New Pratesi note on Cary-Yale


New Pratesi note (now two) on Cary-Yale

The name Franco Pratesi is not unknown to this forum. The search engine gives 51 threads mentioning him. In the late 1980s he made several important contributions to tarot history, including finding the first document assigning divinatory meanings to the tarot subjects, in Bologna around or before 1750. He also brought the attention of the Marziano "game of the gods" to the tarot community. Over the past 28 years he has built up an impressive list of publications, both in print and on the web, as can be seen by viewing his website,

Recently he wrote a "note" of 19 pages on the Cary-Yale tarot deck, "Elucubrazioni sui tarocchi Visconti di Modrone o Cary-Yale", i.e. "Ruminations on the Visconti di Modrone or Cary-Yale Tarot". The Cary-Yale is of course the oldest extant tarot, and no stranger to this forum, with over a hundred threads mentioning it. The thread closest to the topic of this "note" is one in 2003, "Reconstructing the Cary-Yale", However Pratesi's effort--actually, in part something of a collaboration between him and me--represents a different approach.

I open this thread to discuss the issues raised by his "note". In the current post I am going to give some introductory material on the circumstances in which it was written, including a brief run-down of a few of his recent essays written in Italian and a very interesting email discussion we had about the meaning of "carte a trionfi", which is the form in which the term "trionfi" first appears in relation to a pack of cards, in 1440 Angieri/Florence ( Then I will offer a translation of the essay itself, which Franco wrote in Italian. My task is made easier by the fact that it includes some lengthy quotes by me in English.

First, you need to know about Marziano's deck.The deck itself has not survived, but several testimonies of the time exist, including a detailed description by the designer, Marziano da Tartona, written for Filippo Maria Visconti, the same duke of Milan he is assumed to have commissioned the Cary-Yale. Marziano verifiably died in 1425. It is tempting to wonder if there are connections between the two packs.

In Marziano's deck, 16 Greco-Roman "deified heroes"--i.e. gods and demigods, as described in various classical texts--arranged in a precise hierarchy could, when played in a trick, beat any card in the regular suits. At the same time all of them also belonged to the four regular suits, to which he gave the names of allegorical birds. They look like this:

Suit, Suit-sign_________ Gods, in order from most to least powerful
Virtues, eagle:__________1 Jupiter, 5 Apollo, 9 Mercury, 13 Hercules
Riches, phoenix:_________2 Juno, 6 Neptun, 10 Mars, 14 Eolus
Virginities, turtledove:____3 Pallas, 7 Diana, 11 Vesta, 15 Daphne
Pleasures, dove:________4 Venus, 8, Bacchus, 12 Ceres, 16 Cupido

To the extent that there is a hierarchy of 16 superior cards, this deck resembles the tarot deck, although the cards, from their description, looked nothing like any early tarot cards.

Second, you need to know about a few new essays by Pratesi on early documents using the words "minchiate" and "germini" (the latter thought to be another word for the same game, or one very similar). Before recently, it was thought that minchiate was a product of the 1540s. Earlier, it is true, there was only a not entirely secure mention of minchiate in 1466, in a letter of Luigi Pulci to Lorenzo Il Magnifico (F. Pratesi, The Playing-Card, Vol. 16, No. 3 (1988) pp. 12-15; the letter was first published in 1866, but the original got lost in 1956. But because of the 80 year gap between the two references, it was assumed they were to unrelated games.

Then the gap started narrowing. Franco found references to minchiate in the 1530s ('Italian Cards: New Discoveries", n. 5, The Playing Card, Vol 16 (1988), p. 78-83). He also found it included in a Florentine ordinance about games of 1477 (The Playing-Card, Vol. 19 (1990) pp. 7-17), and in another document, a conviction in 1471 for a crime in 1470 (for playing to close to a church: L'As de Trefle, N. 52 (1993) pp. 9-10). (All of Franco's essays are at http;// Also, Andrea Vitali found a reference to "Sminchiate" in c. 1510, in a context where it appears to be the name of a game ( That makes another reference to "Sminchiate", by Berni in 1526, more likely one to the same game, as opposed to a particular play in the game of Bolognese tarocchini, as Dummett had supposed (viewtopic.php?f=11&t=1019&p=15179&hilit=Berni#p15179). Then Huck found references to germini in 1517 and again 1518--actually 1519 in our way or reckoning, in which the year starts January 1 instead March 25 confirmed by Franco in an essay at

Then came Franco's 2015 finding: another reference to germini, in 1506 (The Playing Card vol. 44 no. 1 (2015), pp. 61-71). This last narrowed the gap between notes to less than 30 years.So the essay seemed to me worth translating.

I translated Franco's essay on the "germini" reference; you can read a corrected version at Soon Franco was emailing me about some of my guesses as to what certain 15th century terms meant. I corrected them. But one thing still puzzled me. He said that "carte a trionfi" in Florentine documents should not be translated as "triumph cards" but rather as "triumph-style cards". I wondered, what was the difference? He said that "triumph-style cards" meant cards done in the style of pictures with triumphal scenes or characters on them, for example the triumphs of Petrarch, or the cards described by Marziano (email of Dec. 20, 2015). Then would all "carte a trionfi" have been such cards? No, he insisted; it is 99% certain that only some of the cards were of such scenes. It is also only 99% certain that these were cards for use in card games; but such use is not part of the meaning of "carte a trionfi". He thought of an example. A "vaso a fiori" is, in Florentine speech, a vase with pictures of flowers on it. A "vaso di fiori" is a vase with flowers in it.

In English we have no such expression in three words or less for a vase with flowers on it. However we do have "checkered box" versus "box of checkers", with clearly different meanings, one referring to what is on the box and the other to what is in it. By that analogy, it seems to me, we could make up a term "flowered vase" meaning one with flowers on it. Another example is "deviled eggs", which means eggs part of which has spices added to part of it in a kind of paste: "deviled" is to suggest the spiciness of the result. Not all of the egg is "deviled", just the yoke. But the result is still a "deviled egg". On that model, "carte a trionfi" would mean, if we had the expression, "triumphed cards", i.e. cards of which at least some, and most likely only some, have been given triumphal scenes. They would not include only the triumphal cards, for which the term "carte trionfali" would have been appropriate.

So what were the antecedents of "carte a trionfi", an expression whose first documented use is 1440 Florence? Cards, to be sure, including playing cards, but also something else, triumphal scenes, such as those of Petrarch's poem I trionfi, or Marziano's cards. Later he added, in addition to these, "the same triumphal subjects that were used at the time for cassoni, deschi, and so on" (email of Jan. 1, 2016). An example that occurs to me is illustrations from Petrarch's De viris illustribus (Of illustrious men. He also once mentioned cards of saints.

This discussion reminded me of an old hypothesis of mine that I had not talked about in years. My idea was that the connection between Marziano's deck and the Cary-Yale was that both had 16 special cards that functioned both as the highest members of one of the four suits and as a hierarchy among themselves as far as which one, if two in different suits were played, won the trick. I had much difficulty getting this idea taken seriously as an hypothesis. The problem was that nobody believed that the Cary-Yale had such a structure. I gave them evidence, of a sort, but nobody thought it was credible.

Another part of my hypothesis was that the 16 cards included, as one part, the 6 triumphs of Petrarch's I Trionfi plus one more triumph.derived from Boccaccio.

