Ross wrote,

First I'll have to disabuse you of the notion that all these near-complete trionfi decks were circulating, and somehow ended up coalescing, remarkably containing all the subjects of the near-complete decks, as if the inventor toured Italy, collecting this deck of 14 here, this deck of 16 there, this deck of 20 here, and then mixed it all together into a super-Tarot, and then re-flooded Italy with various different orderings of this synthetic set of trumps. That seems to be the tough part, despite all the hoops and invented scenarios necessary to come up with to account for this "late standard" trump series.

The simple and only necessary answer - lost cards. All decks with any standard trumps are incomplete, in both suit and trump sequences. There is no need, and certainly no benefit, to posit multiple unknown chains of evolution. The analogy with the missing exemplar(s) that you cite for textual transmission is, in Tarot history, the 22-standard itself. The decks are missing pieces of this standard, that's all.

You've oversimplified my concept of the possible "evolution" of the deck, When someone copies a precious manuscript of a work he's never seen before, he doesn't compare it to all other available versions of the same text. He just copies, with errors, the one he's lucky enough to get. Someone else might have copied the same manuscript, with a different set of errors. But if the one copied from were lost, scholars wouldn't say that the later copy was copied from the earlier one, just because of the numerous commonalities between them.

Similarly, if two works of art a few years apart seem to be different elaborations of the same core design, art historians don't just say that the one later in time is a variation on the first, but rather that each might be a variation on a simpler work of art now lost.

Likewise, when someone is designing a deck based on a previous deck, he doesn't compare it with other decks. He modifies what he has in front of him in accordance with his own ideas. He changes the order, changes the concept of a card, adds one or more, etc. I am imagining different 14-trump decks expanding to 16 in Milan, but that expansion not being popular. Someone else tries a 20 trump deck. Someone else later modifies that into a 22 trump deck. These latter two are all in one line of transmission. There can be many such lines, radiating out from various points, but downward only, meaning later in time.

If a text based on one manuscript gets popular, people want to see other versions of that text. They even travel great distnces to get them, such as Switzerland, France, or Greece. If such are available, at that point manuscripts are produced that reflect a variety of manuscript traditions.

So there are two different types of manuscript transmission, from a single manuscript when a text is not popular--let us call it type 1--and from many manuscripts when a text is seen as important and others can be obtained--call it type 2.

However the analogy with manuscripts breaks down here (although not with artworks). With manuscripts, the assumption is that there was an original one that can be reconstructed by comparing all the various copies and eliminating each one's errors. But with games, and inventions generally, that is not usually the case. Person A can copy person B's innovations by incorporating them within person A's existing pattern so as to make something that is not a closer approximation of anything that pre-existed both. For example, if inventor A has an automobile that looks and works great except for a ridiculously fuel-inefficient, large, and malfunction-prone engine, he can incorporate inventor B's marvelous engine and no one will think he has reconstructed something that existed in the past but was lost.

So we might have the situation that one particular type 1 modification of the original deck becomes in demand: in 1452 or so if Milan, perhaps earlier if elsewhere. That is one that expands the 14 to either 20 or 22. The basis for thinking that such might have happened is that certain cards not present in the Cary-Yale (assuming that 2 virtues, Time, and Fortune are missing) are invariably in the same relative place in the order in both the A and C orders (Popess, Pope, Hanged Man, Devil, Arrow, Star, Moon, Sun). That this is the case is accounted for by its being a type 2 transmission (like putting B's engine into A's car). I say "20 or 22" because the Devil and Arrow are not present in any of the PMB-type decks. O'Neill calculated the odds of that happening by chance. The math suggests, although not definitively, that it is not by chance. To be sure, it could have been by choice.

Then, once there are 22 trumps, further regional changes occur, of type 1, giving us the three main types and many differences within them, also type 1. These Dummett has accounted for. There are also a few more type 2 changes (i.e. influences from another region), most notably in Piedmont and Sicily. (The governor of Sicily who introduced the game was most familiar with Roman Minchiate, but spent the 2 previous years as governor of Lombardy.)

The question then arises, are 14, 16, 20, and 25 viable trump numbers for a trick-taking game with five suits? I include 25 because that is one theory about the number of cards in the Cary-Yale.
That is where the rest of what you said in your post is relevant. You say:

With these principles accepted, you can see that the game does not work symmetrically for any other number of players. 3 or less gives too many cards to each player, while 5 or more too few.

Something is wrong. There is the Regle of 1637, which says that there should be three players at most, and that when two play, there is a "dead" hand. Yet there are 22 trumps here, the same as in Bologna.

Mais auant que d'entrer plus auant dans le destail de ce jeu, il me semble a propos de dire qu'il n'y faut estre que trois personnes au plus, & qu'il n'est pas fort agreable a deux, estant mesme encore necessaire d'y en supposer vn troisiesme que l'on appelle le Mort, duquel l'on tire selon le hazard autant de cartes que les autres font de mains pour estre emportées par celuy qui est le plus fort. Neantmoins ceux qui l'ayment extresmement s'y peuuent quelquefois diuertir de la sorte.

(But before going further into the details of this game, it seems appropriate to say that there should be three people at most, and it is not agreeable for two, it being then necessary to assume a third called the Dead, which is dealt by chance as many cards as the other hands, to be taken by the person who is the strongest. However those who love it extremely can sometimes be entertained so.)

There is also Galeazzo Maria's letter home from Ferrara in 1457, where he says he played "cards and tennis" with his host Francesco Pico della Mirandola (Lubkin, p. 309, http://books.google.com/books?id=NUoR2W ... en&f=false). No mention of others playing with him, and Lubkin (p. 92, http://books.google.com/books?id=NUoR2W ... ot&f=false) gives this note as a comment on his life-long passion for tarot,

Your argument for four persons and 22 trumps also assumes the designer thought in terms of the "best case" and "worst case" scenarios. I didn't follow all of your argument, because you talk of the possibility of being dealt all the "counting cards"--19 or 20--and then you switch to the possibility of being dealt all the trumps, i.e. 21. Which is it they have to have, 19, 20, or 21 cards? I will assume the latter, since it seems most desirable to win all the tricks, and that is what you have the dealer getting.

In the Regle, there in fact does seem to be an effort to give every player (of 3) at least 21 cards. It suggests shortening the deck. Immediately after my previous quote, it says:

Mais afin de le trouuer plus agreable il est bon d'oster douze cartes inutiles des quatre peintures, c'est a dire trois de chacunes, sçauoir les dix, neuf, & huict des couppes & deniers, & les trois, deux & az d'espées & bastons qui sont les moindres de chacun de ces points, par ce que les hautes de couppes & deniers ne sont pas de plus grande valeur que les basses des Espées & bastons.

