## From Marziano to the Cary-Yale and the Ludus Triumphorum

### Re: From Marziano to the Cary-Yale and the Ludus Triumphorum

61
Thanks, Iolon. Based on a comment I made on your site, I read your webpage on the Visconti tarot back in March, 2016, when, yes, you had reached the same conclusions that I (and Franco Pratesi) had come to about the missing cards of the Cary-Yale, although I had one criticism that no longer applies, if it ever did.

My further concern, of course, was to deal with the suit assignments that were on the Beinecke website and what they might reveal about the place of the Cary-Yale in the development of the game. That meant coming to grips with the order of the sequence, something I am still doing. The following is the continuation of my previous post.

9. The Beinecke grid in the context of predecessor decks

The next question is, assuming the plausibility of the conclusion in my previous post, how it fits into the development of the tarot, both in Lombardy and in relation to the other two regions. First, would it have been the first deck with the C order virtues?

The suits of the Cary-Yale are contrived to make 16, perhaps to match Marziano's 4x4 grid, perhaps to equal the number of triumphs. In that regard it makes sense to look at what a 14 triumph version of this Cary-Yale would have looked like, without the theological virtues, since there are the 14 cards per suit of the Brera-Brambilla deck, done around the same time, probably by the same workshop. 16-3 = 13, so one more triumph must be added to make 14. But where, and what? It can't be in the first column, because that will kick Justice into the second column, where it is too close to Fortitude to prevent there from being two virtues in one column. It has to be in the third column. If so, there aren't many possibilities. One is a second card for Time, that of an individual life as opposed to the cosmos.
What commends the grid just proposed is that two cards for Time would explain how Time came to be before Death in the C order, but in that case more likely after the grid of the Cary-Yale. I suppose that an alternative might be the Hanged Man, as a kind of triumph of the established order against traitors, but that is rather far from Petrarch.

Another possibility would be to add the Pope in third place, but put the Empress and Emperor in their own row beneath the others, in two columns each. The Empress would be the low card in both Coins and Cups, on the principle that these are "feminine" suits. In minchiate those two suits had feminine pages; they also went from Ace high to 10 low. In that regard they are similar to Marziano's suits of Riches and Pleasures, probably not coincidentally, given that Coins relate to riches and Cups to food and drink. Batons and Swords, being weapons, are the "masculine" suits. In that schema, the 14 card grid would look like this:
Another solution would be to suppose that Prudence was the first card, later replaced by the Popess. In this case the virtues would be in Aquinas’s order in reverse, perhaps to indicate the order in which they are most needed in life. In this case, however, the suits corresponding to each subgroup of triumphs is quite different from that of the Cary-Yale.
And of course the suit assignments could earlier have gone by rows instead of columns, closer to Marziano's own system. That is easy enough, if Fortitude is 6th or Prudence 12th in the first grid above. Or, if two cards for Time is a problem, the Pope could be there, to be removed when the theologicals are added between Wheel and Chariot.
A problem with these solutions is that they are still at some remove from the basic structure underlying a 14 card sequence, that of three conceptual groups, 4 papi + 4 virtues + 6 Petrarchans. This is a sequence reflected best in the A order of Bologna and Florence. Below is an approximation of that order of the archetype. Whether any actual games were played with such an order is another matter.
You might ask, why 4 papi? That is the traditional number in Bologna, but it is not documented until after most other places had already documented four, with the notable exception of Florence. The Rosenwald is sometimes cited as evidence, but it is probably from Perugia (http://naibi.net/A/601-UMBRIA-Z.pdf) and otherwise has the Bolognese order. So if anything, the Rosenwald counts in favor of four in Bologna early on. The Cary-Yale probably had only two, and the Brera-Brambilla at most three. For four, besides Bologna I can only refer to other games and the parallel with suits and virtues. "VIII Imperatori" in 1423 sounds like it could be 2 imperatori for each suit, upper and lower, male and female, good and bad, etc., perhaps extensions of the suits, themselves perhaps associated with the 4 ancient empires. 4 papi/imperatori would be a reduction by half. Also, it takes 4 for the total to be 14 (4+4+6), the same as the number of cards per suit. It is true, however, that there are other ways of adding up to 14. The extreme would be only 1 imperator at first, as in a Spanish game reported by Ross, to which could be added 7 virtues and 6 Petrarchans. The argument for 4 is not as strong as for the 4 cardinal virtues and 6 Petrarchans.

