I set up this blog in 2021 to work out a hypothesis for the origin of the tarocchi, as the tarot is known in Italy. The result was long and meandering. Since then I have been trying to write a shorter version that would include only the essentials of the argument, without exploring all the possible variations. The result is below, as of Aigust 2025.
FROM MARZIANO TO THE LUDUS TRIUMPHORUM
1. A "misunderstanding."
Until the end of March, 2019, scans of the trumps of the Visconti di Modrone, the oldest extant set of tarot cards, appeared on the website of the Beinecke Library, where they are held, in a very odd way. Of the 67 surviving cards, each with a title and short description, first came scans of the suit cards of Swords; then “Empress of Swords,” “Emperor of Swords,” and “Love (Swords)”; then the suit of Batons, then “Fortitude (Batons),” “Faith (Batons),” and “Hope (Batons)”; then Cups, followed by “Charity (Cups),” Chariot (Cups),” and “Death (Cups)”; and finally Coins, ending with two unnamed cards recognizable as the World and Judgment, in that order.[1] These suit assignments are reminiscent of another game, that described by Marziano da Alosio (Marziano da Tortona) for the same duke of Milan twenty or thirty years earlier, which also seemed to link groups of trumps with suits, this time of four groups of four Greco-Roman gods and demigods, linked to four suits of different types of birds.
After I emailed the Beinecke library about the Modrone cards, curator Timothy Young replied (May 28, 2008), “Cataloging information about the cards was received with the collection when it was given by the Cary family to Yale. The author of the printed catalog to the Cary Collection used their descriptions when he created fuller catalog records.” That seemed to imply that the order and suit attributions would have come with the cards when Yale acquired them from the Cary family. But there was no surviving documentation. I checked with him again in 2016. He said he had looked in vain for documents that came with the deck; he also referred me to a 1981 catalog of the collection.
The order, albeit without suit assignments, in fact occurs in William Keller’s 1981 A Catalogue of the Cary Collection of Playing Cards at Yale University: “Empress, Emperor, Love, Fortitude, Faith, Hope, Charity, Death, World, Judgment.”[2] Keller’s bibliography for the deck shows only one item, “Yale University Library Gazette 52, 4 (April, 1978), 227, 22, and no. 51.” That article says nothing about either the order or any suit assignments. Its bibliography mentions a 1974 Yale MA thesis by Martha Wolff, “Bonifacio Bembo and the Minchiate cards painted for Filippo Maria Visconti.” The Minchiate deck, of course, is the only other one known to have the theological virtues, although in a different placement and order than Keller’s. The other trumps do appear in the same order in minchiate as in Keller’s catalog; but of course there are many others in between, missing from the Modrone’s surviving cards. Nor does minchiate have two knights and two pages in each suit, male and female, as is seen in the Modrone.
In March of 2019, revisiting the subject, I contacted both Wolff and Keller (as well as again Young, who had nothing new). Wolff replied (March 20, 2019) that she had not been concerned about the order (it was an art history thesis) and did not recall any in particular, but she did not know how the cards were housed. She contacted Yale and had them make a pdf of her thesis. There she lists the trumps twice, both based on Ferrara orders, first (p. 5) the “Steele Sermon” – Moakley had thought it applied to any 15th-century deck - and then Bertoni. Only her second (in Appendix A) has the theological virtues; they are before Death, as Keller had them later. In contrast, her source, Sylvia Mann’s Collecting Playing Cards, had them in their correct order after Death. There is nothing about suit assignments. In fact, in 2019 Wolff was surprised to see them on the website.
On March 21, Young contacted me with new information: he had looked at how the cards were stored, in four boxes, each containing the extant cards of a particular suit plus those trumps that the website associated with that suit. So a cataloger must have somehow assumed that the trumps in each box belonged to the suit that was also there. The boxes, he added, were from the 1970s or 1980s. In light of what was obviously a misunderstanding, Young decided to remove the odd references to suits in the titles of the trumps. Over time, the order of the trumps and their alternation with suit cards changed as well.
On March 28, Keller emailed me about the order. He couldn’t remember precisely where he got it, but he did not make it up; it was probably from some article by Dummett or Decker. Sure enough, Michael Dummett’s 1974 “The Order of the Tarot Trumps (Part II)” proposed 24 trumps for the Modrone, marking those extant with asterisks: “*Angel (Judgment), *World, Sun, Moon, Star, Tower (House of the Devil), Devil, *Death, Hanged Man, Hermit (Time), Wheel of Fortune, *Chariot, *Charity, *Hope, *Faith, Justice, *Fortitude, Temperance, Prudence, *Love, Pope, *Emperor, *Empress, Bagatto.”[3] Those with asterisks have the same order as Keller's list.
Where did Dummett get this order? It was a hypothesis. He thought that it reflected an original order of trumps shared by all three regions, that of his “A” region as exemplified by the handwritten numbers on the “Charles VI” deck.[4] It lacked the theological virtues and Prudence, but it would make sense to put them with the other three because in that order, excluding minchiate as a later creation, the virtue cards are all one after the other. Prudence is there because if six virtues were there, probably the seventh was, too; the Popess is not there because it is just too bizarre, and a female pope was not part of standard tarot imagery; that there are twenty-four instead of twenty-one maintains a 3:2 ratio of trumps to cards per suit, adjusting to the Modrone’s unusual sixteen cards per suit versus the usual fourteen.[5]
To that extent, the list is his hypothesis based on reasonable assumptions, if not the only ones possible. The most questionable one, no longer made in 1980, was that one order must have been shared by all regions at the outset. His 1974 reasoning was that it was originally an aristocratic game, and aristocrats from one region would have wanted to play the same game when they visited aristocrats in another region.[6] In 1980 he was emphasizing the “extreme localization” of play.[7] Correspondingly, he did not propose a specific order for the Modrone. However, the plausibility of the assumptions has no bearing on the present question, which was to explain the origin of the order that Keller presented. That much has been accomplished.