(But in order to find it more pleasant, it is good to remove twelve unnecessary cards of the four colors, i.e. three each, i.e. ten, nine, eight cups and coins, and three, two and ace swords and batons, which are the smallest of each of these colors, for the high ones of cups and coins are not more valuable than the low of swords and batons.)

I'm not sure if this is meant as a requirement of the game or not, but a practice of removing 12 cards, for 3 players, is consistent with your point; now, instead of just the dealer, every player has the possibility, however remote, of getting all the trumps.

But I don't see why it had to be part of the original design, as opposed to something thought of later (since the probability is so low). Also, without knowing how many cards were used in a tarot deck in Ferrara 1456 (70 or 78) with four players, we don't know how many cards each player got then. If it was a 70 card deck, as suggested by the Ferrara note of 1457, each person would get 17 except the dealer with 19. You can't be sure of getting all the tricks with that, unless there are only 20 trumps, including the Fool.

While the idea of getting 21 cards in a game with 21 trumps is a nice symmetry, and one indeed held onto, I don't see why it should have been part of the game from the beginning--it is not part of your seven principles--and it is especially troublesome in 1456 Ferrara, given the 70 card tarot pack note of 1457.

The lack of a "verzicole" rule in Bologna would have definitely been suggestive of some sort of priority, I'm not sure what. But it was there. And the symmetry with 21, kept even in 1637, is indeed interesting, assuming a 78 card pack with 4 players and 66 with three. I just don't know what can justifiably be made of it, regarding the "original designer". So for the present I'm back to not being able to give priority to anywhere, or to say that there were more than 20 trumps in 1456.

mikeh wrote: You've oversimplified my concept of the possible "evolution" of the deck, When someone copies a precious manuscript of a work he's never seen before, he doesn't compare it to all other available versions of the same text. He just copies, with errors, the one he's lucky enough to get. Someone else might have copied the same manuscript, with a different set of errors. But if the one copied from were lost, scholars wouldn't say that the later copy was copied from the earlier one, just because of the numerous commonalities between them.

Similarly, if two works of art a few years apart seem to be different elaborations of the same core design, art historians don't just say that the one later in time is a variation on the first, but rather that each might be a variation on a simpler work of art now lost.

Why must it be "simpler"? The point is that the Tarot trump sequence is not simply a painting, but a sequence, a hierarchy, of very specific subjects. Every instance of this sequence that survives contains standard subjects, in a fragmentary form (no deck is complete, Sola Busca is not standard). It is impossible that the same subjects would appear in all of these fragmentary survivals unless, when complete, they exhibited the full standard sequence, and only the standard sequence. Random designers in different locations don't choose from the same limited set of subjects, very precise subjects, unless they are basing themselves on a model.

It is probably better to put the ball back into your court - what does lost cards not explain?

What it does not explain is one thing only - the presence of three extra trumps - the Theological Virtues - in the Cary Yale deck.

The reasons the Cary Yale give rise to speculations about evolutionary models can be reduced to two things: its date and its place of origin.

There was a long tradition, invented by Cicognara, who introduced the deck to the world, that the deck was made in 1428 for the marriage of Filippo Maria Visconti and Maria of Savoy. This fiction persisted long enough and was well enough established that even Michael Dummett seriously entertained it. Before a more accurate dating, in the mid-1440s, became the consensus, and before Dummett systematized the evidence and laid the foundation - as well as building most of the edifice - for the study of Tarot history, this dating seemed as good as any other. But we know better, and we know it is not so early, nor the earliest evidence of the game, which is documentary, and has no relation to Milan.

For Milan as the deck's home, the importance of the Cary Yale becomes related to Milan's claim as the place of origin for the game itself. That is, if the game originated in the Visconti court, and Cary Yale is the earliest example of the game there, then it seems reasonable to suppose that the Cary Yale represents the form of the game in the court that invented it. It was an expanded form of the normal deck, with 2 extra female court cards in each suit, and the trumps included the standard set as well as the Theological Virtues. Later this proto-Tarot was shortened, in both suits and trumps, and became like the Visconti-Sforza, also in Milan. After this, the deck went public in the Visconti-Sforza standard form. This was Dummett's theory (or one of them), and it was consistent with the evidence, or a reasonable reading of it (since the connection between the courts of Ferrara and Milan was close enough to assume the transmission of the game one way or the other), arguably until the late 1980s, when Franco Pratesi found the first documentary reference to the game in Florence, as well as evidence of a cheap form of the game in Marcello's text, both dating to before the time that the standard based on the Visconti-Sforza was supposed to have circulated.

Putting the Cary Yale in its place, then, demands that we recognize it as a luxury commission, expanded upon the standard model, rather than a witness to a different kind of standard in a localized court that existed for an indefinite period of time before 1440, when trionfi cards are already attested in Florence.

The fact that it is, emphatically, an expanded deck, both in the suits and trumps, as well as in physical size, makes this argument more probable than the subtraction model. The fact that it dates after the existence of the game in Florence, as well as, probably, in Ferrara (not to mention Burdochio's deck, wherever he got it from, Bologna or Florence) - since "1441" is based exclusively on the "marriage-deck" legend started by Cicognara and not on any other considerations, and goes against the consensus of art historians like Sandrina Bandera who date it to 1443 at the earliest - means that it cannot be considered as a survival of an earlier standard, unless we continue to hold to the theory that the game was invented in the Milanese court earlier than the evidence outside of Milan.

As you know, I hold that the A family, probably in Florence, has a much greater claim to be the original type of game. But this is beside the point - the only question that lost cards as the explanation for the fragmentary state of surviving trump series does not answer is why the Cary Yale has those extra cards. You need a lot of presuppositions, a theory of Milanese invention and exclusive Milanese possession for some time, to believe that it represents an earlier evolutionary form of the game. The simpler explanation, fully consistent with the evidence and chronology, is that it is an expanded luxury commission, a unique revisioning of the whole deck. In this view, it is itself therefore indirect evidence of the existence of the standard sequence of 56 suit cards and 22 trumps.

The 70 card note in Ferrara is not direct evidence of another standard either, since we don't know how it was composed. Of the three solutions proposed, Huck's of proof for 14 trumps, mine for clerical error, and Michael Hurst's for a shortened deck (two pips removed from each suit), based on the overwhelming trend of the evidence, I favor my explanation, but only slightly more than Hurst's. In any case, it will be useless to argue further since no minds will change on this one. Unless someone new comes in with an open mind and wants the evidence and arguments aired, everybody here already knows what they think.

