jmd wrote:ONE suggestion, that I made earlier, is also quite mundane and practical: after woodblocking the sheet, it may have been observed that the stencils (plural - one for each colour) had been cut for the Emperor as mirror image. IF such was the situation, it would have been simpler to whitewash the one card (still uncut) and re-apply a new small individual woodblock to the area with the image mirrored to what we now have, and then be able to apply the stencils.
I'm afraid this scenario is not logical.
The stencils were made from extra prints. The black key line image was used as a reference, and holes were cut with a knife. Then it was varnished for durability. It didn't take much effort to make the stencil, and it is virtually impossible to cut one card "backwards" on a stencil since you are cutting on the printed image itself.
Re-cutting a woodblock in reverse to accommodate a backwards stencil is far, far harder than re cutting a paper stencil.
Also, you cannot easily "whitewash" ink that has been printed onto cardstock. And you would not re-print a single card in the midst of a page of other cards. If for some strange reason you wanted to re-print a card, you would make a new block with that card on it, possibly repeated several times, and just print replacement pages of emperors.
Kaplan's set of Dotti woodblocks are notable here:
Five of the blocks each contain 15 cards.
The sixth block has three card images repeated five times. (5 coins, ace swords, ace batons.)
You need to make one print of this block for every five of the others.
Thus also demonstrating a case in which decks were NOT assembled from isolated, intact printed sheets.
jmd wrote:
What the normal 'ghosting' explanation does not do is explain how only one card is so clearly marked with such precision.
If the card was in the middle of a row of five, and the sheets neatly stacked, it easily explains the precision.
As far as it being the only one in the deck....
If anything, I think this whole episode suggests that decks at the Noblet house were likely assembled from stacks of each card on a table, and not from individual sets of intact sheets. This is actually quite a sensible way to manage the variety of defects that would occur. If one or two random cards got ruined, the usable ones from the sheet could still be used. At the end of the run, you would print out only the blocks that contained cards that were missing from the table, and so make up all of your replacements at once.
The other ghosted cards were most probably either thrown away, or landed in other deck(s)....