In Petrarch's poem, Petrarch's Love for Laura is triumphed over by her dedication to Chastity, this Chastity in turn is triumphed over by Death, Death by Fame (i.e. living on in people's memory), Fame by Time (one's memory fades over time), and Time by Eternity (the Resurrection of the Dead). .Love, Chastity, Death, Fame , and Eternity are, on my hypothesis, surviving cards in the Cary-Yale. Chastity is represented by the card later known as the Chariot, which the Cary-Yale depicts as a lady on a chariot. Fame is a scene with knights below and a lady on top along with two trumpets, instruments traditionally associated with Fame.

The missing Petrarchan triumph is Time, which we know from the PMB (and Charles VI), where it is represented by an old man with an hourglass. It seems to me that there is one other likely missing card of this sort, derived from Boccaccio's version of the succession of triumphs in his Amorosa Visione, namely Fortune. I say this because Fortune is indeed one of the surviving cards of another deck closely associated with the Cary-Yale, the Brera-Brambilla. It also appears in the PMB, from the same Bembo workshop in Cremona.

Another part of the 16 would have been the four cardinal and three theological virtues of medieval Christianity, of which all three theologicals and one cardinal are in the surviving cards.

Finally there are the the Emperor and Empress, which form another group, of separate derivation; we do not need to know what it was because they are in the preserved cards and there are no other cards to postulate from that source. The game of "VIII Emperors" is one possibility (described at That is a game we know from a document of Ferrara, about a production in Florence, 1423. Franco speculates that it is another game in which the suits are extended, in this case with 2 emperors per suit (; see the table there).

This part of what I presented to Franco is from Huck's "chess theory" of the Cary-Yale, which he presented in 2003 ( Later he developed a pictorial version, at As he wrote it in 2003, there would have been 16 triumphal cards that corresponded to the 16 chess pieces, the 7 virtues plus Love as the pawns, the Empress and Emperor corresponding to the chess King and Queen, the Judgment and World cards corresponding to chess's rooks (because they both had trumpets on them, and the Chariot and Death cards to the chess Knights , because they had horses on them. What corresponded to the chess Bishops wasn't clear, because those cards were missing. The Pope and the Popess were the possibility he suggested, an attractive suggestion because then each pair would have one male and one female representative, corresponding to "queen's side" and "king's side" in chess.

The difficulty is to assign Petrarch's Time to both a chess piece and a card, such that it would pair up with another chess piece and card, the 16th card, the one I assign to Fortune. I assign Time and Fortune to the bishops, on the grounds that both cards, in their earliest representations, have old men on them, and bishops were customarily senior church officials. My Cary-Yale reconstruction has no Pope or Popess. (Below is the PMB "Old Man", with his hourglass, the BB Wheel, and the PMB Wheel.)


I had the idea from somewhere that Franco did not like Huck's "chess theory". But I thought maybe if the non-chess aspects, namely the "triumphs" of Petrarch/Boccaccio and the seven virtues, were put in a different context, namely that of four groups, from Marziano, Franco might be interested. So I referred Franco to one short section of an old blog of mine (originally 2008, partly rewritten 2012; it is at, with the relevant part the section "The Cary-Yale in Relation to Michelino and Petrarch's 'Triumphs'"; it connects Marziano's deck (which I called the Michelino, from the painter known to have done the work) with the Cary-Yale.

Franco immediately started writing a new "note" (I'd call it an essay; it's 19 pages long). He showed me drafts and tables for my response. He didn't accept everything I said, and developed his own version based on minchiate (which I agree would be connected), but he did find my proposal stimulating and not something to be rejected. Franco works fast. He was done in a week.

In my next post I will present the first half of his essay, in my translation (with a little help from Franco); I will present the second half after that. The "note" is on in Italian (except for a couple of pages by me), the second note of 2016, starting with the word "Cremona". For the Cary-Yale itself, see First you click on "view all images", underneath the card displayed. Then when you get to a triumph, if you click on the image, you will see more information, including the part that interested me.


Now I am ready to begin posting Franco's "note" on the Cary-Yale. Words in brackets are from me, either giving the Italian original or some short explanation. I include Franco's page numbers and breaks between pages, in case someone wants to refer to the Italian. The oriiginal essay is at I didn't have a lot to do with this part of Franco's "note".

Ruminations on the Visconti di Modrone or Cary-Yale Tarot

1. Introduction

The tarot pack discussed here is part of the Milanese Visconti-Sforza tarots [tarocchi] illustrated and discussed in countless books and articles, often at odds with one another for attribution, date, and interpretation. So much attention is justified by the extraordinary workmanship of these precious cards and even more because they represent the principal ancient specimens of tarot cards that have been preserved. The pack in question is that which, out of all of them, presents the most puzzles to be solved in order to seek a convincing understanding. The name of Visconti di Modrone comes from the Milanese owners who possessed it for generations; the name of Cary-Yale, now more used internationally, comes from the American Cary family who purchased it from the Visconti di Modrone, from whom Yale University acquired it, where it is now preserved in the Beinecke Library in New Haven.

The present contribution does not come from the discovery of new documents, but only from reflections on what is presented and how the pack could have originated; the discussion will proceed with some deviations and parentheses, in a nonlinear manner. The credit for this study (but one could say the blame) is Michael Howard, who has stimulated and assisted in it. In truth, that assistance would have more necessary and more useful if I had the opportunity and desire to write a big book on the subject, rather than a short note. I must acknowledge in this regard, and make the reader aware, that the bibliography on this theme, including discussions on the internet, is vastly more extensive than that used and cited here.

2. The Courage of Sylvia Mann

The figure of Sylvia Mann has been fundamental for historical research on playing cards. The fact that she found herself working with an author of the caliber of Michael Dummett had as a consequence

that she soon took a back seat, and her enormous contribution to the material could not be recognized.

From an organizational point of view, Mann was the person who brought life to the IPCS [International Playing Card Society]and to its official organ, The Playing-Card, a journal that continues to be published today after 40 years. In that journal she was proud to offer a secure reception for studies on the subject, not easily published in academic journals. Personally I owe a great debt of gratitude for her encouragement to continue my research, by including card games along with board games, chess at first, whose literature and history I studied for years. It was she who gave the title “Italian cards - New discoveries” to the series of my articles, and assisted me more than once in their revision, also from the point of view of language.

Her most important contribution was in my opinion to fix with precision and force a line of demarcation among playing cards, one very useful for further research. Of playing cards Mann was primarily a collector (and I seem to remember also stamps before, like many other people). What is usually meant by an "object of collection" [oggetto da collezione] in general, and thus also in the particular case of playing cards? If it is a postage stamp, what is considered "of collection" [da collezione] is not the most common one that you can put on a letter every day, but an unusual specimen, a commemorative, striking precisely because it is unusual, even before its possible beauty.

One can also go back to the Schatzkammer, or treasure chambers of the princes, with their precious objects, as many extraordinary ones as possible, able to fascinate any observer. As always, an "object of collection" is in fact an object out of the ordinary, which never or almost never occurs in daily life except, possibly, in the very poorest versions. For playing cards there is the same reasoning: they deserve so much more the name "cards of collection" ["carte da collezione"]as they are different from those that can be bought in local stores and used in the family or in traditional games with friends. With cards of collection it is instead probable that no one ever played with them; for the collector, there are still artists and publishers who make special packs, and among them you can even find whole types of packs: travel, advertising, erotic, fantastic, round, and so on.