So that's it - lost cards leaves one piece of evidence in a dubious position, depending on your theory of ultimate origin of the game. Otherwise, it explains everything. It is demonstrably true, in that all surviving Tarot decks are fragmentary. It does not demand the invention of scenarios of greater or lesser complexity of unknowns involving numbers of trumps, and improbable survival of whole trump sequences. It simply says - all the surviving decks were standard, and they have all lost cards in the suits and trumps. The Cary Yale is an expanded deck - also demonstrably true. The absence of the Devil and Lightning cards from all copies of the Visconti-Sforza model is best explained as a choice of decorum, I agree. The Cary Yale itself, along with the presumably absent Devil and Lightning, show that the commissioners and players of these cards were more concerned with appearance than the best mechanics of play that would have exercised the mind of the game's inventor.

Ross wrote,

It is probably better to put the ball back into your court - what does lost cards not explain?

There are many things that "lost cards" does not explain. The only question is, do they need explaining?

For example, the 13 cards of Ferrara 1422, the 14 figures of Ferrara 1441. These are explained by "evolution", and are consistent with one of the explanation for the 70 cards of 1457. But do they need explaining in relation to tarot at all? The 13 cards have have another obvious explanation--replacements; the 70 cards have other reasonable ones, which Ross mentions; I don't know of any as good for the 14 figures. But perhaps if we knew more, we'd have one. This is not much by itself, I have given one argument in previous posts, based on what is variable vs invariable in the three orders, suggesting what I call "type 1" and "type 2" transmissions. I will move on to other reasons, grounded in, yes, the incomplete Cary-Yale deck, plus, later, the Borromeo fresco.

But perhaps I should first discuss the dating of the Cary-Yale, because that is, indeed, the main basis, in my mind, for the "evolution" hypothesis. The 1443-1445 dating by the experts is based on the coins, that they are imprints or at least copies of actual coins not made before 1443. I have yet to see any examples of such coins. The "rearing horse" was used by the Visconti even in the 14th century (viewtopic.php?f=11&t=917&p=13797&hilit=rearing#p13797). And the lettering on the sides does not correspond to any known coins, as Marco showed (viewtopic.php?f=11&t=917&p=13797&hilit=rearing#p13807). Moreover, it is uncertain who the artist was; it does not have to have been Bonifacio Bembo.

There are other reasons for thinking that it is a deck commemorating a Visconti marriage or betrothal. We have gone over all this before. The fountain on the man's chest of the Love card suggests Sforza. The apparent Sforza stemmi in the suits of Swords and Batons, symbols of weapons, and Visconti stemmi in Coins and Cups, symbols of wealth and love/spirit, suggest the Visconti-Sforza marriage (or betrothal) in particular. There is the long-standing Visconti tradition of commemorating marriages with manuscript miniatures, now extended to another medium. The pose on the Love card is similar to another Visconti-Sforza marriage commemoration, 1464 (http://1.bp.blogspot.com/-2iyICN4VZ_I/U ... e+1441.jpg). The lady on the chariot drawn by white horses, carrying an object much like a overlarge Visconti coin, led by a groom much like the man on the Love card; Bianca Maria was noted for riding a white horse (I don't remember where I read that description) and was rich and chaste.

That the two banners above the couple on the Love are those of Visconti's marriage rather than Sforza's is explained as a commemoration of the previous marriage, thus a reaffirmation of the alliance between houses. Also, the Cary-Yale minus the Sforza stemmi might have been a Visconti tarot deck format now lost. Given the subject of the card, a union between sexes, an interpretation of the viper as Milan and the cross as Pavia, both Visconti stemmi, would not fit, as implying an incestuous marriage, unless someone knows of some other prominent marriage with these stemmi. Such a lost deck, or deck type, is not a mere "logical possibility" or "Russell's teapot", as the card itself (in the context of known Visconti marriage commemorations), the Marziano deck of much earlier, and Filippo's known commissioning of commemorative miniatures in the late 1420's (documented by Kirsch) all make it reasonable enough,

Also, given that we have no information about the composition of triumph decks before the Cary-Yale and the Brera-Brambilla, both pre-1447, to a certain extent it doesn't matter when, before 1447, the Cary-Yale was done. We have no better information about tarot decks in Florence in 1440 or Ferrara in 1442, or Bologna whenever in the 15th century--unless, of course, you assume the point at issue, which is that decks before 1447 in a particular city had the same composition as decks after 1447, a supposition we know is false in the one example we have, Milan.

Now I want to give a couple of new arguments--at least new to me--pertaining to Milan and 16 vs. 25 trumps. The Borromeo fresco shows 5 people playing triumphs. With 16 trumps, 80 cards, each player will be dealt 16, enough theoretically to be dealt all the trumps. With 25 trumps, each player will be dealt 17 except the dealer who gets 21. That's not enough, Only if each player gets 16 except the dealer with 25 will there be enough. But that's an absurd advantage for the dealer. If there are 14 trumps and 14 in each suit (as I imagine the Brera-Brambilla), and 5 players, each player will get 14 cards, enough for them to be all trumps. The symmetry is preserved. With 13, the same thing. This is an argument I just thought of today, applying the principle you enunciated earlier with 4 people to the Borromeo's 5.

There is also the ideological component, when considered in the context of a trick-taking game. This is another argument I just thought of. It would seem that the game was intended early on for children and their parents or guardians. But the full 22 have the Devil, the Arrow, and the Hanged Man (assuming it is negative) trumping the virtues and a lot else. As I argued in my last post, that is a rather Macchiavellian perspective. It would be difficult for a child to grasp (unless the Hanged Man is one's own virtuous ancestor), and unwise as well, before he or she had grasped proper Christian morality. Also, the Pope and the Popess, who are beaten by the virtues, are too low, unless you wish to teach that the pope and clergy are corrupt. The full 22 (with these five cards and the Fool) befits an older, more sophisticated generation and involves a second layer of thought, as I explained in my last post. A 16 trump game, however, is perfectly understandable from an orthodox Christian perspective; the theologicals beat the cardinals, with charity (St. Paul's "the greatest of these is love") on top; and the later Petrarchan triumphs beat the earlier. Every other trump beats the Emperor and the Empress, in part because they are of dubious morality, but more because they represent the child himself or herself at the beginning of his or her journey, which his more advanced self trumps over. The same logic applies to a 14 trump game, simply removing 2 of the cards (Prudence and Fortune come to mind first, but Empress and Emperor are possible) or smaller numbers of figures.

