own colour, while providing a
contrast with the ink.
THE OTHER LOCATIONS
Conservation and imaging in Leipzig and St Petersburg followed
the same procedures and documentation in conservation of the
manuscript. In fact, all conservation was carried out in the presence
of staff who had been involved in the British Library assessment, to
ensure consistency of practice. Imaging in St Petersburg was carried
out in 2008 by the same photographer (Laurence Pordes). The
Leipzig images, taken by a different team using different equip-
ment, of leaves which had existed under different conditions since
1844, produced rather different-looking results (compare Q34-F8V
and Q35-Fir). Imaging was carried out at St Catherine's in the
summer of 2008 by Michael Phelps, using the monastery's new
Stokes Imaging digital camera and cradle.
Every page was imaged in two ways: ‘normally’, using light
shining onto the surface from a broad angle; and with raking light’,
the light sources being at such an oblique angle that many features
in the surface of the parchment are shown very clearly by their
shadow. The technique is similar to aerial photographs, which often
reveal most about a landscape’s development when they arc taken
early in the morning or close to sunset.
In addition, some difficult pages were selected for imaging using
ultraviolet light. Experiments using multi-spectral imaging (search-
p. 177
ing an area with light from varying points in the spectrum, by which
it is possible to detect ink visible at one point in the spectrum but not
another) were disappointing. Ultraviolet, which was used to good
effect by Milne and Skeat, is sometimes able to reveal text sub-
sequently washed out or erased.
The lilies of the
field correction and
the initial omission
of the ending of
John arc visible
only with ultra-
violet lighting.