But mysteries remain. Why hadn't the Beinecke cataloger responsible for the website given a title and suit for the last two cards, which Keller had named World and Judgment, even while putting them in Keller’s atypical order, Judgment last? It seems too obvious a lack to be simple carelessness; could the cataloger be copying an omission in his or her source? Also, a division into four groups of trumps, each associated with a suit, is still there in another form, from the cards’ being in four definite boxes. Is this distribution, two or three cards with a particular suit, simply random?
Before Marziano’s treatise had been studied in detail, one might have thought so. But until 1989, Marziano’s treatise was misunderstood, based on an incorrect summary by Durrieu that attributed just sixteen cards to the set.[8] Even in 1980, Dummett thought that there had been only sixteen cards.[9] In 1989, Franco Pratesi did give the relevant details of Marziano’s game,[10] that besides the four groups of four there were also four suits of birds,each associated with a group of four gods.
Dummett speculated about a connection to the Modrone in the very next issue of the journal. After first stating why he thought a different deck, the Visconti-Sforza, would not have originally had only fourteen trumps (presumably to match fourteen cards per suit), he added: “We know, however, that the composition of the Visconti di Modrone pack did not conform to what later came to be standard: the hypothesis that it contained only sixteen trumps is accordingly a real possibility.”[11] This, of course, is quite a different hypothesis than the one earlier. It fits my initial question of whether there might have been a connection between the suit assignments at the Beinecke and the suits.
2. Marziano's game
As Pratesi presented Marziano's treatise, later confirmed by a full translation by Ross Caldwell, Marziano’s god-cards can be put in the form of a 4x4 grid in which all sixteen form a hierarchy by columns, while the rows define the four groups (see table below). For the rows, what correspond are in its second paragraph, beginning “The first order is indeed of Virtues, which consists of Jupiter, Apollo, Mercury, and Hercules.” He then describes three more orders, each with four gods. Then he introduces four orders of birds: “And subordinated to these [gods] are four kinds of birds, being suited by similarity.” These four orders are in the last column below. The text then has a very compressed sentence whose sense is more fully expressed by adding what I put in brackets. The fifth column below deals with what I have put in brackets from that sentence:
Thus to the order of virtues, the Eagle [is subordinate and suited by similarity]; [to the order] of Riches, the Phoenix [is subordinate and suited by similarity]; [to the order] of continence, the Turtledove [is subordinate and suited by similarity]; [to the order of] of pleasure, the Dove [is subordinate and suited by similarity].[12]
Finally, towards the end of the same long paragraph, he adds, “But the gods are held to this law among themselves: that he who will be first designated leads all the others following in sequence.”[13] The number next to each of the gods corresponds to the number Marziano gives it in the exposition that follows. I do not want to dwell on the choice of gods: suffice it to say that they correspond to twelve of the thirteen gods proposed in ancient times as "Olympian." Some lists included Vesta; others had Bacchus replacing her. Marziano chooses both and leaves out Vulcan. To make sixteen, there are four demigods that can be fit into his four categories.
Table 1: Marziano’s game, groups by rows, trumps 1 high (most powerful) to 16 low, orders in bold.
Gods Orders of gods Orders of birds
|
1 Jupiter |
5 Apollo |
9 Mercury |
13 Hercules |
<Virtues, suited by similarity to > |
Eagles, with their Kings |
|
2 Juno |
6 Neptune |
10 Mars |
14 Eolis |
<Riches, suited by similarity to > |
Phoenixes, with their Kings |
|
3 Pallas (Athena) |
7 Diana |
11 Vesta |
15 Daphne |
<Virginities, suited by similarity to > |
Turtledoves, with their Kings |
|
4 Venus |
8 Bacchus |
12 Ceres |
16 Cupid |
<Pleasures, suited by similarity to > |
Doves, with their Kings |
It is not hard to see a parallel to the Modrone, if it originally had sixteen trumps divided into four groups, each connected with the suit in the same box “by subordination.” But what could this subordination be, and would it have been expressed in the game?
Here again is Dummett in February 1990. (Its page references are to Franco Pratesi’s 1989 account; “p. 34” refers to the subordination by groups, “p. 35” to the ranking from 1 to 16, all above all the birds and their kings.)
Were the gods trump cards?
.... Their assignment to suits (p. 34) suggests that the gods were merely extra
court cards, ranking above the Kings. Their ranking in order (p. 35) suggests
the contrary, that they functioned as genuine trumps. Marziano's statement (p. 34)
that they beat the Kings and pip cards (ranks of birds) could be read either
way. If they were trumps, their assignment to the suits is pointless; if they
were superior court cards, their ranking among themselves is pointless. Of the
two hypotheses, Signor Pratesi's, that they were trumps in our sense, seems the
more probable. But there are other possibilities: for instance, that, when a
King or pip card was led, the trick could be won by a god only if it was of
that suit, but that, when a god was led, it could be beaten by any higher god.