For the Petrarchans, the order to start with would be Petrarch's. For the virtues, there were various ways they were ordered. They were in a hierarchy, as befits a trick-taking game. Aquinas had prudence highest, then justice, then, fortitude, then temperance. Plato had temperance lowest, governing the appetites, then fortitude, governing the spirited part of the soul, then wisdom (Aquinas defined prudence as "wisdom for man"), and justice as what governs the whole, each doing its proper work. Wisdom of Solomon 8:7 had a list: temperance, wisdom, justice, and fortitude. Etc. I have taken Aquinas as the most hierarchical, but the original designer could have been inspired by any of the sources.

There is also question of suit assignments. Marziano had a grid from 1 to 16 on the one hand and four rows of a grid on the other. The virtues define a connection to the suits, but the other cards in the rows, Petrarchans and papi, do not have the obvious thematic connections to the virtues as Marziano's gods do to his categories. But there still could be mnemonic narratives connected virtues with at least the Petrarchans, and the papi as well if they were not all equal, as in the Bolognese game.

For example: Instinctual love is controlled by Temperance, victories are achieved by fortitude, justice is administered by old men and has death as the supreme penalty. prudence dictates bearing in mind the Angel of Judgment and can result in fame. In my opinion the narrative with Justice high works better, because it is the principle governing the Last Judgment and the best kind of fame for rulers, while Prudence is a virtue to be exercised so as to avoid Death and reach Old Age.

There may not have been any suit assignments in the A order at all. If there were, however, they would have been by rows, because it is impossible to distribute even three virtues one after the other into separate columns. But we don't know the distribution of cards per row. It might have been 4+3+3+4, for example. And we don't know where the blanks would have been. 4+4+4+2, the last two up above the grid doesn't take much reflection; it was the first thing I thought of at any rate, with the last two either going with the highest virtue or not connected to a suit at all.

I want to emphasize that by the archetype I am not talking about any actual decks. The original game could have been more like what is seen later, in one way or another, departing from this archetype in one way or another. I mean what the designer would have started with, even in his head.

The A order is closer to the archetype than any actual or hypothetical Lombard orders, if only because its virtues are in sequence, in a standard order, and it does not insert the non-Petrarchan Wheel. In Bologna, moreover, there are clearly four papi, perhaps from the start. But even the A order deviates from the archetype. Time is below Death, and the Chariot card in its A order depictions never has suggested Pudicizia, but rather Fame, which if so is in the wrong place. Pudicizia in fact seems to be missing, perhaps replaced by the virtue cards, especially Temperance and Fortitude, so that Love, in Petrarch's order, will be before them.

Did Filippo or some other in the C region decide to improve the A order's game by instituting suit assignments by columns, associating virtues with Petrarchans in the vertical direction, with a strict Petrarchan order for the latter, but adding the non-Petrarchan Wheel in the process? Or did one of the A region's cities decide to rationalize Filippo's order, thinking perhaps to make it easier to remember and more profound, including the reform of his poor order of virtues, while also improving upon Petrarch in the ordering of triumphs by giving the triumphal chariot to the one who triumphs over others, whose fame was celebrated by a chariot in republican Rome, i.e. the military type, as opposed to a woman celebrated for her virtue, and being truer to Marziano in having suit assignments by rows? The same questions can be asked of the 16 triumph game with the theological virtues. Either the 14 or 16 could have started in either place.

What speaks against Lombardy, at least for the 14 triumph game, is the artificiality of the columns needed in order to make suit assignments: having two cards for Time would seem to require an awareness of both the A order’s placement and Petrarch’s; the alternative, with two columns each for the Emperor and Empress is also rather artificial. Also, I am inclined to think that order, being the most important part of a trick-taking game, is more important than the concept involved in a card’s artistic depiction, which can and does vary even with the same order. To that extent, the A order is probably earlier.

Along these lines, the Cary-Yale’s order of cardinal virtues – Justice, Fortitude, Temperance, Prudence – looks more likely to have been a rearrangement of Aquinas’s order than the reverse, because two are in Aquinas’s order, Fortitude and Prudence. When inserting the virtues alongside Petrarchans, it makes a kind of sense to have Temperance as a triumph above Death, because temperate living serves to forestall death. And it makes sense for Love to be ruled by Justice, because marriage is a contract between two people, which is probably what is meant by the two people shaking hands on the Cary-Yale Love card. We should recall that Filippo had his first wife tried in a court of law on the charge of adultery: her execution was a matter of justice.

