Lake discusses in particular Scribe D as follows:
"Reserving, however, out of respect to the opinion of so distinguished a palaeographer, the possibility that Barnabas is by a different hand, it is tolerably clear that Scribe A originally wrote all the text of the NT except Hermas, which was the work of Scribe B, and that Scribe D wrote the text on the conjugate leaves, ff. of 10, 15, 29 & 30, 88, 91, and possibly on part of folio 126. Specimens of these three scripts, A B D are arranged side by side on Plate III.
There is possibly room for legitimate doubt whether Tischendorf was right in distinguishing Scribe A from Scribe B, but personally I entirely accept his judgement, for after the prolonged acquaintance with the style of Scribe A, necessitated by photographing each page, I felt while watching the script 'come up' on the negative in the developing tray, that the first page of Hermas was different from the others, as it seemed to 'come up' differently, though from the nature of the case I did not know until afterwards which this particular plate was.
The same thing was still more noticeable in the case of the Scribe D plates. It would be too much to claim that this purely personal experience ought to weigh strongly in the judgement of others, and I admit both that I am unable to analyse satisfactorily the difference between Scribe A and B, and that it is not so clear to my own perception now as it was when I was spending the greater part of each day in the company of the MS.
The discrimination of Scribe D from Scribes A/B is easier and admits of no reasonable doubt. There is a distinct difference in the script, though it is more easily perceived than described; possibly the letters are somewhat squarer in Scribe D than in A - the height being less in proportion to the breadth - and Scribe D is altogether prettier than A. But the decisive point is that Scribe D constantly fills out the end of a line with the sign >, which is rarely or never used by Scribes A/B. A specimen column of Scribe D is third on Plate III.
Scribe D = Vaticanus?
It was Scribe D whom Tischendorf identified with the scribe of the NT in Codex Vaticanus; (see the 4th column on Plate III), and it will probably be at once conceded by those who compare this with Scribe D that there is no real trace of justification for Tischendorf's theory. The wonder is that the fine eye, which saw the difference between A, B, D, - differences which anyone might be excused for overlooking - could ever think for a moment that the script of Scribe D was identical with Vaticanus.
'Cancel-Sheets' added in the Scriptorium
The conjugate leaves written by Scribe D are clearly 'cancel-leaves'; that is to say, they were written after the MS had been completed, in order to take the place of others, written originally by Scribe A, which were for some reason imperfect or spoiled. Such replacing of rejected leaves would naturally form part of the διορθωσις (correction) of the MS in the scriptorium, and that this was the case is rendered practically certain by the fact that Scribe D actually wrote the whole of Tobit & Judith in the OT, so that he was clearly a member of the scriptorium. The importance of this point is that it shows that any work done on the MS before the 'cancel-leaves' were added must also be regarded as work done in the scriptorium too, and it is convenient at this point to indicate the details of which this can be proved:
(1) The Eusebian Apparatus must have been added before the cancel-leaves in Matthew (folios 10/15), as these leaves, and these only, lack the Sections and Canons. Thus the scribe who added the Eusebian Apparatus belonged to the scriptorium.
(2) The Stixoi: Similar reasoning shows that the scribe who added the στιχοι in the Epistles belonged to the scriptorium, for, after the Epistle to Romans, these are only omitted in 1st Thess., the last page of which is one of the cancel-leaves (folio 88).
- Kirsopp Lake, Codex Sinaiticus Introd. (1911), p. xviii fwd
Aside from establishing that Scribe D was one of the original scribes who created the codex, and that his work must have taken place in the original scriptorium (Caesarea) where they worked, Lake gives good evidence to show that Scribe D was likely a διορθωσης, or Overseeing Corrector of the manuscript.
Major Renovations to Manuscript:
Nonetheless, while minor corrections, even insertions of a few lines in the margins are expected, large-scale replacement of whole sets of pages was apparently as rare as it plainly is drastic. This was not normally required, even when making an important manuscript for a powerful client, like a wealthy patron or emperor.
What was going on, that required such extreme editing of what should have been a normal copying process? The answer seems to be that while the main scribe, Scribe A was in fact doing what he was supposed to, namely faithfully copying his master-copy, this was not satisfactory to the Overseer, who then took over the job at key points, one of them being the Ending of Mark.
Here it seems obvious, and all textual critics seem to be in agreement, that Scribe D removed several pages from Scribe A's original work, and began replacing that text actually a few pages prior to the actual ending of Mark.
Why start so far back? One of the reasons seems to be that Scribe A had already raced ahead so far (and had already begun copying Luke), that simply taking over and continuing at the expected place was not adequate.
Scribe D seems to have decided that the only way to 'fix' the text was to re-copy several previous pages. He did this in a way that is as unusual as it is transparent, to a trained eye. He has spread out the last few pages, with more space, paragraphs and shortened lines, so that the text written spreads out to fill the pages, the number of which is basically fixed by the Quire layout (each Quire made up of 4 double-folios folded in two, giving 8 folios, or 16 pages per Quire).
To remove a text by removing permanently whole pieces of vellum would seem to require deleting 4 pages at a time. Alternatively, he could have trimmed out a single folio, to remove 2 pages of text. Scribe D obviously felt that was not an option, because it would disrupt the standard way the book was to be bound, and the amount of text being adjusted was not large enough to justify folio removal or insertion of extra folios (the folio count seems to have remained the same).
Instead Scribe D has apparently removed one folded double-folio at least, and replaced it, and re-written the text to spread the modified text to more or less fill the space without too much defacement of the style and layout. If the text in dispute was in fact Mark 16:9-20, or part of it (we'll get to that possibility later), this would only involve a column and a half or so, and might explain why the text needed to be stretched somewhat to minimize the blank space at the end between Mark and Luke.
Mark 16:9-20 Apparently Can't Fit:
Some who have studied the strange changes in letter-counts for the replacement columns have suspected that the original text could not have contained the traditional ending of Mark. But this may be beside the point.
It was likely that the number of folios and quires needed were pre-calculated, and standard 'tricks' were used to keep copying on track, such as pre-designed layouts or charts, keywords for column and page heads etc. If the person who calculated and alloted the quires for the work had assumed that Mark's ending was to be left off, then a problem could have arisen near the point where Mark's ending was approached.
If Scribe A was using a copy of Mark that had the full ending, he may have gotten to a certain point and suddenly noticed there would not be enough space.
Scribe A would then have consulted his superior, the διορθωσις, likely Scribe D, who probably was responsible for calculating the number of Quires and folios needed.
Scribe D now took over to rectify the situation, first removing a few pages of Scribe A's work, and then attempting on the fly to spread the text according to the original plan, possibly while still using the exemplar of Scribe A.
The actual features of this part of the text of Sinaiticus will be discussed in more detail later. What we can come away with for now is a general idea of what might have taken place to account for the replaced sheets, the need for re-writing the ending of Mark, and the basic reason for the switch between copyists.