Mann instead taught all collectors of playing cards - or at least all those, perhaps few, who have assimilated her lesson - that there was a different way to find the extraordinary in cards of collection [carte da collezione]. Mann’s revolutionary proposal was very simple: all the "ordinary" cards can and indeed must become "extraordinary", collectible: it is sufficient to leave the familiar framework and procure ordinary cards of distant countries and times! Indeed, on closer inspection, precisely those cards are to be collected and then studied in their historical and geographical evolution. As an example it can be very revealing for us to consider some strange Japanese playing cards, that with a little attention we can succeed instead to understand not only how ordinary they were, but also clearly how they are derived from the cards, also ordinary, of the Portuguese. In short, Mann has deserved much more than my personal recognition; all the historians interested in this material owe her gratitude. We will see shortly how this parenthesis is not so far outside the subject as it might appear.

3. Experimental Character of the pack under study

The pack in question presents itself as a unique example among the preserved ancient tarocchi, not so much for its workmanship or style, as for the figures on the cards. Already the pips [carte numerali, number cards] are not the usual ones, for example, with the usual staves or scepters, but here represented by arrows; also among the triumphal cards [carte trionfali] appear unusual ones. Perhaps even more characteristic is the fact that in addition to the queen, there are other female characters [personaggi] among the court cards [carte figurate]. Alongside the male pages [fante, which the Beinecke translates as I have given] are the corresponding female pages [fantine, again following the Beinecke]. That happens in other cases, starting with minchiate; but everyone knows that minchiate was introduced later, and above all, in those cards the two female pages take the place of two missing male pages, while in the CY pack there are both the one and the other. Not only that, but here the knights, too, have beside them their feminine counterparts, not in substitution but in addition, and this seems precisely a unique case among all the playing cards, which deserves a separate reflection.
1. S. Mann, Collecting Playing Cards, Wimbledon, 1973.
2. S. Mann, V. Wayland, The Dragons of Portugal, Sandford 1973.

In conclusion, all the evidence leads us to consider the CY pack as a pack of collection, from two points of view: it is obvious that today it is because of its age, rarity and beauty, but also, it looks so unusual that at birth it must have been an object to impress anyone who saw it. You can then take into consideration the great teaching of Mann: with all its beauty, being an unusual pack not intended for a traditional game, it certainly deserves the attention of historians (not for nothing have thousands of pages already been written on cards like these), but, precisely because of its extraordinary nature, it can bring much useful information for our reconstruction of the history of playing cards and their development. In the case of the CY pack, however, it cannot be excluded that it was instead a precursor to a triumph pack [mazzo di trionfi] that in its final form did not precisely exist; in which case, by the same criterion of Sylvia Mann, it would appear of enormous historical interest, while remaining an experiment, since it would be pointing toward a pack which at this point needed only to establish its standard characteristics.

From the above it is evident that for the history of playing cards it is essential to propose as accurate a dating as possible for the CY: a difference of a few years can transform it from an insignificant variation on a well known theme to a pioneering experiment destined for a great future. To resolve the problem, the skills of historians of playing cards do not seem sufficient. To be convinced, it is sufficient to read what was written in this regard by one who can be considered the greatest of all (3)
It is impossible to determine if the Visconti di Modrone pack was an isolated experiment, which was detached from a standard already established, or if it is the only surviving example of a primitive stage in which the tarot pack had not yet acquired the structure that was to later become canonical. If it does represent a primitive stage, it is also impossible to determine if it was of a time when there coexisted significant variations in the composition of tarot decks or one in which a precise standard prevailed, different from what was observed later. One hypothesis, advanced, for example, by Dr. Algeri and Mrs. Gertrude Moakley, can be excluded with certainty as completely anachronistic, and that is that it was a Minchiate pack.
3. M. Dummett, Il mondo e l’angelo, Napoli 1993, p. 52. [original in Italian]

Michael Dummett having passed on, we can turn to another scholar of a high level, Thierry Depaulis, whose last book updates with simple precision much of our knowledge about the history of the tarot (4). Here is what we can read on the CY deck:
The first [Visconti di Modrone] has only 67 cards, including eleven triumphs [atouts], but offers some unexpected pictures [figures], male and female knights, maids and valets, and the three theological virtues, Faith, Hope and Charity, which are not normally part of the series. This atypical tarot could be a kind of trial run, especially the presence in two suits - batons (here, in fact, arrows) and swords – of emblems of the Sforza (the fountain and the quince, mela cotogna) while the other two - coins and cups – bear the emblematic of the Visconti, seems to be explained by the union of two families that the Love card could represent. Only one possible date, 1441, when Francesco Sforza marries, at Cremona, Bianca Maria Visconti, the only child, natural but legitimated, of duke Filippo Maria. It would then be the oldest preserved tarot deck.
The attribution seems reliable to me, because we know that Cremona played a significant role in the production of triumphs in Lombardy; but some conclusions must still be drawn, and some contraindications are still found. A particularly significant one is that very same Thierry Depaulis, also with the book just quoted, has brought to the attention of playing card historians the quotation from the Diaries [Giornali] of Giusto Giusti, in which a triumph pack [mazzo di trionfi] was produced in Florence in 1440 for use in and around Rimini. But if in 1440 "normal" triumph packs already exist, it must be deduced that the CY pack, precisely because of its exceptionality, coexisting among objects of more common use, is of secondary historical significance. If we think instead of a prototype destined to obtain shortly afterwards considerable success in a normalized form, it is necessary to go back to earlier dates, like the year 1428 supported by others. Ultimately, the discussion on the subject is not closed, if we find traces recurring until the last days.

I tried then to seek an answer in one of the forums on the internet that are dedicated specifically to the subject 5. In this case the search for contributions on the subject is easy and assisted by powerful search
4. Th. Depaulis, Le Tarot révélé, La Tour-de-Peilz 2013, p. 20. [original in French]

engines. The problem is that if you insert “Cary” you get in response: "The following words in your search query were ignored because they are too common words", while if you insert “Cary Yale” the answer is: "Search found 400 matches". There was a time in which the second answer would make me very happy, but now even 40, rather than 400, would be too much; at least for me, and at this age, not only the verba volant, but also volano these scripta [allusion to the Latin adage verba volant, scripta manent: “spoken words fly away, writings remain”].

Let us then try to consult different historical skills, those of the historians, who have devoted considerable attention to the Visconti-Sforza tarot.

4. The questionable discussion of art critics

There is, to my knowledge, no academic discipline with the name “History of card games”, and thus as specialists in the history of card games and playing cards are met only amateurs, with very few exceptions. But as for the history of art, there are many chairs, academics abound, and their writings fill libraries; thus a history of playing cards has everything to learn from experts on the history of art. From the foregoing it is evident that the problem of the dating of the CY pack, which has been shown of enormous importance, one could reasonably expect to find already solved in the writings of historians, who also have paid recurrent attention to this topic.

Unfortunately, the contribution of the art critics does not resolve the problem, but might be said to complicate it more. It seems that for an art historian, the dating of a work of this kind can vary in intervals so large as to render unnecessary and unreliable their contribution, at least until a proposal becomes really more convincing than all the others, which to everybody 'today would not seem accomplished. In this regard have to mention at least one recent publication on the Visconti-Sforza decks, which unfortunately leaves behind the scenes the specific deck at issue here, the CY; however, the treatment of data and the bibliography should still prove useful to anyone wishing to deepen [one’s study] (6).
6. S. Bandera, M. Tanzi (eds.), Quelle carte de triumphi che se fanno a Cremona. Milano 2013.