It seems probable that the Wheel was added in Lombardy. The Visconti had a large fresco of Fortuna at her Wheel in their palace on Lake Maggiore. There is also the specific placement of the Wheel in minchiate and the handwritten numbers of the Charles VI, so unlike its placement anywhere else but in my hypothesized Cary-Yale, suggesting at least a causal relationship with the latter one way or the other. But that is only one card.

The seven virtues seem more prevalent in pre-1440 art in Florence and Bologna than in Milan. However, Milan did have, as Wolff points out in her MA thesis, the profusely illustrated Song of the Virtues and Liberal Arts, (for which see the thread viewtopic.php?f=11&t=862 done in mid-14th century Bologna but soon acquired by the Visconti archbishop-ruler of Lombardy. And it could not have been the only model, because while some of the illustrations of the seven are strikingly like those in the Cary-Yale, others are not, for example Fortitude, which is also quite different from how that virtue is illustrated in the other regions (but which is similar to another Bolognese 14th century depiction of the seven virtues, viewtopic.php?p=12161#p12161, which also found its way to Milan). So even if the 14 triumph game started in the A region, the 16 card game, i.e. the proto-minchiate, may still have started in Lombardy, filling out Florence’s 4x4 grid with the theological virtues.

In favor of such an early date for both trionfi and minchiate is the satirical poem thought to have been written around 1440 by Burchiello, using both "minchiatar" and "trionfi" within a couple of lines of each other (viewtopic.php?p=15174#p15174). In the literal sense, the words don't refer to games, but this type of poem typically engaged in double meanings. Such a transfer might also have happened later, when Florence and Milan were allies. As I will argue in the next section, it is not unreasonable to suppose that even the PMB, in the 1450s, had the theological virtues among its triumphs.

10. From the Beinecke order to the full C order of 21 plus the Fool.

Another puzzle results from taking the Beinecke order and suit-assignments seriously. How could the C order as later reported ever have evolved from an order in which the cardinal virtues were in C order but the Petrarchans in minchiate order? Why would five of the Cary-Yale's Petrarchans change their placement by the 16th century (Angel, World, Time, Chariot, Wheel), but not the virtue cards?

Various explanations are possible. One, not very persuasive, is that in the minds of the players the placement of the virtues is what made their games unique to their region, part of its regional identity. The dropping of Prudence is then a bit surprising, but the placement of the other virtues is distinctive enough. But would enough players have known how distinctive their placement of the virtues was compared to other regions, or if they did, have found it worth preserving? Another explanation, even weaker, might be that once the virtues were spaced much further apart than cards usually were, they were more resistant to change. But Justice and Fortitude are quite close together in the C region, and Temperance and Fortitude in the B region. If in the 16th century the Old Man and Fortitude could switch places in the C region (9th and 11th), and the Chariot could go from one side of the virtues to the other in Bologna of the 17th century, why not two virtues one or two cards apart?

With neither of those explanations very persuasive, the hypothesis I want to examine is that the cardinal virtues were different from other triumphs in their function in the game, one that continued for some time, linking roughly the same groups of triumphs in the same order to the same suits. For that, these cardinals had to be in much the same place in the order, even as the size of the columns increased.

The next known Lombard deck after the two already discussed is the PMB (Pierpont-Morgan-Bergamo, also called the Visconti-Sforza), or more properly, the 14 triumphs of that deck in an earlier style than the others that have come down to us. Done in the 1450s under the Sforza regime, the extant cards both include subjects not in the previous decks and lack subjects seen earlier.

With 14 cards per suit, the 14 extant triumphs might be all of them, with none missing. If so, only one of them could be construed as a cardinal virtue, and even it is somewhat ambiguous, since it differs from most Justice cards in having a rider on a white horse in the background. Riders and a figure holding scales was also present on illustrations of Petrarch's "Triumph of Fame." So the possibility arises that it is a deck with no virtue cards at all, as Huck has proposed. If so, a Marziano-style grid is still possible, but the cards that link rows or columns with suits are not there. Here is what such a grid would look like, indicating the corresponding suits by both rows and columns
From this array it is not possible to infer any particular placement for a virtue. By columns, we have no idea where it would be in the column, or where the blank spaces were, and the same by rows. If such a deck existed, it is fairly useless, as regards to preserving the order of the virtues, as a transition from the Cary-Yale to the C order. I am not saying that the PMB was not of that type, only that if it was, there must have been other decks with a similar order that did have the virtues in their usual places.