"Evolution" explains (a) the transmission pattern for the order of cards, comparing A and C; (b) why 25 doesn't fit 5 players in Milan, but 16, and in general the same number of trumps as suit cards, does; and (c) why the 5 trumps I mentioned don't fit, in trick-taking, the orthodox Christian ideology of the trump sequence. I think these are things that do need explaining. It also, in passing, explains the "14 figures" of Ferrara 1441.

Moreover, the "evolution" theory does not have to posit anything as the "original tarot". It only posits a preceding form, more or less. What there was before that, if anything, is deemed beyond rational speculation.

I don't adopt any of this as established, or even a "working hyothesis", as there is a vast amount I, and we, don't know. I just see no way of deciding among certain alternatives.

mikeh wrote:

Moreover, the "evolution" theory does not have to posit anything as the "original tarot". It only posits a preceding form, more or less. What there was before that, if anything, is deemed beyond rational speculation.

Then the question has to be asked, why would you reject a theory that works - original standard-lost cards - for one that doesn't - unknowable stuff-parts of standard are actually complete-coalescence into standard?

What do you gain by this preference, except a mess? The evolutionary theory explains less, so much less that you are forced to accept things that "are beyond rational speculation". So, instead of solving apparent anomalies like the Cary Yale through the greater explanatory power of the standard model - it is an expanded variation of the standard imagery, like the later Minchiate, just as the Sola Busca and Boiardo (and the Animal and genre scenes Tarots of later centuries) are iconographic variations of the standard number - you throw it aside for a less explanatory theory that not only forces you to make up ad hoc theories for every instance, but also forces you to invent unknowable prior conditions. The theory's unwieldiness defeats itself, by your own admission.

It's as if someone rejected Kepler's elliptical orbits because they didn't explain why they are elliptical, so perhaps we should just toss it all and go back to Ptolemy and epicycles (the 1422 "13 cards" and 1441 "14 figure", for instance, which are irrelevancies). If Newton had thought this way, he never would have seen gravity.

It's just a perverse choice, preferring the less explanatory just because. Perhaps because it's more exciting to come up with ad hoc stories, and the history is interesting to read and dream up scenarios in any case.

The only rational choice for the theorist and the researcher who uses theory to guide him is the one with vastly greater explanatory power, in the light of which everything can be rationally explained and doesn't demand additional unknowns piled on unknowns, all depending on ultimate unknowables. This theory is the "standard model" - an original standard 78 card deck, of which whole sequences of popular versions, and fragments of custom, luxury versions, have survived the centuries. It's simple, it's fully explanatory, and the few conjectures it requires don't beggar belief.

I don't adopt any of this as established, or even a "working hyothesis", as there is a vast amount I, and we, don't know. I just see no way of deciding among certain alternatives.

There is indeed a vast amount we don't know, but the amazing thing is that what we do know is sufficient to build a coherent and explanatory theory that not only accounts for the knowns, but can also successfully predict unknowns (for instance my 2007 prediciton, based on the standard model, that if any discovery of trionfi earlier than 1442 is found, it will be within 3 to 5 years before. Nothing had been found earlier for 133 years - since Campori in 1874 - so it was a long shot to think that anything might, but lo and behold, only 3 years later Thierry found Newbigin's Giusti, and it was as predicted - within 2 years, 1440. I didn't pull "3 to 5 years" out of my ass, or because of some a priori rule of thumb - it was from the pattern of the data of the standard model that I made my deduction and placed my bet. This is what I mean by predictive power - testable, falsifiable theories, rare in history because we depend on accidents of preservation, not repeatability. We should treasure such methods where we can make use of them. Another prediction - if any trionfi cards are found before 1440, or if any new ones from the first 5 years or so are found at all (say 1439-1443), they will be of standard subjects, and not even the Theological Virtues like in the Cary Yale. The latter is a unique design, probably the only one like it ever made).

mikeh wrote: But perhaps I should first discuss the dating of the Cary-Yale, because that is, indeed, the main basis, in my mind, for the "evolution" hypothesis.

Now I want to give a couple of new arguments--at least new to me--pertaining to Milan and 16 vs. 25 trumps. The Borromeo fresco shows 5 people playing triumphs. With 16 trumps, 80 cards, each player will be dealt 16, enough theoretically to be dealt all the trumps. With 25 trumps, each player will be dealt 17 except the dealer who gets 21. That's not enough, Only if each player gets 16 except the dealer with 25 will there be enough. But that's an absurd advantage for the dealer. If there are 14 trumps and 14 in each suit (as I imagine the Brera-Brambilla), and 5 players, each player will get 14 cards, enough for them to be all trumps. The symmetry is preserved. With 13, the same thing. This is an argument I just thought of today, applying the principle you enunciated earlier with 4 people to the Borromeo's 5.

For an explanation of the number of trumps in the Cary Yale being 24 (25 with Fool), I think Dummett's the most elegant. Namely, the person who designed the deck reasoned that since the ratio of trumps to a suit in the standard deck was 14:21 or 2:3, then by adding two court cards to the suit, for 16, means that the trumps should have three more, for 24 (plus Fool), thus 16:24, or 2:3. For the subjects, there was a ready-made self-contained group available, the three Theological Virtues. Insert this group, and you have your number. Note an additional elegance, namely that the total number of trumps would then match the total number of counting courts, 24.

I can believe that the designer was thinking of a larger group of players than the standard number, four, and that this is what made him choose to enlarge the deck. The fact that the additional courts are female, and that the Borromeo shows a courtly game with mixed men and women, prompts the idea that this unique production was conceived precisely for this sort of milieu, and is not based on a popular deck with the same kinds of additional subjects. If the Devil and Lightning were missing from this deck for reasons of decorum, which is a reasonable conjecture given their absence in all Visconti and Sforza decks, then they may have been replaced by other subjects, such as the famous Ship card.
http://pre-gebelin.blogspot.fr/2007/11/ ... liano.html
If this is so, then it is a remarkable coincidence that the Sicilian Tarot chose the same subject to replace the Devil two centuries later - unless the designer, moving in courtly circles, had seen such a card in a Milanese luxury deck. It might even be that the lightning was removed and only the tower, unmolested, remained in the Cary Yale and similar decks, then, as this is what happened in Sicily.