Among all the stylistic particulars and details in the depictions, none drew the attention of the experts more than the alternating coats of arms on the tent in the background of the Love card; while one is definitely Visconti, the other was variously attributed to the House of Savoy, the city of Pavia, and perhaps to other noble houses. The importance of the attribution is linked to the hypothesis (shared by the majority of the historians) that the deck had been produced on the occasion of a marriage between a Visconti and a Lady X, with its related dating easy to find. Usually, in the case of assignment to the House of Savoy, the wedding would be in 1428, between Filippo Maria Visconti and Marie of Savoy, but even this is uncertain because other critics argue that it is indeed the House of Savoy, but the marriage would be in 1468 between Galeazzo Maria Sforza and Bona of Savoy. The distance between the two cases is such that our request for help is unmet.

[Translator's note: Here is the middle section of the CY Love card, with the coats of arms on top:


The Visconti coat of arms is the viper with the red man in his mouth.]

A different discussion was of the clothes of the characters depicted on the cards. According to some historians the CY deck should be considered prior by a few decades to the other Visconti-Sforza decks. in order to justify the significant difference in the style of clothes. Again art historians have not reached agreement, and in any case the for playing cards it is commonly known that dating the epoch of production on the basis of the clothing of the characters leads to the possibility of big mistakes.

When, as often happens, we are in the presence of historians who do not know the history of playing cards, we find ourselves faced with curious attributions, like that of considering the deck CY a kind of minchiate; it is easy then for a Dummett is to counter, as in the quote above, that minchiate was of an epoch still to come. It is a situation that I know well, because I found myself there, bogged down in my first encounter with the comedy by Notturno (7); also then it was minchiate that appeared before being invented; it is a bit as if one found a manuscript of the Gospels dating from the first century BC ... Nevertheless, the appearance of the female page in the CY deck, and other details that are not secondary, such as the presence of the theological virtues, the interweaving of the swords, and others, cannot but recall the minchiate pack.
7. F. Pratesi, The Playing-Card, Vol. 17 No. 1 (1988) 23-33


Here is the second half of Fanco's essay, "Ruminations on the Visconti di Modrone or Cary-Yale Tarot", from the Italian at We are at the top of page 8. My contribution begins in section 8. In the conclusion, there was one sentence I am not sure is translated right; I think it is, so I give the Italian in brackets.

5. Connection with chess.

Recently the triumphal cards in the CY pack have been put in relation with the figures of chess, and it has fallen to me to discuss this proposal, advancing reservations (8). However, it is precisely for the CY deck that it is inevitable to recall some reference to chess, even before considering the triumphal cards. In order to advance interpretative hypotheses about the preserved cards, let us try preliminarily to proceed in reverse, trying to "build" a deck of cards on the basis of the pieces on the chessboard. A premise is necessary as to what historically was the case when chess as well as cards passed from the Islamic world to the European one. Originally there were no women on the board, or even civilian characters; they were all soldiers of various degrees around the king. Soon, however, the army of chess in Europe became a representation of the environment of the court, with the queen by the king and then the judges [giudici or bishops [vescovi] and knights [cavalieri], with a maximum of eight pawns [pedoni, also meaning pedestrians] still with the function of soldiers.

In playing cards there was a similar transformation, although later, because cards came after the transformation had already taken place in chess. Before, they were king, senior officer, and junior officer; then they became king [re], queen [regina] and jack [fante]. For the new pair of king and queen, the situation is closely analogous; for the other chess pieces, in playing cards we can accommodate a limited number: only one in normal cards, only two in the tarot type, while in chess there would be three (bishop, knight, rook [alfiere, cavalo, torre]). Not only that, the three characters that accompany the pair of king and queen are in turn couples, i.e. 2x3 pieces, which would require six cards and then eight including the royal couple. (A further problem is encountered if the couples are separated: including the royal pair, there are four, while the individual pieces in chess are not four but five, for the necessary distinction between the king and queen.)

Even today in chess, the Queen's side is distinguished from the King's side; we speak, for example of the queen’s bishop or the king’s bishop. It is not a long way to get to a change of sex of the pieces initially located to the left of the white queen or right of the black queen! So, if you want to fit a deck of cards to the typology of chess

pieces, the easy solution is to construct a deck with eight pip cards corresponding to the pawns and eight court cards corresponding to the chess pieces initially in the first row of the board. It should be borne in mind that a pack of cards has four suits instead of the two sides present on the board (so that historians like Rosenfeld have argued, with little success toward the truth, that chess in four was the primitive form and had the greater influence); thus we get a hypothetical deck of 64 cards, twice as many as the chess pieces.

6. Adding the triumphal cards

In the search for a hypothetical construct a priori of the CY type pack one must proceed with the necessary additional cards with the function of trumps. For these cards it is not even easy to find a suitable name, because the name “triumphs” [trionfi] was used mainly for the entire deck, and the name “tarot” [tarocchi], which was initially adequate, soon passed into indicating, similarly, the full deck. To avoid confusion, the terms “triumphal cards” [carte trionfali] or “higher cards” [carte superiori] might be used. How can these new cards be characterized? As mentioned above, even for these cards an origin has been proposed associated with chess pieces, but I cannot be convinced of the validity of such a proposal. In my opinion, a derivation from chess can serve for the court cards and possibly for some of the higher ones, but not for the whole triumphal series [serie trionfale].

We would be in complete darkness regarding any proposed hypothesis were we not provided the schema of the higher cards designed by Marziano a few years earlier, again in the same environment of the Milanese court, perhaps in collaboration with Duke Filippo Maria Visconti. The number of 16 triumphal cards is also in accordance with the ratio of 1:4 with the other cards in the deck, which can be found in decks of succeeding Bolognese triumphs (9). With this background, it is not difficult to reconstruct the hypothetical full deck. We need "only" sixteen other cards and also, for their type, certain requirements that have to be met, especially two.

The sixteen additional cards must first of all correspond to a triumphal series, with a hierarchy that presents itself logically and easy to remember, and perhaps if possible organized like the series of the Trionfi of Petrarch, with a clear justification for why each triumphs over the previous card and is overtaken by the subsequent. If the succession in trick-taking power is not clear, it would be necessary to add the sequential numbers on the cards, as was done in the tarot later. The requirement indicated may be the only one necessary, but if following the example of Marziano is desired, there is a second to be fulfilled: the sixteen additional cards must also be able to be situated [dovrebbero essere situabili] into four groups of four triumphal cards, each at the head of one of the four suits of the 64 card deck identified earlier.

7. Discussion of the pips and courts.

One can proceed to a first comparison between the hypothetical deck imagined before and that of the CY. As for the pips in the deck the CY contains thirty-nine of forty, and nothing suggests that the one missing card (the 3 of Coins) was originally absent. So, compared to the eight pawns of chess we definitely have two extra cards. One might think to take as a model chess existing on 10x10 board, but accommodating the pips in that way departs even further from an accord for the court cards, so you have to find “another way.”

The court cards in the CY deck are six per suit, and this is a real record compared to the three in ordinary card decks and the four usually present in tarot decks; the number of six courts in each suit, although unusually high, cannot suffice to establish a one to one correspondence with the major chess pieces. If we want to insist on the usefulness of the comparison, or the opportunity to find it, there remains the possibility shown in Fig. 1, in which the cards with the numbers 1 and 10 have taken the place of the rooks in chess. (With the final F is indicated the feminine correspondents of the Male Page [fante, in the diagram PM] and Male Knight [cavallo, in the diagram KnM]. To respect the chess positions each P should trade places with a Kn, but here the hierarchy of cards is being respected.) As weak support for the hypothesis in such circumstances, it is possible to advance a few considerations. From the side of chess, it is not easy to Associate

the members of the court to the rooks as it is to the other figures, so much so that also historically, various types of associations have been proposed, and often rooks were artifacts of defense and not characters [personaggi]. [Translator's note: here I have replaced the Italian abbreviations for the chess pieces and cards with their English equivalents.]