On my hypothesis about the early decks, the A, B, and C placements, more or less, existed from the start, or near to it, in each region and continued in much the same form up to the time when they are recorded in definite orders. This is because the players are basically conservative, as the constancy of the order in different places over centuries shows. Change happens only when the game goes someplace new. How one virtue managed to stop being used altogether, if the Cary-Yale is any indication of general practice everywhere, has yet to be explained. In any case, I think we can infer that two of the cards normally included in Lombardy but which are absent from the surviving PMB cards were Fortitude and Temperance. Prudence may be missing as well. The only other triumph lacking from this set but present earlier is the World card. That makes a total of 17 or 18 as a minimum. It might have gone to 22.

Another issue is whether this deck would have descended from a deck like the Cary-Yale with the 3 theological virtues, or one like the Brera-Brambilla which, with 2 fewer triumphs, probably did not. In the previous section I outlined just one hypothetical grid for the Cary-Yale but 3 different ones for the Brera-Brambilla. It will be necessary to examine each one in turn, to see whether the PMB could reasonably have developed from from any of them, and from there, perhaps, to the whole 22, while still holding onto suit-assignments.

One of the grids I suggested for the Brera-Brambilla type was by columns and had 2 cards for Time.
The problem is that the cards to be added the Fool, Bagatella, Popess, and Pope, all of which come before Love. A new column would have to be added, with no possible cardinal virtue. The only solution I can think of is to run the papi along the bottom row, as in the second grid I proposed for the Brera-Brambilla, this time without the Empress and Emperor having to be in two columns each. In this case the Old Man can be either 11th, as below, or in the last column, as in the Cary-Yale. Since we don’t know what the World card looked like, we don’t know if it would have represented Fame, as in the Cary-Yale, or been a second card for Eternity, as would fit the PMB “2nd artist” card.
It is then easy enough to exchange Wheel and Chariot. Everywhere except in Florence (and sometimes even there, as in the Strambotto order), the Wheel went after the Chariot. It is also possible to suppose only three virtues, since it is only they whose positions need to be remembered for the later C order. Below I have the Old Man 11th, but he again could be 16th.

It is also possible to suppose only three virtues, since it is only they whose positions need to be remembered for the later C order. Below I have the Old Man 11th, but he again could be 16th.
Whatever special privileges or obligations that accrued to Swords, Batons, and Cups would then not accrue to Coins. There is an odd asymmetry among suits, but it is workable enough in play.

Another grid I proposed for the Brera-Brambilla was by rows instead of columns. This is also possible with the PMB cards. I cannot imagine that once a game was played by columns anyone would go back to rows, but perhaps the Cary-Yale type was not widely known compared to the Brera-Brambilla type.
It works, even if it involves learning the rows all over again in this configuration. Death is 13.

The last grid was that of the Cary-Yale itself, with the theological virtues. The problem again is how to fit all 4 papi into a 4x5 grid with Justice in the first column. The solution again is a row across the bottom. Either the theologicals are before Death and Time, or the other way around. The latter is shown below, in which Death is 13.
Another alternative would be a 6x3 grid, or even 7x3; either way there are the necessary 6 spaces from Popess to Justice, if the Bagatello is left below the grid. Again, the theological virtues could be either before (Beinecke) or after (minchiate) Death; if they were before Death, Death couldn’t be 13, if that is important. The Old Man has to be 11th, to keep Temperance in the 4th column.
I have been assuming that the theologicals were present in the PMB. Is that a safe assumption? If they were, there is some reason for thinking they would have been moved to conform with their placement in minchiate. In the tarocchi, the three celestials are in precisely the same place as the theologicals in minchiate. The replacement of one set by the other could have been a reason for three of the "second artist" cards. In fact, there are even certain visual resemblances between the three theologicals of the Cary-Yale and the three celestials of the PMB's "second artist" cards. Both Hope and the Star feature a lady looking up to something in the upper right corner of the card emitting rays. Both Faith and the Moon feature a lady with her right arm up and left arm down. Both Charity and the Sun feature an infant and someone holding a round object. https://blogger.googleusercontent.com/i ... Sgyv=s1982. These visual similarities would have helped the players remember their places in the order, but only if they were in the same places in the order as the theologicals had been in minchaite.