I wouldn't think that the Cary Yale designer re-imagined the game, the rules of play, since it seems unlikely he had "five" in mind as some kind of perfect number. He just knew that in courtly gatherings usually more than four people played cards, mixed groups of men and women, so had to come up with a larger deck, but not one impossibly huge like the Minchiate, which demanded a complete rethinking of the game, and most importantly, presumed numbering the now vastly expanded set of trumps. Nobody was expected to memorize the Minchiate sequence, and it wasn't necessary. But the three Theological Virtues are a self-contained group, no burden on the mind, and all that had to be done was to agree on a hierarchy among them, which was probably the same as the Biblical order, Faith, Hope and Charity. There was nothing "clever" in it, like in Minchiate. It was a light-hearted pastime for mixed groups, as the Borromeo fresco shows (in the context of other games they played together, see their in situ arrangement:


http://www.rosscaldwell.com/images/borromeo/sala1.jpg

Note that there is no reason to believe that they are playing with a deck like the Cary Yale, or even trionfi, but both speculations seem harmless, so they are not likely to mislead, and the latter in particular seems plausible given the size of the cards.

Ross G. R. Caldwell wrote:

mikeh wrote:

Moreover, the "evolution" theory does not have to posit anything as the "original tarot". It only posits a preceding form, more or less. What there was before that, if anything, is deemed beyond rational speculation.

Then the question has to be asked, why would you reject a theory that works - original standard-lost cards - for one that doesn't - unknowable stuff-parts of standard are actually complete-coalescence into standard?

Not "unknowable stuff-parts" but "unknown decks". It seems to me that it is not to that extent wrong to defend a theory that does not depend on unknown facts (about the non-existence of prior decks), as opposed to one that does (namely, on the supposition that there were none such). The theory you espouse depends on unknown, and probably unknowable, assumptions about the non-existence of previous decks. One of the alternatives I espouse, the "evolutionist", has no such dependency. How that is a drawback I have no idea.

Both the "evolutionist" and "creationist" (if I may so dub yours, meaning dependence on a single genius in one act of creation) theories "work" in the sense of fitting the known facts. I do not reject your theory; I simply say that the facts do not justify either its adoption or rejection. That is my agnosticism. Neither of the two is like the "flat earth" theory, for example, which we can safely reject. Or even the geocentric cosmos theory. Or so I have tried to argue.

You say that there is much evidence that the tarot was invented within 3-5 years of 1442, based on your graph. I have dealt with that graph elsewhere (viewtopic.php?f=11&t=939&p=13781&hilit=chart#p13781), but I will try to repeat what I said in briefer terms. If I have left out something, see my other post. Science assumes that the past is like the present and the future. There is nothing wrong with that. But we cannot assume that the years before 1442 are like the years after 1442, regarding the invention of the game of tarot, because by definition we are dealing with something new, something that was not the same in the past as in the future, counting from a specific date as the present. To make predictions regarding past and future about something new, an invention, we have to study other inventions, and records of inventions, and generalize from these, if possible. You have not done any of this, Ross. Without it, however, your predictions derive from the wrong sort of data, a future that by definition is unlike its past. If your prediction turns out to be right over the long term, it does not confirm your theory. Such predictions about data are routinely explained in other ways, such as the known destruction of evidence in key places, and the lack of documentation generally for new inventions in the medieval period/early modern period, for 15 years or more (sometimes, in fact, hundreds of years). The correctness of your prediction may derive from things other than your theory.

Ross wrote,,

It's as if someone rejected Kepler's elliptical orbits because they didn't explain why they are elliptical, so perhaps we should just toss it all and go back to Ptolemy and epicycles (the 1422 "13 cards" and 1441 "14 figure", for instance, which are irrelevancies). If Newton had thought this way, he never would have seen gravity.

It's just a perverse choice, preferring the less explanatory just because. Perhaps because it's more exciting to come up with ad hoc stories, and the history is interesting to read and dream up scenarios in any case.

Let's go over this again. Kepler's facts are the various observations of planetary motions, chiefly those of Tycho Brahe. The hypothesis is that the planets move in ellipses, but at varying speeds, depending on where they are in their orbit.

What is parallel in our case is, for facts, the existence of lists of 22 trumps, starting in the late 15th century, lists that parallel the subjects of extant cards from the 15th century and later. I do not challenge these facts, of course. I only say that the explanation you offer for them is one possibility, and "evolution" is another. Calling the latter "irrelevant" is begging the question.

People after Kepler did challenge his hypothesis. The planets do not move precisely in elliptical orbits. Their orbits are affected by the positions of the other planets in the solar system. So Newton postulated a mysterious force called "gravity" which seemed to account for these deviations, as well as for Kepler's "three laws", and which also predicted hitherto unknown planets. Newton's theory had more explanatory power. It was also vastly more complex than Kepler's, and in fact required him to invent a whole new branch of mathematics, (For a comparison of the complexity, see http://en.wikipedia.org/wiki/Kepler%27s ... ary_motion.) To Galileo, the idea of such a hidden force, from the heavens as well as the earth, constituted occultism and he rejected it on that basis. His theory of the tides, of course, was a mess. Fortunately Newton was also an alchemist (you can still read his translation of the Emerald Tablet) and had no problem with occult forces.

In our case, I would say you are in the position of Kepler, yes. But does that make more complex theories more like Ptolemy or more like Newton? It is in part a question of explanatory power, which I have already addressed and will address further. Also, my theory, like Newton's in relation to Kepler and unlike Ptolemy's, includes yours as one subset (one "or" clause). There is also the problem that history is not like the solar system: decks of cards do not stay pretty much the same over the spans of human time. We have to think in terms of other disciplines besides physics and astronomy. Most obviously, there is Darwin and the "origin of species" Darwin had to take an idea more complex than that of Genesis seriously long before he had enough evidence to present it. The theory of manuscript transmission also offers helpful analogies, as I have suggested..There is also the history of inventions.All of these have their limitations.

You mention an abundance of data, allowing you to make predictions. it is true that we have a lot more data now than when Dummett wrote Wicked Pack. But it is the wrong sort of data. If it were data about the history of inventions, that would be relevant. If it were data about light-hearted games for mixed company, probably also including parents and their children, among the nobility of Milan, Ferrara, Bologna, and Florence in the period 1425-1447, that would be relevant, if only in a negative way, to reduce the probability of ithe game's having arisen in one or more of these cities. Data after 1450 is less relevant, and gambling games (where the main object is to win money or humiliate people) less so than social games. What I mean is data that mentions many other games but seems to exclude tarot. Even then, certain reasonable possibilities will probably not be excluded, for example that Filippo Maria had the game invented and made as a gift to his wife and her relatives, and/or to play with his mistress and daughter. The difficulty in finding such data, and in what it would show, is one reason why I tend to find discussions about explanations, as opposed to interpretations, of tarot in general unfruitful.

