I wish Marco would say a little more about how he sees this Castello card in relation to the topic of this thread. I don't understand.
Also, I had missed reading Ross's article. Thanks for calling attention to it. But again, I don't understand.
I don't understand why someone would invent a game that is made to order for attacks against it, i.e. with just the same number that Bernardino and others had associated with sin and the devil. A cardmaker is in business to sell something people will want and enjoy, and thus if possible something he or they won't get in trouble for having. So why would he design a game that appears diabolical, because of its allegience to the evil number 21? As for "defeating fortune," the only way to defeat fortune, according to the preachers, would be not to play cards. It's a little like re-using the name "Edsel" for a new make of car, but insisting that it is the opposite of the old Edsel, a rather self-defeating move on the part of a manufacturer.
However I can see 21 as a good number in the context of fortune-telling, aimed at those that didn't put too much stock in fire-and-brimstone preachers. The reason is that suggested by Robert Place in The Tarot, History, Symbolism, Divination
, p. 25f (in Google Books). Huson makes a similar point in The Mystical Tarot
, p. 46ff (also in Google Books). There were booklets listing fortunes corresponding to each of the 21 outcomes of the dice. These booklets could then be used in simple sortilege: you draw a card and get a fortune, the one in the booklet corresponding to the number on the card. If you get the Fool you draw again. There were similar books for 56 outcomes. There were also correspondences between the 23 letters of the alphabet and 23 fortunes. The preachers probably got some of their ideas from looking at the booklets they were burning, adapting them freely (I have never heard of an alphabet with 21 letters; the one in use then had 23, as far as I know).
On this view, the tarot would probably not have had 21 plus 1 trumps at the beginning, precisely so as to avoid the association with evil dice. Especially when first in production for the masses, it needed to be as pious as possible, so as to avoid the condemnation of the Church: hence all the religious imagery, suggesting a ladder to heaven rather than hell. Later, once the decks were established, cardmakers would have seen the appeal of a deck with additional possibilities for fortune-telling, and the deck with 21 numbered trumps would be the one to thrive.
Gambling and divination would have been seen as intrinsically connected. Classical tradition held that dice-like things were used in fortune-telling before they were used in gambling. Robert Graves makes this point in his Greek Myths.
A source in Google Books is Harvard Studies in Classical Philology
Vol. 83, p. 9 (I got to it with the search words "Apollo priestesses pebbles." There are many other such sources.) Multi-colored pebbles were shaken in water by the priestesses of Apollo. And when soldiers cast lots--not only to divide the spoils, but to see who would go on a particularly dangerous mission--they were thinking that in that way the gods would signify their will. I suspect that this point was known in Renaissance Italy. The humanists read the same books as Graves etc. Apollo invented divination, and Hermes debased it as gambling, as the Homeric Hymn to Hermes hints. So for e.g. the superstitious prostitutes who used the tarot as a come-on with erudite clients (the context in Aretino), divination with tarot would be a way of discerning God's will, the "divine" in "divination."
This divinatory use of tarot is not documented until Venice 1527 (if Place p. 25 is correct); before that--1507 Gianfrancesco Pico, as Ross discovered, and c. 1505 Strassburg, per http://trionfi.com/0/p/41/
--it is just "playing cards." But I would think that the practice started even earlier. Trionfi's documentation of a 1485 German oracle book with fortunes related to 51 or 52 animals looks suspiciously like it might have been adapted from cards, especially since the fortunes resemble those in the later oracle books for cards.
But the production of a deck with 22 trumps might originally have had nothing to do with dice or fortune-telling. It might have been the result of an effort to standardize the cards among cities, what with Milan having a card others didn't (Papess) and others having cards that Milan didn't (Devil, Tower). 20, or 20 plus the Fool, is a rounder number than 22, more consistent with the "Mantegna" and the fashionable Christian Pythagoreanism of the times (which imagined the descent and ascent of the soul through 10 spheres). Then dice, perhaps along with a tantalizing suggestion that the cards were the letters of God's language, Hebrew, might have made 22 as the number to survive.