With the theologicals changed to celestials, and Lightning and Fire added as well, we get:
A problem with all these reconstructions is that, while possible, they include cards which not only seem to have come out of nowhere but also, in assigning suits to subgroups of triumphs, mostly created problems in fitting them into a 4 by something grid. Why have these new cards at all? What was wrong with the previous game, that new cards should be needed? And especially, why invite trouble by having a Popess, who could only be an embarrassment to the family who commissioned the deck, if not because of the legendary Pope Joan, an insult to the papacy, then because the Visconti had long been charged, in a series of papal bulls, of supporting the Gugliamites, a sect that among other things proclaimed that the time had come for a woman to be Pope? Not only that, as the papacy and the Visconti surely both knew (although it was not said publicly), the particular woman the sect had had in mind was the wife of a Visconti. It is true that the Popess could be explained in a perfectly orthodox way as the Church, legally the wife of the Pope, or simply as the female counterpart to the Empress; but her heretical associations with the Visconti Popess or Joan would make her a particularly inapt choice for a ruler who was in the process of negotiating a treaty with the pope so as to keep the peace now that he'd won his war.

Another new card was the Hanged Man. He, too, had a representative in the new ruling family (who probably commissioned the deck), namely Francesco's father Muzio "Sforza" Attendola, whose name, along with the depiction of a man hanging upside down from a gallows, antipope John XXIII had plastered on all the bridges of Rome. In the picture, as Moakley told the story (p. 95), the man held a scroll on which was written, "I am the peasant Sforza of Cotignola, traitor, who have committed XII treasons against my honor; promises, agreements, pacts I have broken." It is surely no coincidence that the Hanged Man is number 12 in the series, right before Death at 13. And it takes precisely the new cards in the front of the sequence to make those numbers possible. But it only works either in a complicated new grid assigning suits by rows, or else a grid assigning triumphs by columns to only three of the suits, dropping Coins. And why remove Prudence, Aquinas's highest virtue, rather than Temperance, Aquinas's lowest? Why all that, just to feature another family ancestor whose reputation would not inspire much confidence in his son's faithfulness in upholding a treaty with a pope? I turn to Ferrara.

11. The B region

In late 1440 Ferrara was graced by a visit from 13 year old Bianca Maria Visconti, with presumed sincere intent on the part of her father, apparently to acquaint her romantically with its young heir-apparent Leonello, who gifted her with "14 figures" on New Year's Day 1441. The visit turned out to be a trick to induce the man she was already engaged to, Francesco Sforza, to re-enter Filippo's service and take her off the marriage market. I do not put it above the Estense to make some additions to the 14 they had, precisely to embarrass both sides of this family, not dreaming that the couple would one day rule Milan. But there could have been other reasons as well for the additions. With the rule of the Estense fairly secure in Ferrara, Pope Joan would have been a good joke. And they had troubles of their own with possibly traitorous relatives, in this case the legitimate sons of Niccolò, who resented his choice of the illegitimate Leonello as his heir. It is reported that Niccolò did not formally designate Leonello as his choice until just before his death (https://en.wikipedia.org/wiki/Leonello_ ... of_Ferrara).

I will start with the basic 14, without additions, in a Marziano-style grid. There is no difficulty with a grid linking virtues to suits by rows (Marziano's conception) in the B order. The Wheel must be there to give enough separation between Fortitude and Prudence, which means that only three papi can be there. Also, while in the B order the Old Man is always before rather than after Death, in a grid operating by rows it is not an option: he has to be there, to keep two virtues from being in the same row.
If it is desired that Justice be in a high row, to suit its exalted status as the highest virtue (as it was for more than one d'Este duke of this period) and second highest triumph, a 3x4 grid will do it, with World above the grid and the Bagatella the very first card below. It does not seem to me that one of the papi would have been there, because such dignitaries would have come from positions of honor in some kingdom. A card below the grid would still be more powerful than the kings, to be sure, but with no authority in any kingdom. Emperors invariably came from one of the territories within their jurisdiction. "Below the grid" means, in essence, not holding office in any kingdom. Traveling entertainers had that affectation: free agents who went from place to place without a social position in any of them. If they had power over kings, it was by the nature of their trade, creating illusions and for that reason sought out, reminding them that despite their precautions they were always vulnerable to being tricked.
In a 4x4 grid, all four papi can be accommodated, and again the Bagatella will be needed to make 16. This time he is part of the grid, and so in this case has that special relationship to one kingdom; sometimes they were granted privileges by some governing body, or became attached to a particular court.
With the meaning of "little thing," the Bagatella's title suggests the card's place in the sequence, as the lowest triumph. I take its preexisting double meaning as a sleight of hand player, from the "little things" he used, as the reason for the depiction of one on the card. Ferrara is the only place where the term with its usual spelling was used. In Bologna it was Bagattino, meaning a small coin, with no obvious relationship to the personage being depicted. Lombardy and Florence used the word bagatello, which seems to me a means of giving the Ferrarese term a masculine ending. (It is true that Renier spelled it "Bagatella" in transcribing the "Pavia" verses, but he was mistaken; viewtopic.php?f=12&t=2141). With the masculine ending, the word has no history except in relation to the game.