If this seems complicated, we should remember that evolution sometimes goes in
the direction of simplicity; we should recall also the complicated rules about
the trump suit in Karnöffel. This hypothesis would make Marziano's game
ancestral to Tarot, but at a considerable remove.[14]
Dummett imagines the god-cards winning tricks in two ways, as members of a hierarchy of sixteen and as the highest cards of their suit. This use of the word “suit” does not correspond to any one word in Marziano; it is the combination of an “order” of gods with an “order” of birds “suited by similarity” to it – virtues with “stately” eagles (as Caldwell and Ponzi explain this idea[15]); riches with phoenixes, associated with “oriental wealth”; virginities with “chaste, shy, faithful” turtledoves; and pleasures with “gregarious, promiscuous doves”; Dummett’s word “suit” in reference to Marziano’s game is then simply an abbreviation for any of those four combinations of birds, kings, and gods. So far, so good.
On this proposal, however, if any card other than an Eagle, for example, is led, even the lowest pip card of that other “suit” would win a trick over even the highest card in Eagles. This consequence would give the god-cards very little power outside their own “suit”: since the main feature of trumps is an ability to take cards of any suit, they would not be much like trumps at all.
We could try modifying Dummett’s proposal: all god-cards beat all suit cards of whatever suit, but a god-card of the suit led takes priority over other god-cards. For example, if Doves are led, Jupiter beats any Dove, but not any god in the order associated with Doves. Even Cupid beats Jupiter if a Dove is led. Such a rule would give a ludic equivalent to the last sentence of the treatise, where Marziano says that even Jupiter was not able to escape Cupid’s arrows.[16] Likewise, if Turtledoves are led, Daphne can triumph over Apollo, as Marziano describes her doing.[17] (However, he also says that Mars is “never defeated, but by Jove,”[18] which has no ludic equivalent; Marziano’s descriptions of power relationships don’t always fit my proposed rule.)
It might then be asked: if such a rule applied, one that favored groups of gods of the suit led, why wouldn’t Marziano have spelled it out? He was writing an edifying discourse emphasizing virtue and guiding the artist’s depictions, not a rule book. Did one have to follow suit if one could? It is not said. However, it may be replied, other trick-taking games did have such a rule, and it didn’t have to be stated. These other games did not have trumps, however. Yet even this may not be true. The account records in Ferrara for 1423 list the purchase from Florence of a card game called “VIII Imperadore.”[19] The word “Imperadore” suggests cards whose power ranges, at least potentially, over more than one kingdom, and “VIII” suggests eight of them. With four suits, they could have been assigned two to a suit, more powerful than the suit’s king, yet unlike it also able to take cards in other “kingdoms.” If so, there may have been a rule that emperor-cards in the suit led took priority over other emperor-cards in winning a trick. In that way, it would be like Marziano’s game as I have imagined it.
If this solution is not persuasive, there is another way each order of birds could be subordinated to a corresponding order of gods: in the scoring, with extra points for sequences that went from gods in an order to the corresponding Kings and below. Points for combinations were probably already known, in card games as in dice. They certainly were known later, both in trumps and courts, although not going from one to another.
In relation to my proposed trick-taking rule, another consideration has to do with what is known as the “equal papi” rule in tarot games. In “VIII Imperadore,” if there were two trump cards per suit, there would have been no need for them to be ordered into a hierarchy across suits. Besides a rule that emperor-cards in the suit led took precedence in a trick over those of other suits, there could be a rule that if two of equal priority were played to the same trick, then that played last won. This last is an actual rule in the Bolognese game, the “equal papi” rule. There it is limited to the four “papi,” as they are still called in Bologna, a rule that also survived in 1565 Piedmont.[20] It is a rule so contrary to the usual practice that it is likely to go back to the earliest days wherever it was part of the game. Since Piedmont was rather distant from Bologna and ordered its trumps in the Lombard manner otherwise (except for having Judgment last, another Bolognese characteristic), it probably would have learned the game from the adjoining duchy of Lombardy rather than from Bologna. If so, the same rule probably would have held in Florence, where the 1423 “Imperadore” deck was produced, and where Judgment was also the final trump. I am proposing that the “equal papi” rule, once followed everywhere, might have originated in the game of “VIII Imperadore,” a game with trumps but no order of trumps.
3. From Marziano to the Modrone
Going from Marziano to the subjects of
the Modrone, there is an allegorical aspect as well as a ludic one. What “similarity"
could there be between the trumps of the tarocchi and its four suits of Swords,
Batons, Cups, and Coins? Actually, there is a set of correspondences well
enough known to serve the purpose. Gertrude Moakley
pointed out that Justice in
medieval representations held a sword; Temperance two vessels; Fortitude 
often
a column or staff; and Prudence a mirror, which resembles a coin.[21]
She cited Innocenzo Ringhieri’s 1551 Cento giuochi liberali et d'ingegno[22], which proposed a game in which Justice corresponded with Swords, Temperance with
Cups, Fortitude with a suit called Columns, and Prudence with one called
Mirrors.