There is to be sure also the question of which theory about the tarot has greater explanatory power. I gave several ways in which the "evolutionist" theory explains more, namely the various notes with numbers attached to them before 1460, the variations and constants among the various orders, suggestive of what I called "type 1" and "type 2" transmissions, and the fact of the Cary-Yale. You only discussed the last of these.

Here I would first note that in the latter instance even Dummett did not think that the "evolutionist" hypothesis should be rejected (Game of Tarot p. 430l):.

It is possible that the Visconti di Modrone pack was no more than a freak, and that what was later the standard composition of the Tarot pack was standard from the time of its first invention. But it is also possible that the Visconti di Modrone pack represents the original form of the Tarot pack, and that the 78-card pack as we know it is the result of a modification adopted early in its history. If so, the standard set of twenty-one trumps must itself be the slightly mutilated remnant of the original, and possibly larger,set. In that case, we could not expect any ordering of the trumps in the standard set to make perfect sense; even if there was any particular symbolic intention underlying the original sequence of Tarot trumps which there may not have been, we could expect fully to understand it only if we knew which subjects the original set contained and in what order they were arranged. It is unlikely that we ever shall.

Some may interpret this as a reducio ad absurdum, but it seems to me unflinching logic. Dummett imagines here the pack's evolving from a larger to a smaller number of trumps--although he does say "possibly larger", In this particular I do think there are other reasonable alternatives; that only increases the difficulties. While we may well never know what subjects the "original" had or their order, I think we can nevertheless make reasonable hypotheses regarding what preceded (in lists) and filled out (in decks) the facts as we have them. Your "creationist" hypothesis is one; the "evolutionist" is another. Dummett says that we might in this case (the "evolutionist" one) never know fully the original order or even what the orignal subjects were; he is thinking that the Popess might not have been original, but Prudence was. More problems. So I tend to agree with his last two sentences. All we probably will be able to do is to interpret the sequence, and its possible antecedents, in various historically appropriate ways, and not understand fully how it came to be.

You say that the hypothesis of 25 trumps in the Cary-Yale is accounted for by the "elegance" of a 3:2 ratio of trumps to cards per suit, and how the number of trumps (excluding the Fool) equal the number of courts. These are to be sure "elegancies", but there are other such "elegances", for 16 trumps Having the same number of trumps as cards per suit is also "elegant". In addition, 16 has the further "elegance" that the number of cards dealt, in a 5 person game, is the same as the number of trumps total; that is a principle that is found in the 3 person game of 1637 and perhaps also (as you point out) in the 4 person game of 1446, and so may have been a very early feature..There is also the additional "elegance" that the Cary-Yale trumps as I have proposed feature cards depicting 2 towers, 2 horses, 2 old men, and 2 royals, plus 8 virtues (if love is considered a virtue, as it seems to be in its Cary-Yale depiction), exactly parallel to the 4 pairs and 8 pawns of chess.

There is also the question of the Devil and Arrow cards inelegantly not fitting the philosophy of the sequence as a whole, in virtue of their trumping power, i.e. overpowering, not only some very virtuous cards, but also numerous "counting cards" (and even more, if sequences count). You suggest that the Devil and Arrow cards might for that reason have been replaced by a Ship and a Tower not struck by lightning, as in Sicily (even though the numerous extant partial PMB-type decks have no Devil or Arrow but also no cards that might have replaced them). But my point had to do with the initial plan. Supposing all 22 were created at once, why would these two subjects be there in the first place? For them to be able to trump all the virtues, and authorities, and many counting cards, individually and in combination, is a contradiction to the orthodox Christianity otherwise permeating the deck. You so far have no explanation.

My reason for thinking up or endorsing such elegancies and inelegancies is not that they are exciting or visionary. It is rather to show that there are many historically appropriate forms of elegance, and to fix on one of them is merely to fall in love with one's own image.

That said, it still seems to me, reading over what I have written, that the "evolutionary" hypothesis has the greater explanatory power, despite its greater complexity and number of blank spaces and clauses connected by "or". But because of them I also don't think that the greater explanatory power is much to brag about.

More on "counting cards". Dummett, Game of Tarot Ch. 20 p. 419f:

Minchiate was, of course, a deliberate invention, a conscious variation upon existing Tarocco games; but we know that in both Florence and Rome the 78-card pack continued to be in use down to the seventeenth century. Our best clue to the nature of such games consists in those played in Sicily; the Sicilian game was probably first imported from Rome, and the traditions of Tarot play in Rome seem to have been strongly linked with those of Florence. If this hypothesis is sound, then what Minchiate shared with the game played in Florence and Rome with the 78-card pack was not the versicole but the fact that, in addition to the Bagatto and the Fool, all five top trumps, the arie, were counting cards.

So it appears that at the top end of the trumps there were five counting cards in Florence and Rome, just as in de Gebelin. For me the significance is that a group is thereby defined, solely from the rules of play, although perhaps driven by association in meaning. De Gebelin, to be sure, had five counting cards at the other end, too, whereas Dummett only has two. But Dummett goes on to say:

In Sicilian Tarocchi, and therefore, on our hypothesis, in the 78-card game as formerly played in Florence and Rome, there are seven, the five top trumps, the Bagatto and the Fool; and in Minchiate there are twenty-one, the eleven top trumps, the five lowest ones, four others and the Fool.

There remains a question in my mind whether in Florence and Rome with the 78 card pack the five lowest trumps might have been counting cards as well, expanded in Minchiate to 9 plus the Fool on one end and 11 on the other. These five would make another group, again independent of the meanings of the cards but perhaps driven by association in meaning.

For Minchiate the scheme simply follows the numbers.

The first 5 mirror the last 5 (1-5 and 36-35) - Papi and Arie

The next top 5 (31-35) - Rossi

10-20-30, cause this in harmony with 40 - interesting numbers, if you part the trumps in 4 parts

13 (13th card from bottom) and 28 (13th card from top) - interesting numbers, if you part the trumps in 3 parts

0 Fool

*******

2-5 have 3 points
36-40 have 10 points
all other counting trumps (1-10-13-20-28-30-31-32-33-34-35) have 5 points
The Matto and all 4 Kings have 5 points

Well, it might have been an interest, that the half (= 1/2) of the trumps (20) has points, and that 24 of 96 (= 1/4) of the cards (without Fool) and the Fool have points.

*************

So it's easy to learn. But ....