Here, without yet adding the Hanged Man, Death is 13. Then, from the preceding precedent with 14, there could easily be another card necessarily below the grid, i.e. the Fool, but given no number, not even against suit cards. Or perhaps the Bagatella and Fool were not yet differentiated. The face of the d'Este Fool looks somewhat like that in the famous portrait by Fouquet of Niccolò's Fool Gonella, who was no fool.

In adding cards with assignments by rows, the triumphs in each row change. For that reason the task of memorizing new subgroups is likely not to have been popular, and in that case it is no wonder that 70 card triumph decks, suggesting 5x14, could exist in Ferrara at late as 1457 (http://trionfi.com/0/e/00/). With all 21, there can even be a 4x5 grid by rows, but I cannot see that having to memorize such long rows of randomly distributed triumphs would make for a very fun or popular game.

By columns, the subgroups are much easier to remember. Above, the first 14-triumph grid by rows in this section (with the Pope card) is also an example by columns, because it has one virtue both by rows and by columns. The columns can also be made more equal, with quite a bit of flexibility about what the added card would be, replacing the theologicals.
By columns it is possible to add the Bagatella and Popess for 16 only if they are either both below the grid (so as to keep Temperance in the first column) or part of a row across the bottom, as seen below.
This is a rather complex grid, so probably not as popular as a 14 card version.

A 4x5 grid is not only thinkable but also easy to remember. Since there has never been any indication of the theological virtues in the B order, I assume that the celestials were there instead.There are two possibilities, with and without the Hanged Man. Below is the grid without the Hanged Man, but with the Bagatella.
With the Hanged Man, there would be 22 and the World above the grid. Without the Bagatella, every triumph up to Prudence moves down one place, creating a space for the Hanged Man at either 11 or 12.

Since Temperance is 6th, Fortitude 9th, and Justice second from the end, a 3x6 grid (three columns, for three virtues in 6 rows, shown below) will be possible once the virtues have been reduced to three, if the Bagatella is put below the grid (shown) or the World above it. In a 3x7 grid, with one more row, both will be inside it. To keep Death 13, the Hanged Man has to be there.
It is evident that the Hanged Man has replaced Prudence. In all of the B order grids, whether the shorter ones by rows or the longer ones by columns, Prudence is invariably in the third column, because of where the other three virtues are, and if by rows it has to be just before Death, so number 12. The reason John XXIII picked the number XII for Muzio was probably that it was the number of Judas, always put last in the gospel lists, and so applicable to traitors generally. So 12 is just the right number for Prudence to be changed to the Hanged Man, depicting the prudent action of a ruler against traitors and a warning to prudent subjects. As I mentioned in an earlier post, a certain Vincenzo Imperiali, in a Risponsa to Lollio's famous Invettiva, c. 1550 (and since there is no record of this Imperiali otherwise, perhaps Lollio wrote both), includes a section that goes through all the triumphs in B order from the top, although some are only alluded to and not mentioned by name. One of those not named is the Hanged Man. Instead, in precisely that place in the order, he speaks of prudence, saying (as reported by Pratesi at http://naibi.net/A/03-FERRARA-Z.pdf):
Vien poi la Morte, et mena un’altra danza,
Et prudenza e malicia attera,
Et pareggia ciascuno alla bilanza,
Ma 'l vecchio saggio la Fortun' afferra…”