Another text was a funeral oration for Giangaleazzo Visconti, Filippo Maria Visconti's father, who died in 1402 Milan: "O chiara luce, o specchio, o colonna, o sostegno, o franca spada, che la nostra contrada mantenevi sicura in monte e in piano!" [O clear light, o mirror, o column, o support {or sustenance, for Cups}, o confident sword, you kept our territory safe in the high places and the flat! – trans. MSH] (Arch stor tomb, anno xv, p. 792).” This seems a metaphorical way of alluding to Giangaleazzo as the embodiment of the four cardinal virtues. (It would be of interest to know if “clear light” referred to a suit of cards as well as a kind of superior vision.)
An allusion to these virtues would have fit a title of Giangaleazzo’s, “conte di virtu”; his first wife’s dowry had included a county by that name. His son Filippo inherited both; so when Marziano in his introduction says that by observing the “deified heroes,” Filippo will “be ready to be aroused to virtue,”[23] he is probably making the same allusion. The Greco-Roman deities are a mixed bag as far as virtue, except in the Latin sense of a special quality or excellence. A game centering on Christian moral virtues would suit a Christian prince.
Fortitude was in the box with the Modrone’s suit of Batons, consistent with this virtue-suit correspondence. In a 4x4 grid, Swords and Cups lack one trump each, Coins two. The other cardinal virtues, on the present hypothesis, could take three of these spaces, each with its corresponding suit. Moreover, such placements are just how they fit in the Lombard order: after Death for Temperance and after Love for Justice, leaving Prudence with Coins. Exactly where the two remaining spaces, in Batons and Coins, would occur, and for what cards, is not yet determined. Below, they are 5 and 16 merely for convenience. I give the Emperor and Empress the possibility of equal status because they would likely have been subject to the "equal papi" rule. Since there were two knights and two pages, male and female, in each suit, they likely would have been subject to the same rule - and maybe even the Kings and Queens.
Table 2. Groups by columns, “Susio” placement of cardinal virtues, other trumps in accord with Keller and the Beinecke’s boxes, hypothesized cards in italics, cardinal virtues and suits in bold, remaining cards as ????
|
16 ???? |
12 Temperance |
8 Hope |
4 Justice |
|
15 Angel (Judgment) |
11 Death |
7 Faith |
3 Love |
|
14 World (Fame) |
10 Chariot (Chastity) |
6 Fortitude |
2/1 Emperor |
|
13 Prudence |
9 Charity |
5 ???? |
1/2 Empress |
|
Coins |
Cups |
Batons |
Swords |
Two cards remain
to be identified. An educated guess can be made by looking at the cards already
known. Five of the six Petrarchan triumphs are there already (it is mainly their titles that would have been the inspiration). Love
and Death are obvious from their titles. Eternity is represented by the Angel of Judgment. Chastity, in many manuscripts Pudicitia, the Roman
virtue of avoidance of shame, is represented by the lady on the Modrone Chariot,
who holds a shield in accord with Petrarch’s description of her fight against Cupid.[24]) Fame is the lady on the World card holding a winged trumpet, a conventional attribute of Fame, in one hand and a crown in the other, with a mounted knight below. Below are the most relevant details of these two cards: notice the indentations on the sides of the shield, for the jousting lance; for the whole cards, see section 5 below.
Petrarch had six triumphs: Love overcome by Pudicitia, overcome by Death, overcome by Fame, overcome by Time, overcome by Eternity. A card for Time would complete the series, and the Vecchio, or Old Man, was sometimes given that title. An old man, minus the hourglass, did in medieval iconography sometimes represent the deteriorating effect of Time; Petrarch also refers to himself in such terms.[25]
The usual placement for the Vecchio was just before Death. In the grid, however, that would put Temperance in the same column as Prudence. Since the rest of Petrarch’s six occur in his order, Time would likely do the same, going between the World, as Fame, and Judgment, as Eternity. It might have been the same image of an old man, but another image of Petrarch's was that of the Sun, which opens that Triumph; in that vein, a Milanese Triumph of Time thought to be of the 1440s – or, by others, 1450-1475) shows only a young sun-god in his chariot.[26] It is true that the Sun later occurred as the third highest triumph in the Lombard order, and Judgment second. But such variations of one space are common in orders otherwise similar. For example, Alciato seems to have had the Chariot below Fortitude, whereas an anonymous writer in Pavia of the same period puts them the other way around.[27].
The remaining empty space is in the column with Fortitude. There is more than one possibility, but my speculation is the Wheel, Boccaccio's main triumphator in Amorosa Visione, as a kind of seventh Petrarchan-like triumphator, It is also one of the two surviving trumps in the Brera-Brambilla deck, of the same milieu as the Modrone, and the subject of a large fresco in one of the Visconti palaces.[28] Putting that card lower than the Chariot is foreign to the Lombard order but would fit that of minchiate, if the theological virtues weren’t there. We need only suppose that minchiate moved them and Prudence to be with its other unique trumps. In that case, the Modrone might well be a proto-minchiate, but with a Lombard order for the cardinal virtues. A card game called “minchiate” is documented in 1470 and 1477 Florence, and decks of “germini” in 1506.[29] (The card next to Fortitude could also have been something else, such as Worldly Fame as opposed to Immortal Fame.) With these additions, we get:
Table 3. Table 2 completed. Groups by columns, “Susio” placement of cardinal virtues, other trumps in accord with Keller and the Beinecke’s boxes, hypothetical cards in italics, cardinal virtues and suits in bold.
|
16 Angel (Judgment) |
12 Temperance |
8 Hope |
4 Justice |
|
15 Time (Old Man or Sun) |
11 Death |
7 Faith |
3 Love |
|
14 World (Fame) |
10 Chariot (Chastity) |
6 Wheel |
2/1 Emperor |
|
13 Prudence |
9 Charity |
5 Fortitude |
1/2 Empress |
|
Coins |
Cups |
Batons |
Swords |
This grid preserves the Beinecke’s placement of the triumphs in its four boxes. If the corresponding suits, “suited by similarity” to the virtues, are put below the groups, the result is precisely their placement in the Beinecke’s boxes. The groups are also easy to keep straight: they begin and end where there are two cardinal virtues. The placement of the cardinal virtues through their correspondence “by similarity” to the four suits, moreover, results in a striking coincidence with the known Lombard orders, a coincidence too great to be dismissed.