Also we have, that the summary 12+50+55+25 = 142, which is roughly a 3/2 relation to 97 (97*3/2 = 145.5), If one would exclude the Fool and his 5 points, the summary 12+50+55+20 would be 137 and the relation would be (96*3/2 = 144), so worse.

But if one assumes an earlier state of the game, in which the 4 papi (2-4) also counted 5 points instead of the degraded 3 points, one would have without Fool the summary 20+50+55+20 = 145 and this in some better number harmony with 144 and 145.5

So .... Counting the Fool with a 1/2 point and the 4 Papi with 5 instead of 3 points would give precisely 145.5.

This looks like the "original counting rule". Somebody found it necessary, that the Fool robbed a few points from the 4 degraded Papi.

The "number elegance" suffered.

************

In the discussion of the Rosenwald Tarocchi ...

Franco Pratesi: ROSENWALD’S FOURTH SHEET
http://trionfi.com/rosenwald-tarocchi-sheet

... there was the idea, that once a 96 card game existed (in which Fool and Pagat were merged to one figure). For this the total number of points for trumps in the above described manner would have been 145, and the 3/2 ratio to 96 would have been 144.

"145" as sum of the points would have had then the advantage, that a draw ("unentschieden") was impossible.

In the early Fool discussion it likely played a role, that "zero" wasn't in much use (the Roman number system still was preferred against the invasion of Arabic numbers. With 15th century the use of zero advanced... the Sola Busca Tarocchi (1497) knew a zero on the matto. For the trumps Roman numbers were used. The suit cards used Arabic numbers.


see the small Arabic 3


Roman numbers on trumps


Mato with "0"

*******************

The Rosenwald Tarocchi has a Fool/Magician with a Roman number "I"



Inside the 5x14-theory it's assumed that the Fool presented the number 11 in the work of the first painter of PMB

Inside the Chess theory it's assumed, that the Fool in the (likely Florentine) Charles VI tarocchi appeared without Pagat ... similar to the suspicion about the (likely Florentine) Fool/Pagat in the Rosenwald Tarocchi

Inside the Mantegna Tarocchi the beggar got the Roman number "I"

Inside the Hofämterspiel the 4 Fools got the Roman number "I"

**************

Perhaps one should persecute the first use of Arabic numbers in art.

The European acceptance of the numerals was accelerated by the invention of the printing press, and they became widely known during the 15th century. Early evidence of their use in Britain includes: an equal hour horary quadrant from 1396,[22] in England, a 1445 inscription on the tower of Heathfield Church, Sussex; a 1448 inscription on a wooden lych-gate of Bray Church, Berkshire; and a 1487 inscription on the belfry door at Piddletrenthide church, Dorset; and in Scotland a 1470 inscription on the tomb of the first Earl of Huntly in Elgin Cathedral. (See G.F. Hill, The Development of Arabic Numerals in Europe for more examples.) In central Europe, the King of Hungary Ladislaus the Posthumous, started the use of Arabic numerals, which appear for the first time in a royal document of 1456.[23] By the mid-16th century, they were in common use in most of Europe.[24] Roman numerals remained in use mostly for the notation of Anno Domini years, and for numbers on clockfaces. Sometimes, Roman numerals are still used for enumeration of lists (as an alternative to alphabetical enumeration), for sequential volumes, to differentiate monarchs or family members with the same first names, and (in lower case) to number pages in prefatory material in books.

http://en.wikipedia.org/wiki/Arabic_numerals


These are often the same names, so likely not a sign of a very large distribution
http://de.wikipedia.org/wiki/Indische_Zahlschrift

From Georges Ifrah (specialist on number development I get, that "zero" appeared first in 1491 (printing date; De Arithmetica opusculum by Philippi Calandri ... according a text made for Lorenzo de Medici before)


http://commons.wikimedia.org/wiki/Tratt ... uselang=de
as PDF download here ...
http://www.bsb-muenchen-digital.de/~web ... nav=0&l=de

.... , just in the year, which is given to the Sola Busca. The same year 1491 is given by Ifrah to the French change of the use of the word chiffre, which before meant "zero" and since then got the modern meaning. As original names for zero Ifrah gives "Sifra, Cifra, Cyfra,Tzyphra, Cifre, Cyfre etc.", all derived from the Arabic Sifr.

Ifrah takes the position, that Arabic numerals were already in much use earlier, though more for calculations written in sand (instead using the Abacus). He speaks of prohibitions for the use of the method ....

In contrast we've well running Abacus schools in the likely "very modern Florence" around 1450.

Well, that's a complex problem. Perhaps later.
Renaissance isn't the theme of Ifrah, this appears at the end of his book, original "Histoire Universelle des Chiffres 1981".

mikeh wrote:More on "counting cards". Dummett, Game of Tarot Ch. 20 p. 419f:

Minchiate was, of course, a deliberate invention, a conscious variation upon existing Tarocco games; but we know that in both Florence and Rome the 78-card pack continued to be in use down to the seventeenth century. Our best clue to the nature of such games consists in those played in Sicily; the Sicilian game was probably first imported from Rome, and the traditions of Tarot play in Rome seem to have been strongly linked with those of Florence. If this hypothesis is sound, then what Minchiate shared with the game played in Florence and Rome with the 78-card pack was not the versicole but the fact that, in addition to the Bagatto and the Fool, all five top trumps, the arie, were counting cards.

I don't know on what basis Dummett makes the statement the 78 card game was used in Florence until the 17th century. As far as I know, always willing to be corrected, Francesco Berni's Capitolo of 1526 is the last time tarocchi is mentioned in Florence.

Dummett says as much in Il Mondo e l'Angelo (1993), p. 257: "Dopo il Berni non sentiamo più parlare da alcuna fonte fiorentina del gioco dei Tarocchi, ma solo dei Germini." After Berni we hear no further Florentine source speak of the game of Tarocchi, but only of Germini.

Even in 1980 he knew this, so I have to consider it an unguarded statement on his part. Recall that he didn't know much about the role of Florence or the early spread of the game southward in 1980; on the same page (419) he says of Tarocchi: "By the beginning of the seventeenth century it had spread to Rome." We now know that this date was at least by the 1450s; and of course Florence now has the earliest reference of all. The spread of the game was much faster and further than Dummett could have imagined in 1980.

Not that it matters for your inferences, but it is important to remember the context of passages like this.