On the side of the cards, it can in fact be recalled how in many card games the ace and the 10 had special roles. It is true that in the current Italian game of briscole [trump] the 10 is not used (of course, in the normal deck it no longer exists and was replaced in that function, oddly enough, by the 3); however, it is equally true that in many foreign trick-taking games with trumps it happens that the ace is in fact the highest card and 10 the second, followed in order by the courts. You can push beyond the association with chess, considering the power of the court cards, given that at the time, before the spreading of modern chess with the new queen, precisely the two rooks were the most powerful pieces on the board (like cards 1 and 10 in games of trumps).

In conclusion, a way can be found to move from a fairly logical structure of these number and figure cards and from the obvious type 8 + 8, as in chess, into a type 10 + 6. Fig. 2 shows, according to the diagram, all the number cards and figure cards still present in the CY deck.[Translator's note: In the Word document version of this table, which I used to put in the English abbreviations, the top quarter slid to the right from where it is in the pdf online. I was not able to fix this problem, and I don't think it matters much.]


8. Discussion of the triumphal cards

The sequence of preserved triumphal cards in the pack is the following: World, Angel, Death, Chariot, Charity, Hope, Faith, Fortress, Love, Emperor, Empress; at least so their order is indicated

in the book cited by Michael Dummett. There remain to be discussed in particular the two highest positions in the Milanese environment, which appear consistent in that order when compared with succeeding packs of the same origin. However, it does not seem that this series is a complete sequence of triumphal cards, although in principle it could be. Dummett has proposed various lengths for the hypothetical original sequence, considering also the practical convenience of maintaining a relationship between the ordinary cards and the triumphal ones that is not too different from what exists in known complete packs.

But we have a guideline to take into consideration for the triumphal sequence, provided by the system described by Marziano with its sixteen deified heroes. Here the characters are clearly changed and already incomparably closer to those that are commonly found in various types of tarot. The only rather unexpected presences are the three theological virtues, which in the tarot are present only in the Florentine (of which, however, it is believed that they were still a long way from birth). In place of the gods of Marziano one can try to insert other cards in the sequence but supposed lost in the original pack, being guided by the Trionfi of Petrarch, as well as other known sequences. The "solutions" may be different, but my attention was drawn by the proposal of Michael Howard, presented in an updated form in the table of Fig. 3, after being presented by him and discussed in past years (10).



In the second column of Fig. 3, the cards supposed lost are listed in italics and with only one capital letter; the content of the third column will be explained later. The procedure followed for the reconstruction is described by the author.
On the Beinecke Library website the cards are divided into four groups, which correspond almost precisely to the four suits. They are in the following order, with the captions as given:

Swords: Empress of Swords, Emperor of Swords, Love (Swords).
Batons: Fortitude (Batons), Faith (Batons), Hope (Batons),
Cups: Charity (Cups), Chariot (Cups), Death (Cups)
Uncaptioned, Uncaptioned

The last two are first, a scene with knights and castles usually designated “World” (Mondo), and finally, one of the Last Judgment, corresponding to the card known as “l’Angelo” in later lists.

When I emailed the Beinecke about this arrangement (Aug. 25, 2008), curator Timothy Young replied, “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 catalogue to the Cary Collection used their descriptions when he created fuller catalog records.” We may wonder how far back these descriptions might go. Nonetheless it is a start. Its principle of division,
11. M. Howard, Personal communication, Jan. 15, 2016

or even why there is one, is not obvious, simply from looking at the cards or from knowing what came later.

First it is necessary to determine what cards are missing. The existing cards correspond to 5 of the 6 Petrarchan triumphs and 4 of the 7 principal virtues of the medieval Church, the 4 cardinal virtues and 3 theological. The missing Petrarchan triumph is that of Time, and we see that in later packs with an old man (Vecchio), holding an hourglass. Together with the Empress and the Emperor, that would make 15 cards. However in the Brera-Brambilla pack, done just a little later than the Cary-Yale according to current thinking, there is a Wheel of Fortune. This also was one of the triumphs in Boccaccio’s Amorosa Visione. That would make 16.

These cards must then be inserted in their proper places. In Marziano’s pack, each of the suits was associated with a bird and an allegorical theme: eagles had the theme of Virtues; phoenices that of Riches; turtledoves, that of Virginities; and doves, of Pleasures. So the CY’s organizing principle might be the four cardinal virtues, each related to three other of the cards. How the cards would follow each other within the groups cannot be inferred precisely, except by reference to the Beinecke’s order, but probably the cardinals would be either at the beginning or the end of the group they dominate. Since the Angel is a fitting climax for the sequence, and Justice was never, in the later lists, put before the Empress, it is possible that Justice was at the end of its group and Prudence at the beginning of its group. If the Beinecke order is strictly followed, Fortezza would have to be at the beginning of its group, while Temperanza could be at either end. After Death is where it is in later Milan lists; so, somewhat arbitrarily, that is where it will be put.
Personally I find it reasonable to set the total number of triumphal cards at sixteen, but I do not see why one cannot rely directly on the minchiate sequence for the reconstruction instead, with the same elements, a different sequence; why not use the one of minchiate if, among other things, precisely the theological virtues are present there? Once we admit the comparison, the match is found immediately. So my proposal, alternatively, is that of Fig. 4, where among other things the order of the two highest cards is reversed against the list of Dummett, supposing that in Milan or Cremona the Florentine order was still in force.


Now I cannot write that then minchiate existed (nor am I convinced regarding the final form of 97 cards), but I cannot also rule out that some kind of Florentine triumphs already existed and would serve as a model for any reproductions and variants.


9. Subdivision of the triumphal cards into suits

With all the uncertainty of the case, the previous point can at the same time be considered a starting point for a further step. Again having the pack of Marziano in mind, it is natural to consider the possibility that the sixteen reconstructed cards can also be seen, in addition to the sequence of numbers 1-16 shown, as formed by four groups of four cards each connectable to one of the four suits of ordinary cards. Though I have tried with the series that I have proposed, I was not able to select four groups in the desired manner; so I give up further attempts in that direction. However, I can continue examining the proposal of Michael Howard, who in some way has managed to carry through for the stated purpose the subdivision into four suits. To achieve this result, Howard used an association of the triumphal cards to the four suits that is present in the archiving of the museum and the documentation that comes from the Cary family, joined with the same pack at the museum. Again in this

I can use, with some relief on my part, the author’s description (12).
The third column [of Fig. 3] follows the Beinecke groupings and order within groups, with the additions.