It is true that the order is not precisely the same as the later orders. But small variations existed even within regions, and most of the variations in the order in Table 3 from what came later are readily explainable: Angel (Judgment) before World is how it went in Florence and in neighboring Piedmont, and the two cards were next to each other. I have already explained how the three theological virtues and Prudence’s move from before to after Death in minchiate serves to put the special cards of that deck together. The Chariot’s placement below rather than above the Wheel, in all the Lombard-derived orders, is then a simple exchange of adjoining cards.
4. Rows vs. columns
My reconstruction departs from Marziano’s structure in one important respect: Table 2’s groups of four are simply subsets of the order as a whole, in the same order, i.e., 1-4, 5-8, 9-12, 13-16, corresponding to the columns of the grid. Marziano’s groups are the rows, taking every fourth trump in the order, i.e., 1, 5, 9, 13; 2, 6, 10, 14; 3, 7, 11, 15; 4, 8, 12, 16. The game of Table 2 is what I would call a “game by columns”; Marziano’s is a “game by rows.” Why the change? Let us suppose that the same cards were in an order where every row had a cardinal virtue. The most straightforward way is for them to be one above another, as in Dummett’s original proposal for the Modrone, except for the reduction from twenty-four to sixteen:
|
12 Chariot |
8 Faith |
4 Prudence |
Coins |
|
|
15 World |
11 Wheel |
7 Justice |
3 Love |
Swords |
|
14 Death |
10 Charity |
6 Fortitude |
2/1 Emperor |
Batons |
|
13 Old Man |
9 Hope |
5 Temperance |
1/2 Empress |
Cups |
Actually, as long as the four cardinals are together, the order otherwise does not matter. If it is preferred htat Death be 13, then the Old Man (or Sun) as Time can be in its Petrarchan placement between World and Angel. Moreover, there could be more or fewer trumps than sixteen: for fourteen, there could be blank spaces in the grid (one less card in two of the groups), or one above and one below a 3x4 grid. With twenty-four cards, there would be six columns (6x4). One or more of the cardinal virtues could even be separated from the others, as in minchiate – but in such a way that each had a different row.
The problem for players is that if rows determined the groups of trumps, it would be a challenge to learn even the most basic rule, that of determining which card wins a trick. For example, the bottom group in Table 4 is Death, Hope, Fortitude, and Empress. How to remember such an assortment? Some jingle or lesson in virtue could be created to jog the memory, but there would be four of them, plus another totally different order of sixteen. This game might appeal to parents of children wishing in inculcate virtue in their children in relation to the vicissitudes of life articulated by Petrarch, but even then it is overkill. The game by columns would be less daunting. However, with four cardinal virtues one above the next, or even three, they could not be in four different columns; only the game by rows is possible. Such a game would have quickly dropped out of favor, replaced by one without trump-suit correlations at all. Only a strange residue would remain: the “equal papi” rule, for its Imperadore, from the game of that name.
Besides Lombardy and Florence/Bologna (these two had all three virtues together), there is also Ferrara. Evidence there suggests the possibility of fourteen trumps and fourteen cards per suit.[30] Besides a 4x4 grid with two blanks, a 4x3 grid is attractive, with one trump above and one below, the three theologicals replaced by just one card, the Bagatella, added to the game because this trickster, unlike the Empress, is ungoverned by virtue. The game by columns is straightforward: the virtues need only be separated by two other cards.
Table 5. Game by columns, Ferrara order, fourteen trumps, cardinal virtues and suits in bold.
14 World
|
13 Justice |
10 Prudence |
7 Fortitude |
4 Temperance |
|
12 Angel |
9 Old Man (Time) |
6 Chariot |
3/2 Emperor |
|
11 Death |
8 Wheel |
5 Love |
2/3 Empress |
|
Swords |
Coins |
Batons |
Cups |
1 Bagatella
The grid structure can thus account for all three placements of the virtues: Florence/Bologna in the “game by rows,” and Lombardy/Piedmont and Ferrara/Venice in the “game by columns.” From sixteen or fourteen, the trumps become twenty-one, dropping the grid in the process. It is quite possible that some of these subjects came from a version of “Imperatore,” if it featured “good” vs. “bad” trumps, as the cards’ titles suggest in its namesake of Kaiserspiel (also known as Karnöffel) across the Alps.[31] Of the previous game, only the three ways of placing the cardinal virtues and the “equal papi” rule are left as its residue.
Also from Ferrara is some information that can suggest an intended audience for such a game: in 1442, decks are purchased for the use of the ducal children, aged 8 and 10.[32] A game in which groups of triumphs consist of cardinal virtues plus the six life-concerns of Petrarach's I Trionfi is well suited to teaching the integration of virtue into one's life. It is just the sort of educational game that parents would like to see young children engaged in playing.