(Then comes Death, and makes another dance,
And prudence [prudenza] and malice [malicia] down here.
And everything equals out on the scales [bilancia].
But the wise old man seizes Fortune…)
The term bilancia, “scales,” is not a reference to the Justice card, because Imperiali already referred to it, the second from the top; nor is there any suggestion otherwise that it might be a card of minchiate, in this case one of the signs of the zodiac, Libra. The term is to show that a sovereign who sends a traitor to the gallows (“another dance”) prudently restores the balance between sovereign and subjects. That this description is meant to refer to the Hanged Man is supported by Pratesi, who observes that “the 22 triumphs are deducible with relative ease from the verses,” even if “several are only indicated by paraphrase" (essay just cited, my translation). Andrea Vitali, another researcher who has written about Imperiali in this connection, even entitles his online essay “L'Appeso: Il Traditore e la Prudenza.” Whether or not Imperiali was drawing on a tradition that survived into his time (c. 1550), it is a logical enough connection to make.

So it seems to me that the Hanged Man, on the present hypothesis, would have entered the sequence in Ferrara, and gone from there elsewhere, to replace Prudence. The Bagatella, from its spelling, suggests Ferrara as well, and it also is needed to fill out grids linking virtues to suits in the B order. If the Bagatella, then perhaps also the Fool. "Popess" also sounds right for the Estense, a family that enjoyed satire. As for the other new cards, the five between Devil and Sun, it is hard to say at this point where they would have come from.

As for any particular reason Death was 13, the answer is not so straightforward. But it was, even if the numbers had to leave out a triumph to do it, as in Bologna and maybe Florence; and if it wasn't, the card didn't get a number on it (as in the Rosenwald). I have still found little indication that the number was considered unlucky in Italy, or associated with death then, fact the opposite (viewtopic.php?p=24338#p24338). The only suggestion is from the Modena Perceval, where it was the "siege perilous" at the Round Table. Perhaps it was 13 because the Hanged Man was 12.

12. Summary.

In this post I have tried to situate a Cary-Yale tarot using a Marziano-style grid and the former Beinecke' order and suit assignments within a historical development both in Lombardy and in interaction with the two other regions of the tarot. Comparing 16 and 14 card versions of that grid in Lombardy with the likely 4+4+6 archetype, a grid in the A order is far closer than the Beinecke's or anything resembling it. So it is likely that the game originated in the A region, with or without a Marziano-style grid by rows, and then was modified in accord with the requirements of a grid with suit assignments by columns rather than rows. As for the presence of the theological virtues, however, because the Wheel is more likely to have been a Lombard addition, where it is needed, and its position in Minchiate parallels that of the Beinecke, the suggestion is that an order including them started in Lombardy and then went a fairly short time later (within a decade or so of the Cary-Yale) to Florence, where it endured and expanded into minchiate.

By the 1450s the Lombard tarot had expanded again, with the new triumphs first seen in the PMB. The inclusion of at least 4 triumphs plus the Fool can be accommodated into the grid structure, but with no apparent reason for their being added and several reasons against it: the grids have to be rather contrived, and several of the additions could easily have generated ill will from the papacy if it was thought, as would have been reasonable, that the Visconti-Sforza family was responsible. Ferrara is a more likely source for these additions, where the grids actually invite them, in the right places in the order, and also the reduction to just three virtues, and the same number of columns in the grid, with the Hanged Man substituting for Prudence. Moreover, the additions fit the personalities and general satirical spirit of the d'Este court. However, the grids by columns remain quite contrived, compared with that of the Cary-Yale.

The more unwieldly the grid, the less likely that they that they were used in the game. It might have been only at the beginning, with 14 and/or 16 in the A region by rows, 16 in C by columns, and 14 by rows and/or columns in B and/or C, that they played a role. After that, with Petrarchans interspersed vertically with virtues, the educational objective of applying the virtues in the context of major life concerns was achieved without the grid, and it could be dispensed with. With the three orders established, and their characteristic placements of virtues, their distinctive distributions among the triumphs remained much the same, the virtues changing over the years no more or less than any others of the early triumphs, despite the abandonment of the grid.

Of course more evidence is needed. The present hypothesis is only an inference from what came before the ludus triumphorum, the quite insecure Beinecke order and suit assignments, and the various orders that came about. Aside from the former Beinecke captions, of questionable authenticity, there is no documentation of any tarocchi game played with such a structure. But from my experience with the Beinecke, we can also see how evidence gets destroyed: if there is nothing in prior notes that have been saved, or other easily identified indications that still exist, to back up such notes, then they are simply erased, as if they had never been.