The character of the four virtues fits that of the four suits: Justice has a Sword; while the “slings and arrows of outrageous fortune”, as Hamlet put it, require Fortitude, an unbending will, like a lance or arrow. The Temperance card has its two cups. The round Coin suggests the Sun, a symbol of God, as well as reward for right conduct.
The allegory would then somewhat as follows. What the Empress and Emperor need most is Justice, because of who they are. Love also must be guided by Justice, which requires respecting the wishes of the beloved. But faced with the apparent misfortune, represented by the Wheel, of Laura’s wish for Chastity he first needs Fortitude; and in general, when faced with the Wheel, one needs also Faith and Hope. These virtues do not triumph over the one preceding, but build on one another. Their nature as triumphal cards is in their victory over vice. In numerous illuminated manuscripts, they can be seen stomping the corresponding vice underfoot, usually represented by a particular personality whose name was written in. The CY theological virtue cards are the same. Under Hope’s feet, with a rope around his neck, is Judas, who had that virtue’s opposite, Despair. Under Faith and Charity are two crowned figures lacking those virtues, probably the heretic Muhammed (parts of the first letters are legible) for the first; for the second, Herodus would have been usual (see Leone Dorez, La canzone delle virtu e delle scienze di Bartolomeo de Bartoli da Bologna, Bergamo 1904, p. 82). [Translator's note: the three theologicals can be examined here: An example of virtues overcoming vices is the top row here:]

With Temperance – control over the passions – comes also Charity toward others. All these virtues lead to the Triumph of Chastity, the woman on the CY Chariot card. Moreover, it is Temperance that in this life can delay Death; and her cups in another sense – that of the spirit of the Eucharist. can even overcome it. For the last group, Prudence is the governing virtue. The ability to choose the right means to the right end (Aquinas, Summa Theologiae, IIa. q. 57, art. 5, ans.). – Prudence guiding all the virtues – leads sometimes to renown in this world, which inevitably fades over Time; but with God’s mercy, leads to glory on Judgment Day.

This kind of allegory is of the type we can imagine Marziano writing to Filippo. To remember the sequence, it is enough to remember the allegory and the placement of the cardinal virtues. The story does not have to go in just this way; this is merely one that respects the groupings and order in which the cards were given to Yale.
The result is presented in the third column of Fig. 3. In that table the sixteen cards are actually divided into four
12. M. Howard, Personal communication, Jan. 15, 2016.

groups, associating respectively the first (1 to 4) to the suit of Swords, the second (5 to 8) to the suit of Batons, the third (9 to 12) to the suit of Cups, and the fourth (13 to 16) to the suit of Coins. The four cards associated with each suit somehow maintain their hierarchy, representing a sort of continuation of the court cards, in such a way that the complete pack could be seen - in a manner not dissimilar from the pack designed by Marziano - in the two alternative ways schematized in Fig. 5 for the structure into five suits and in Fig. 6 for the one in four. (In the two figures a peculiarity of tarot games, and also of other trick-taking games, was not taken into account; so the sequence of number cards 1-10 would be correct only for the "long" suits with Batons and Swords, while it would be the opposite, in ascending numerical order, for the "round" suits of Cups and Coins.)


It is not at all certain that the attempt is successful, however, as we know from Marziano, who shows us that in the early days there was actually a possible interchange between the higher court cards and the triumphal cards. [Translator's note: For example, Franco says in discussion, the Emperor and Empress might have derived originally from the King and Queen.] It also appears likely that there were then various uncertainties in the separation between the two groups, with some higher

characters originally belonging to a suit and then breaking away definitively.


10. Conclusion

For the Cary-Yale tarot (CY, also Visconti di Modrone), various options for reconstructing the complete pack were analyzed, possible logical paths that could have led to that particular structure; in particular, of triumphal sequences there were considered those known from the Trionfi of Petrarch and from other ancient packs of triumphs, including in particular the pack of Marziano, and possible associations with the figures of chess. Once a hypothetical pack with the requirements needed was reconstructed, it was determined if and to what extent the CY could be associated with it. [Una volta ricostruito un ipotetico mazzo con i requisiti richiesti si è controllato se e quanto poteva essere conforme al mazzo CY.] There were also presented a couple of reconstructions of the triumphal series considered probable, hypothesized as

sixteen cards of which five have been lost. One of these [reconstructions] is based on the structure of Florentine minchiate, which at the time had not yet been introduced, at least not in the form known later. There was also presented a proposal for division of sixteen triumphal cards into four suits. Several of the aspects discussed seem to favor the interpretation of the CY deck as a precursor of packs of standard triumphs rather than a variant of a type of pack already in common use; but on this point, of great importance historically, significant progress unfortunately has not been made.

OK, it's open for discussion.


Well, it's been 23 hours since I posted the invitation for discussion. It is a meaty piece, not easy to digest. I of course have plenty to say about it. Perhaps if I say some things it will stimulate somebody else. I will start, somewhat arbitrarily, with the chess-tarot question, not in general but in relation to the CY exclusively, assuming the missing cards that both Franco and I agree on (and also, except perhaps for the Wheel, Huck as well).

Some post-Pratesi ruminations on the chess-tarot relationship.

Analogies are just that--similarities, among differences. There are of course major differences between the chess pieces and the CY, as well as between the two games; but there are also similarities.

The first question, I think, is: what is the point of drawing such analogies? I was amused to discover that Gareth Knight, in a book called Tarot and Magic, used the analogy to chess as part of an explanation for why certain cards, in particular the pages as the equivalent of the bishops in chess, had certain divinatory meanings (p. 194, tarot&f=false). Even if true, and probably it is, that is at a later stage. In the case we are concerned with, the CY, chess analogies are thought of as having something to do with explaining why the pack is what it is.

So for Franco there is the question, why are there six court cards per suit in the Cary-Yale? He doesn't pose this question directly; but it seems to be what is involved, because he makes a point of there being a male side and a female side in chess, king's side versus queen's. So the chess analogy could explain why the Cary-Yale is the same. For such a purpose, it seems to me that it is not necessary to carry the analogy out to the bitter end, i.e. rooks being analogous to the Ace and the 10. In the game of tarot, aces and tens are just the highest and lowest number cards. That is a poor analogy to rooks in chess. If there are better analogies in other games, the chess analogy might explain the peculiarities of Aces and 10s in those games; but we are concerned with tarot. But as I say, nothing hinges on how convincing the analogy to rooks is. What matters is simply that we have two major groups in the suits, courts and pips, which correspond loosely to the first and second rows of chess pieces. and a male and female side for the CY courts, corresponding to the same for the chess pieces.

That may or may not be the explanation for why there are twice as many courts per suit, a female for every male, in the CY compared to ordinary playing cards. We don't know enough about what ordinary playing cards looked like then. From 1377, Johannes of Rheinfelden describes, besides the most common deck with three male courts, some variants with females (Game of Tarot, 1980, p. 12L):
...the text goes on to list a number of variants; one in which all the Kings are replaced by Queens; one in which two of the suits have Kings and the other two Queens; one with five suits; another with six; and, finally, one with four suits, but with five court cards in each suit (King, Queen, the two Marshals and a Maid), making 60 cards in all.
We also know that the Stuttgart Playing Cards, c. 1430, had two suits with three male courts and two suits with three female courts (Husband, The World in Play: Luxury Cards 1430=1540, 2016, p. 26). Below, the Under-dame of Stags, identified as such by Cloisters curator Timothy Husband in The World at Play: Luxury Cards 1430-1540 2016, p. 25:


Later on there were decks that had female pages in two suits and male pages in two suits; the female suits had Ace high and Ten low while the male suits were the other way around.

So even without thinking of chess, it is a very small step from male suits and female suits to suits with both males and females. It is even easier if there is another way of symbolizing male and female polarities between suits; in the CY, the usual male suits have what appear to be Sforza emblems, and the female suits Visconti emblems. No chess analogy is needed. It might have played a role in taking the step to combine male and female in the same suit--or not.

When we get to the triumphal cards, as Franco calls them, there is indeed the analogy that Huck points out. But here there is a difficulty. In Franco's case, the King in chess was analogous to the King in cards, and so on. The only mildly problematic card (excluding the Ace and Ten) was the Page, which is somehow analogous to the bishop. In most of Europe, it isn't called a bishop. In Italian, the word for that piece, alfiere, means "standard-bearer" or "ensign". In Spanish, it is alfil, probably deriving from the Italian, since the word has no other meaning. In German it's läufer, meaning "runner". It is a wide-ranging piece, able to move as far as it likes as long as it has an unobstructed path. In English, a page is an apprentice knight, one of whose duties is to deliver messages. In the U.S. Congress it is a young person who delivers messages for congresspersons. Gareth Knight observes that this capacity might have given the corresponding cards some of their historical divinatory meaning of "news" or "letter". And the queen could have a female delivery-person as much as the king could a male.