5. Why not 5x4+1 or 6x4?
An objection might be that it could just as well have been 6x4 at the
start, or even 5x4+1 (the Bagatella below the grid, Prudence in the highest group,
no Popess). Against that is the precedent
of Marziano’s 4x4. Moreover, as Lothar Teikemeier has pointed out,
many of the Modrone trumps are depicted in such a way as to suggest an analogy
with chess pieces, sixteen to a side[33]. There is no particular reason why Death
should be on a horse, nor that Petrarch’s Triumph of Chastity should have a
chariot with horses and a horseman, as the Modrone has them (first two from left below). Likewise,
Petrarch’s Fame and Eternity do not call for depictions of towers (3rd and 4th). Yet the
Italian names for two of the pairs of chess pieces were precisely cavaliere
(or cavalo) – horseman or horse - and rocco, similar to rocca,
the word for a hilltop citadel. Emperor and Empress recall chess’s King and Queen.[34] The seven virtues and love plausibly
correspond to the pawns. To these I would add that the bishop was for Boccaccio
the alfino, which comes from a Hindu word for “elephant” but means “at
the end” in Italian; it fits the old men in both the “Old Man” card (fifth below, from the Visconti-Sforza, a successor deck) and the
Wheel (sixth, from the Brera-Brambilla, another successor, both with Visconti insignia). A chess analogy would have appealed to Filippo Maria Visconti, who "was accustomed from his youth to play various games," as Decembrio recounted in his biography of the duke.[35]
Another factor that seems important is the invariance in the position of certain trumps among the later orders:[36] the part of the sequence from Devil to Sun always goes in the same order everywhere; likewise, the Bagatella is always first, the Hanged Man is always between the Old Man and Death, and the Fool is always outside the sequence. In contrast, the twelve others (excluding the theologicals and Prudence) all vary by one or two places among the various lists. With the four dignitaries, or “papi,” the variation is that in Bologna and Piedmont, they were equal, and elsewhere not. An exception to this variability of original cards is the Old Man; but he may well have been an addition, or moved early on from higher in the sequence. An exception to the invariability of later cards is the Popess. Its variability can be accounted for as a result of a late abandonment of the “equal papi” rule, done after the trumps had reached 21.
Why this contrast between variability and invariability? Variability suggests the period before 1445, when cities were isolated and focused on individual self-assertion; later, with Medici allies taking power first in Bologna (Sante Bentivoglio, 1445) and then in Milan (Francesco Sforza, 1450), an alliance generalized by the Peace of Lodi in 1453, a period of cooperation and greater unity ensued, where standardization would have been desired. The Modrone is on the pre-standardization side of that divide. Florence, as the dominant crafts center,[37] would probably have played a leading role in this standardization. Other cities could have introduced new subjects (e.g. the Bagatella in Ferrara), with one center, probably Florence, determining whether these subjects would be part of the standard and where in the sequence they would go. There might even have been two such standards, one called “trionfi” and the other “minchiate,” with the former replacing the three theologicals with an equal number of celestials and the latter keeping both sets. The addition of the zodiac, ending in Gemini, could have been the occasion for the name change to Germini, so as to distinguish it from the game with a shorter deck.
It might be argued that the invariable placements are simply the most natural ones, given the subjects. “Bagatella” or “Bagatino” (in Bologna) naturally is first, in virtue of its name, “small thing” or “small coin.” Likewise, a hanged traitor is closer to death than an old man. The sequence from Devil to Sun is allegorically one of increasing light. But left to its own devices, surely some place would have had the Moon as the lowest celestial body, being closest to the earth after the clouds’ lightning (an early name for the Tower card), the Traitor lower than the Bagatella, being a worse deceiver than a sleight-of-hand player, or the Devil and the Devil’s House (another early name for the Tower) at the end of the sequence, as sinners’ just dessert.
Still, it could be that only the oddest of these placements, Star before Moon, was a late addition, the three celestials replacing Hope-Faith-Charity. In that case, a 4x5+1 grid for the Modrone is possible. Elsewhere, it just makes the “game by rows” more difficult. For a 4x6 grid, Marziano’s treatise itself can lend support to the hypothesis: if the bird-suits had eleven cards (ten pips and a king), the ratio of trumps to cards per regular suit would be close to Dummett’s 3:2. But whether such a ratio was thought important, as opposed to considerations having to do with the number sixteen, is unknown. Moreover, there remains the odd coincidence that except for the variable Popess it is only invariant cards that happen no longer to be extant in the Modrone
6. Two objections and a summary.
It might be objected that my whole argument for an initial grid structure depends on the questionable assumption that Marziano had a rule connecting his groups with particular suits in the game and that such a rule carried over to the Modrone. It is not an assumption but a “plausible hypothesis,” in Dummett’s phrase,[38] to account for the “subordination” found in the text, and just as Dummett’s 24-trump sequence for the Modrone is a hypothesis. New facts suggest new hypotheses, as Dummett recognized, and new ways of looking at old facts. It is supported but not proven by such facts. That neither Marziano nor the early tarot linked groups of trumps with regular suits remains possible. It is just that the character of the game would have been different, and the coincidence that the Lombard placement of the virtues distributes them into four groups matching the traditional correspondence of virtues to suits in the boxes at Yale remains unexplained.