So it appears that at the top end of the trumps there were five counting cards in Florence and Rome, just as in de Gebelin. For me the significance is that a group is thereby defined, solely from the rules of play, although perhaps driven by association in meaning. De Gebelin, to be sure, had five counting cards at the other end, too, whereas Dummett only has two. But Dummett goes on to say:

In Sicilian Tarocchi, and therefore, on our hypothesis, in the 78-card game as formerly played in Florence and Rome, there are seven, the five top trumps, the Bagatto and the Fool; and in Minchiate there are twenty-one, the eleven top trumps, the five lowest ones, four others and the Fool.

There remains a question in my mind whether in Florence and Rome with the 78 card pack the five lowest trumps might have been counting cards as well, expanded in Minchiate to 9 plus the Fool on one end and 11 on the other. These five would make another group, again independent of the meanings of the cards but perhaps driven by association in meaning.

I don't know how much weight to give reconstructions of original rules based on the Minchiate game. For the first five trumps, remember that one papa is already gone, so it includes Love. Would these original rules then include the Bagatto and the four papi of the Rosenwald, excluding Love? But we have the numbering on the Charles VI and the Strambotto, which indicate already three papi by 1500.

Why does it seem important that the top five and the bottom five should count? They are linked by theme, or group, that much is clear, but I don't see where you're going with it. Does it help you theorize what the designer was thinking when he chose the images of those groups, or their number?

In Bologna, as you know, the Sun and Moon were distinguished as rossi, "reds", but they didn't count for particular points. In a trump sequence, a grande, the Angel and any two of the three following cards, including the Sun and Moon, were necessary (a contatore could replace a missing one except the Angel), but otherwise I am not aware of any special rule regarding these cards.

Note also the earliest rules include the observation that some people try to include Love and the Chariot as papi, following the same equal-papi rule as the actual papi, but that traditional players frown on it. In any case, the papi don't count individiually, but only in a sequence.

It seems to me that speculating from Sicily, back to Rome, then from there back to Florence, in something as slippery as which cards counted for individual points after the invention of Minchiate and centuries after the original game was invented, is extremely uncertain. I am certainly not going to be convinced by this logic.

Ross G. R. Caldwell wrote:
I don't know on what basis Dummett makes the statement the 78 card game was used in Florence until the 17th century. As far as I know, always willing to be corrected, Francesco Berni's Capitolo of 1526 is the last time tarocchi is mentioned in Florence.

Franco Pratesi ...
http://trionfi.com/evx-germini-tarocchi-minchiate
... had detected recently a Florence 1606 document

1. 1606 - In Orto di Santa Maria Nuova

The law under examination is present in a section of the ASF. (2) However, the same text, with only some minor variations in the spelling of a few words, is present in a well-known publication of the Florentine laws. (3)
The Hospital of Santa Maria Nuova is still active as such in the centre of Florence; rather it is indeed the only ancient hospital that has remained active in its original place within the town. Information on its long history is easy to find, because it is associated with significant developments in the history of both medicine and fine arts.
Unknown to me is its Orto, or vegetable garden. At present, any garden in that location should be very small, after the several wings and secondary buildings that have been added in the meantime; we can easily imagine however that in the beginning of the 17th century there was still free space enough around the Hospital.
Whichever the real dimension and shape of that garden, we learn from the law under examination that playing several games was allowed there. Not only, there was also an Ortolano: the name should verbatim corresponds to a gardener, but his function was clearly that of manager of the playing area.
We have thus several positive elements for understanding that some games could be allowed there: the place was open to the public, game sets were available, and there was a responsible manager, who was charged of the control. This was likely the best place if not the only one in which playing some games was completely allowed in Florence.
Now, we are ready to gain knowledge of the games that people could play there at the time. We cannot expect dice or other gambling games, and indeed they are not even mentioned. Only three games were allowed there.
A game that one could easily expect to find, in the open air, is the game of Pallottole. That name is now mostly used with the meaning of bullets, but here the meaning is that of small balls: the game was thus Bocce, one of the several bowl games with a long history.
I am actually much more interested in the two remaining games, both played with cards. It may appear as an extraordinary situation to find two thirds of the games based on playing cards, without even a board game mentioned. Fact is however that these were not common cards, clearly not the cards with which Primiera or similar games were played; and only two packs were available for each game.
The first kind of special cards allowed is fully expected, Germini. It would have been strange if this traditional game were not present among the allowed ones. Somewhat unexpected, to me at least, is the second game allowed by then in Florence, Tarocchi. This is one of the very infrequent cases, or may be the unique one, in which I am finding Tarocchi recorded as used in Florence.
Most important for us is that we find two such packs and games mentioned. This can be considered as the starting point for a subsequent discussion on the members of this family of games: on the basis of the information here provided, it is certain that they were two at least.

In the Appendix:

A1. Bando / In aumento dell’Ordine del 1605, circa i giuo/chi, che sono permessi nell’Orto di S./Maria Nuova. / Pubblicato il dì venti di Giugno / MDCVI.
Il Serenissimo Grandua di Toscana, e per Sua Altezza Serenissima li Spettabili Signori Otto di Balìa mossi da giuste, e ragionevoli cagioni, che l’esperienza ne ha mostrate, aggiugnendo al Decreto, e Ordine per lor Signorie fatto sino sotto dì ventisei di Aprile 1605. per il quale si proibisce, e comanda, che nell’Orto del Giuoco di S. Maria Nuova non possi giocare altri, che solo Cittadini statuali, né ad altri giuoco, che a Germini, e Tarocchi con due paia di Tarocchi, e Germini per volta, e pallottole, proibendo in tutto, e per tutto il giuocare come si dice a chiamare, alla bassetta, alla scoperta, & all’alzata; aggiungendo all’ordine solito, qual resti nondimeno nel suo vigore, & osservanza, proibiscano, e comandano, che in detto giucoco non vadino, se non quelli, che sono di età da venticinque anni in su, quali non vi si possino trattenere se non sino all’ave maria delle ventiquattro ore, né vi si possa giuocare ad altro, che a Germini, e Tarocchi, come di sopra, & alle pallottole sotto le medesime pene, e pregiudizi, che nel suddetto Decreto si contengono, & all’Ortolano che tiene il giuoco di Stinche, di fune, di Confini, & altre pene a dichiarazione del Magistrato sino alla Galera inclusive.
Donato Rofia Cancelliere demandato ss.
Bandito per me Francesco Guidalotti Banditore questo dì 20. di Giugno 1606.

Some overview about the Germini production (without any note about Tarocchi) in Florence and Toscana is possible since 1638 (if I remember the year correctly).

Andrea Vitali mentioned, that he has more Tarocchi material involving Florence "after 1526". I don't remember details for the moment.