In comparison to the courts, however, most analogies between chess pieces and the postulated triumphal cards of the CY are vastly weaker. What do pawns have in common with virtues, other than that they are numerous? Pawns are lowly, virtues godly. Pawns are sacrificed for the sake of a greater good. Virtues when sacrificed may yield temporary gain but never a greater one in the long run. You can look at a pawn as long as you like, and it will never turn into a virtue. It is only if you already have in mind the seven virtues--usually depicted as similar-looking ladies--that you would think to correlate them with seven of the pawns (and hope you have something comparable for the eighth); it is an afterthought, a parallel once you already have the virtues in mind for the cards and are willing to ignore the dissimilarities.

Petrarch did have pairs of triumphs, bad triumphed by good. Bad carnal love was triumphed over by good chastity. Bad death was triumphed over by good fame. Bad time was triumphed over by good eternity. None of these are pairs in the "chess analogy". And as described by Petrarch, the chess analogy's pairs have little in common visually (the poem is at Take chastity and death. They are depicted simply as armed women. Chastity breaks Love's bow and arrow. Death is another woman, armed with a sword and perhaps some poison. There are no horses. Petrarch's image of Fame is simply a "brilliance"; no trumpets. His image of Eternity is God, with no mention of any Last Judgment. To be sure, trumpets are associated with Fame, and also with the Last Judgment. But it takes an artist consciously trying to make the connection to choose those particular images.

It may well be that the chess analogy explains the choice of imagery, such that there is something in common between each of three pairs of cards: Fame's trumpet parallels that of the Eternal, now seen as the Last Judgment. Death was not typically portrayed on horseback; Chastity was associated with unicorns rather than horses. But having Death on a horse and horses drawing Chastity's Chariot make them a visual pair. Petrarch's image of Time is the Sun.The sun, moon, and stars were all images of Time. Another was a clock, in depictions of the virtue of Temperance. Ruins were another way of portraying time. (My source here is "The early Renaissance personification of Time and changing concepts of temporality," by S. Cohen, Renaissance Studies Vol. 14 No. 3.) Time was also portrayed as an old man. But why that image, absent from Petrarch, in particular? The Wheel of Fortune sometimes did have a fourth person on its rim, but he was not usually an old man. Making him so pairs him with Time. In this way the chess analogy of pairs of pieces may have influenced the design of the CY cards. But it was not any analogy with chess that would have determined the subjects of the CY; it would have been Petrarch, Boccaccio, and the medieval depictions of the seven virtues that did that, as 14 triumphal subjects. The card-designer accomplished the "chess analogy" once the subjects had been thought of. The Emperor and the Empress, as ranks above that of King and Queen, then made 16.


I need to add something to what I wrote a few hours ago:

What is on the cards corresponds also to the designs of the corresponding chess pieces. The Emperor and Empress are crowned royalty, like the King and Queen in chess. The Fame card has castles, and that is what the rook looks like; it even has the alternative name of "castle" or "tower". Time and the Wheel have old men, and the bishops, whatever they were actually called, had bishop's headgear. Not only are the designs parallel, but also the power of the cards, compared to their chess correspondents (an important part of Huck's theory). The King and Queen, at that time, were the weakest pieces as far as where they could move and so capture other pieces. The rooks were the most powerful. Similarly Fame and the Angel were the most powerful cards. The others in both cases were in the middle. All of this is part of the elegance of Huck's "chess theory".

However I have learned from both Franco and Huck. From Franco I learn that the suit cards' subjects parallel those of chess, including the individual court cards. From Huck I learn that the triumphs parallel the pieces in the way they are designed. (Added later: I could have also learned that from Huck, if I had found his article to that effect, at So it seems that chess permeates all 80 cards of the CY pack.


I have a few criticisms of Franco's essay, fairly minor, I think, but still important enough to draw attention to.

One issue is the question of whether each triumph has to somehow triumph over the one before it. He derives two principles from Marziano's example:
The sixteen additional cards must first of all correspond to a triumphal series, with a hierarchy that presents itself logically and easy to remember, and perhaps if possible organized like the series of the Trionfi of Petrarch, with a clear justification for why each triumphs over the previous card and is overtaken by the subsequent. If the succession in trick-taking power is not clear, it would be necessary to add the sequential numbers on the cards, as was done in the tarot later. The requirement indicated may be the only one necessary, but if following the example of Marziano is desired, there is a second to be fulfilled: the sixteen additional cards must also be able to be situated [dovrebbero essere situabili] into four groups of four triumphal cards, each at the head of one of the four suits of the 64 card deck identified earlier.

While it might be true that each god is somehow more important or powerful than the one before it in Marziano's sequence, I don't agree that this first requirement has to be followed for a game, and pack, of triumphs to be successful. For a pack to be carte a trionfi some of the cards have to depict triumphal themes. Also, since it is a trick-taking game, some cards should allegorically triumph over other cards. Petrarch's triumphs do that in relation to each other. But virtues do not have to triumph over other virtues, or even the card before it in the triumphal series. They are triumphal by virtue of their praiseworthiness in regard to the corresponding vice. If there is the sequence faith, hope, charity, it is enough that the sequence be not difficult to memorize, not that one virtue dominate over or be more important than another. In my proposal, the theologicals appear in just that order, the same as St. Paul in I Corinthians. So it is easy to remember. In Franco's, there is something similar, but it goes "hope, prudence, faith, charity".


I do not see how each triumphs over the one before it. And it is a harder sequence to remember. Not only is the order different, but there is a cardinal virtue in the middle, prudence. The only reason I can imagine for it to be there is that it had to go somewhere, having been removed from the tarot sequence. It is probably a development after prudence was removed from the tarot sequence, but here put in the same general part of the deck.

The cardinal virtues, unlike the theologicals, were listed in numerous different ways during the Middle Ages. That is one reason they appear in numerous places in the order in different places and decks. In my proposal they go: justice, fortitude, temperance, prudence. This is precisely the order in which they appear in the Nicolas da Bologna c. 1355 illumination.


I am not sure of the provenance of this illumination, but Bologna at that time was ruled by the Visconti, and somehow it ended up in the Ambrosian Library of Milan. Probably it was in Milan at the time of the CY as well. This order, also in the cards, arranges the virtues in an order that puts prudence as the highest, because it governs all the others; in Plato, wisdom was the highest, but the point is the same. Temperance is second, probably because the principle of the mean between extremes, moderation, is another basic principle of virtue, as Aristotle had argued. Otherwise, justice should triumph over love, in the sense that the jealous extremes of passion need to be controlled by one's sense of justice. Also, faith and hope should triumph over the Wheel, when it brings misfortune. And Temperance, in the sense of proper governance of the appetites, can temporarily defeat Death. However the domination of the cardinal virtues is not simply over the card before it, as I read the allegory, but over all the cards in its foursome.

Franco's proposal, based on the later minchiate, has the order "temperance, fortitude, justice". This is the order of presentation in Plato's Republic, once wisdom is removed. Temperance is presented there as governing the appetites, fortitude as governing the "spirited part" of the soul, and justice the soul as a whole. While the appetitive part is lower than the spirited part, which is lower than the rational part, it still isn't true that one virtue triumphs over another, because all are imposed by the rational part.