Another objection might be that if Moakley’s correlation of virtues with suits occurred to me, it could also have occurred to whoever distributed the cards in the boxes at Yale, observing Keller’s order but using Moakley’s virtue-suit assignments and, for the distribution of the cardinal virtues, the relevant pages of Dummett’s 1974-75 articles or the tables in Game of Tarot.[39] To get the distribution of cards into the right boxes, it is just a matter of assigning suits to trumps in accord with Moakley's correlations and the Lombard order. The result will be an alternation of suits with trumps much like that proposed by Moakley in her imagined procession of soldiers in formation alternating with floats with allegorical figures on them.
One problem is that her succession of suit-bearers follows a different order, and her floats of trumps, the first of which was connected to both Cups and Batons, does not correspond to the Beinecke’s groups. It is true that someone could have imagined the missing virtue cards in their Lombard order, disregarding Moakley’s order and groups as well as the non-Lombard aspect of Keller’s, but keeping Keller's order and Moakley's suit-virtue correlations. But it seems to me that without knowing Marziano’s structure, it is unlikely that anyone would think of doing such mental gyrations: of all of Moakley's ideas, her suit-virtue correlations received the least attention (none, to my knowledge). I only thought of them because of their parallel with Marziano. Moakley, like Dummett in 1980, reported Marziano’s deck as only sixteen cards.[40] To that extent, how the boxes at Yale came to be the way they were remains a mystery. We might have known more had the libraries involved kept a record of the apparently trivial matter of how the trumps were stored and on what principle, if any.
Yet even without such information, the hypothesis that the Modrone was designed for a game linking trumps with suits has some suggestive evidence. I will summarize. In Marziano's treatise, it is plausible to interpret his "subordination" of each group of gods to a suit as implying some rule in the game linking these groups with those suits. It is also plausible to assume that "VIII Imperadore" had a similar structure, but with four groups of two, and in addition, a form of the "equal papi" rule seen later in the tarot. Since a game "by columns" is easier to play than a game "by rows", and the placements of the cardinal virtues in the Lombard orders allows for such a game, it is plausible that the Modrone's game had the former structure, and likewise in Ferrara, utilizing the association between virtues and suits that has come down to us. Sixteen would be the same as Marziano's, as well as providing visually for a parallel to chess pieces. Fourteen is suggested by the "14 figures" and "70-card triumph decks" of Ferrara. The order of trumps in Bologna and Florence allows instead for a "game by rows," as Marziano's would have been. It is a reasonable assumption that the "game of rows," due to its difficulty, would have been abandoned in favor of a game that was easier to learn but harder to play well, emphasizing strategy over moral teachings, therefore no links between trumps and suits. To make even more room for strategy, the trump suit could expand to include a papal pair and other additions, leading to both the tarocchi and minchiate, evidence for which is in the invariability of just those cards among all the later orders. A didactic game for children would have been transformed into a game of strategy for adults.
To be sure, this is all hypothetical: such earlier games might not have influenced the tarot at all; it might have been created in final form from the start. The available evidence to the contrary is suggestive but ambiguous. What remains is a plausible hypothesis with some explanatory power; as such, it can at least be a reminder of how little we actually know – the tip of the proverbial iceberg - about the origin of the tarot.
[1] This arrangement is no longer on the website, but I reported it in a note that Franco Pratesi included in his “Milano e Firenze – I,” pp. 519-537 of his Giochi di Carte nella republica fiorentina (www.aracneeditrice, 2016), on p. 531; the essay is also online under a different title at https://www.naibi.net/A/502-CARYYA-Z.pdf.
[2] New Haven, CT: Yale University Press, 1981, vol. 2, p. 53, https://collections.library.yale.edu/catalog/17389206.
[3] Journal of the Playing-Card Society 2, no. 4 (May 1974), pp. 33-49, on p. 47.
[4] Ibid., p. 46.
[5] Ibid., pp. 47-48.
[6] Ibid. pp. 44-45.
[7] Michael Dummett, The Game of Tarot (London: Duckworth, 1980), p. 398. Online at https://forum.tarothistory.com/viewtopic.php?t=1175.
[8] Paul Durrieu. “Michelino da Besozzo et les relations entre l’art italien et l'art français à l’époque du règne de Charles VI,“ Mémoires de I’Académie des inscriptions et belles-lettres 38, 1911, pp. 365-393, on p. 376. Online in Google Books.
[9] Dummett of n. 7, p. 33, n. 2. Online at https://forum.tarothistory.com/viewtopic.php?p=27685#p27685.
[10] “The Earliest Tarot Pack Known,” The Playing-Card 18, no. 1 (Aug. 1989), pp. 28-32, and no. 2 (Nov. 1989), pp. 33-38. Online in Academia.
[11] The Playing-Card 18, no. 3 (Feb. 1990), p. 75. Online at https://naibi.net.
[12] Ross G. R. Caldwell and Marco Ponzi, ed. and trans., “A Treatise on the Deification of Sixteen Heroes, by Marziano da Sant’ Alosio (Lulu, 2019), translation (except for what is in brackets); pp. 23, 25, original pp. 22, 24: “Subordinatunque his quator Avium genera, similitudinibus accomodata. Virtutum quidem ordini, Aquila. Divitiarum, Foenix. Continentiae, Turtur. Voluptatis, Columba.” Table adapted from their pp. 10-11, adding “suited by similarity” and the heading “orders of gods. Pp. 1-34 online in Google Books.
[13] Ibid., pp. 24-25: “Sed inter se dii hae lege tenebuntur, quod qui prior inferius annotabitur, sequentibus praesit.“
[14] Dummett of n. 11, p. 74, online at https://forum.tarothistory.com/viewtopic.php?t=1175. He repeats this hypothesis in Il Mondo e l’Angelo (Naples: Bibliopolis, 1993), p. 168, but using “suit” (seme) to apply only to the bird-orders (including their kings): each god-order, then, is “associated” with a corresponding bird suit. He does not mention the alternative, that the gods might be trumps without any rule correlating them with suits.
[15] Caldwell and Ponzi of n. 12, p. 13 (editors’ introduction). Online in Google Books.
[16] Ibid., pp. 92-93; "non ipse Iupiter effugere potuit" (Jupiter himself could not escape).
[17] Ibid., pp. 86-87: "Apollo frustra tempus, et preces effudit" (Apollo in vain expended time and entreaties).
[18] Ibid., pp. 62-63: “Iove excepto, non nisi Invictus.”
[19] Adriano Franceschini, Artisti a Ferrara in eta umanistica e rinescimentale. Testimonianze archivistiche. Parte I dal 1341 al 1471 (Ferrara-Rome: Corba, 1993). Selection translated by Ross Caldwell at http://trionfi.com/imperatori-cards-ferrara-1423.
[20] Ross Caldwell, Thierry Depaulis, and Marco Ponzi, eds. and trans., Con gli occhi et con l'intelletto, explaining the tarot in sixteenth century Italy (Lulu 2018), p. 17 and n. 8, pp. 31-32, to Francesco Piscina, Discorso sopra l’ordine delle figure dei tarocchi. P. 17 is in Google Books.
[21] Gertrude Moakley, The Tarot Cards Painted by Bonifacio Bembo for the Visconti-Sforza Family (New York, New York Public Library, 1966), p. 41 and n. 1, p. 53.Online at https://moakleyupdated.blogspot.com/2017/03/35-43-ch-3-family-for-whom-cards-were.html.
[22] Bologna, 1551, p. 132 (in Google Books).
[23] Caldwell and Ponzi of n. 12, pp. 22-23: “ut horum animadversione ad virtutem studia propensius exciterus.”
[24] I Trionfi di Francesco Petrarca, ed. C. Pasqualigo (Venice: Typografia Grimaldi, 1874), p. 65: “Trionfo della Castita,” line 119. Online in archive.org.
[25] Ibid., p. 109: “Trionfo del Tempo,” line 60 ff.
[26] Vatican Library Ms. Barb. Lat. 3943, f. 191r, at https://digi.vatlib.it/view/MSS_Barb.lat.3943/0336.
[27] Thierry Depaulis, “The Tarot de Marseille - Facts and Fallacies, Part II,” The Playing-Card 42, no. 2 (Oct.-Dec. 2013), pp. 113-114. Online in Academia.
[29] Franco Pratesi, “De l'utilité des jurons pour l'histoire du minchiate,“ As de Trefle No. 52 (1993), pp. 9-10, and “1499-1506: Firenze - Nuove informazioni sulle carte fiorentine,” The Playing-Card 44, no. 1, pp. 61-71, Both online at https://naibi.net. He also mentions a letter of 1466, the original of which is now lost. Both “minchiatar” and “trionfi” appear in a satirical poem of ca. 1440 Florence a few lines apart, a possible double pun; see https://forum.tarothistory.com/viewtopic.php?p=15174#p15174.
[30] http://trionfi.com/0/f/03/ (14 figures, 1441); http://trionfi.com/0/e/1 6/ (70 card trionfi decks, possibly 5x14, 1457). The sources are “Ferrarese account books” and Franceschini of n. 19, pp. 485 and 823 f.
[31] Dummett of n. 7, p. 188. The names given in a ca. 1450 poem, were karnöffel (meaning something like “Lout”), süw (Sow), babst (Pope), keyser (Emperor), and tüfel (Devil). Online at https://forum.tarothistory.com/viewtopic.php?t=1175.
[32] Franceschini of n. 19, p. 221. Reproduced and translated at http://trionfi.com/0/e/02/.
[34] Giovanni Boccaccio, Filocolo, Book 4 Chapter 96 (of about 1336), for rocco and cavaliere in the context of chess. Online at https://it.wikisource.org/wiki/Filocolo/Libro_quarto/96.
[35] Caldwell and Ponzi of n. 12, p. 106, from Pier Candido Decembrio, Vita Philippi Mariae tertii Ligarum Ducis (written 1447-1448): "Variis etiam ludendi modis ab adolescentia usus est."
[36] For the main orders, see Dummett of n. 7, pp. 399-401. Online at https://forum.tarothistory.com/viewtopic.php?t=1175.
[37] Franco Pratesi, “Studies on Giusto Giusti” (2012), section 6.1, at http://trionfi.com/giusto-giusti and http://naibi.net/.
[38] Il mondo e l’angelo (of n. 14), p. 168: “Un ’ipotesi plausibile.“
[39] Dummett of n. 7, p. 78 for the Modrone, p. 401 for the Lombard order. Both online at https://forum.tarothistory.com/viewtopic.php?t=1175.
[40] Moakley of n. 21, pp. 52, 74. Online at https://moakleyupdated.blogspot.com/2017/03/updating-chapter-3-by-m-howard.html.
No comments:
Post a Comment