What is important, in my model, is that the sequence of cardinal virtues be memorized and as well what other cards belong in each foursome and in what order. Knowing the allegorical connection between the cardinal virtue and the other three cards would help in remembering these groups. Franco's order has a different principle. However it is not simply that of each card in the series triumphing over the one before it, because there are many possibilities.

In Gareth Knight 's book cited in my previous post, p. 201, I was pleased to see that he divided the triumphal cards according to precisely the same principle of one group for each of the four cardinal virtues ( lion&f=false). His exchange in the order of Fortitude and Justice seems related to the Golden Dawn's exchange of the positions of those two virtues in the sequence. It is nice to see that this principle can be applied even to the Waite-Smith, however much it differs from that of the CY. After introducing the four virtues, he relates each of the 22 cards to one of four corresponding "Halls": I cannot tell from Google Books' selection what the first one is, but the second is the "Hall of Strength", the third is the "Hall of Temperance" and the last is the "Hall of the World". Knight identifies the World card with Prudence.


Another disagreement I have with Franco is about the uniqueness of the CY. Most commentators on the CY emphasize that we do not know whether the CY was one of a kind or represented a type with many less exquisite examples. Dummett says (Il Mondo e l'Angelo, p. 52, quoted by Franco),
It is impossible to determine if the Visconti di Modrone pack was an isolated experiment, which was detached from a standard already established, or if it is the only surviving example of a primitive stage in which the tarot pack had not yet acquired the structure that was to later become canonical.

There is also Bandera and Tanzi (I tarocchi di Bembo, 2013, p. 11). After listing the 22 special cards that at one time were called "trionfi", they go on to say:
Nella medisima tradizione, che fa tipica della Lombardia del Quattrocento, esiste poi un altro tipo di tarocchi, composto da un maggior numero di figure: sei anziche quattro, per ogni seme, con l'aggiunta del fante e del cavallo femmina per ogni seme, e, nel gruppo dei Trionfi, delle tri Virtù Teologali, Fede, Speranza e Carità.

(In the same tradition, which is typical of the fifteenth Lombardy, there is another type of tarot, composed of a greater number of figures, six instead of four for each suit, with the addition of the feminine page and knight for each suit, and, in the group of the Triumphs, the three Theolgical Virtues, Faith, Hope and Charity.)
This is rather extreme, simply calling the CY representative of a "tipo". However they may be thinking that even if unique, it is one of a type, just one with one member.

Most recently, there is Timothy Husband, The World at Play 2016, p. 80 (for what it is worth):
It is uncertain if this pack was uniquely structured or if it represents an earlier stage before the tarots were standardized.
Even though we have no documentation of any other of its type, there are reasons for thinking that the CY was not the first deck with features resembling the Marziano in Milan. There is the distance in time between the Marziano and the presumed date of the CY. In 15 or more years, the Marziano structure would likely have been forgotten. That is probably why Dummett in 1993 made the "conjecture" that the tarot was born in Milan around 1428-1430 (in Il Mondo e l'Angelo, for a quote giving c. 1428, see There is also, of course, the 1427 or 1428 marriage, a fit subject for the Love card: There was a Visconti tradition of commissioning illuminated manuscripts to commemorate their marriages (Edith Kirsch, Five Manuscripts of Giangaleazzo Visconti, 1991, cited at However it would not necessarily be Marie of Savoy and himself, because there had been other Savoy brides to Visconti husbands in the past: Caterina of Savoy and Blanche/Bianca of Savoy (Lorredan at The latter is the namesake of the bride in 1441.

Also, Filippo was engaged in a major illuminated manuscript project in 1428, uniquely in his life, that of finishing an illuminated manuscript begun by his father (, again from Kirsch). He did not engage Michelino for this work, but a younger and less expensive artist, named Belbello ( It is not hard to imagine Filippo engaging him or some other to make a pack of cards as well. It would not have had Sforza heraldics on it, but then neither does the Brera-Brambilla. It may have been a game which at first only he played, perhaps with his mistress and daughter, and slowly descended to members of his court, including his condottiere.

The clothing then becomes possibly relevant in fixing the time of the type. It is true that playing cards later sometimes dressed their personages in clothes of an earlier time. Later art did so as well. But later practice is not to the point. What we need to know is the practice at that time, the 1440s. For that other works of art in the same period and geographical region are relevant. I have not myself done a thorough examination. The clothing, much of it, is clearly of an earlier time, around 1430 or earlier (see my blog post at; find "Tolfo". After that time, there was a mood against ostentation. There were even laws against it, called sumptuary laws. I don't have the dates for Milan, but for Venice 1433 was notable (

If the CY is one of a type, it may have been a commemorative, commemorating not only the Sforza-Visconti marriage but also a type now no longer fashionable. In other words, far from being a forerunner of decks to come, it might have been a representative of a type long since superseded. We have no idea. In any case, the precise dating of the CY is now not so important, as long as we recognize that its type is of a certain period. Since the style conforms very precisely to work done by the Bembo workshop in the early 1440s, that is probably when the CY was made. The type may have gone back several years earlier. We simply don't know. We cannot say the CY is unique unless we do know.

Another issue (not a disagreement, just an unresolved issue) is the temporal priority of my proposed order of the 16 triumphal cards vs. Franco's. Here are the two proposals again:

In favor of mine are three points already made. First, the positioning of prudence looks ad hoc, as though once it had been eliminated from the tarot sequence, the longer sequence had to put it somewhere. Another point is that by bunching the three other virtues together, the structure of the sequence as four groups determined by the four cardinal virtues is lost. Unless some other principle governing the four groups can be found, we can at least say that Franco's proposal is more distant from the Marziano model than mine is. Another point is that if the source of the theologicals is St. Paul, his order departs from it.

A fourth point, also rather obvious: if the source of the triumph of Time is Petrarch, putting it before Death departs from it.

Finally, it is hard to identify the Minchiate/Florentine Chariot card as Chastity. Chastity is an especially feminine virtue, because men needed to be sure that their heirs were theirs. With the Temperance card right after Lover, it would seem that it is this virtue that triumphs over Love; if Petrarch is being followed, Temperance is this sequence's version of Chastity. What the Chariot triumphs over is Fortune. If so, the Chariot is simply a symbol of victory in general, and not chastity. And in fact in all the type A Chariot cards, including that of the minchiate, the person on top is male, a Triumphator in the Roman tradition, in which military victories were celebrated by triumphal parades where the victor rode in a chariot. It is a person like Mars in the so-called "Mantegna Tarocchi", who sits on his triumphal chariot. This is another deviation from Petrarch.

To be sure, such deviations are not proof that mine is earlier, because later versions of works sometimes seek to restore things that earlier versions, wanting to be improve on their model, somehow departed from. These are merely considerations that suggest a temporal priority. It is more likely that someone using a source would follow it than depart from it.

Two last points: I am not suggesting that my proposed order for the CY, one that strictly follows that which was given to the Beinecke Library, is exactly right. It is only the division into groups that matters. Second, my proposal should not be construed as an "ur-tarot", that is, the original version of the tarot sequence. If the CY is representative of a type, we have no idea what the original order was. The numbers were not on the cards; so it could easily have been a matter of experimentation. Also, we have no idea what else was around, including at earlier times and other places. Since there are suggestions elsewhere of a 14 card sequence, the ur-tarot might have been 14 cards, corresponding to suits of 14. Or 13 or 12, corresponding to suits of that number. We have no idea.


Franco now has a second note on the Cary-Yale, at In this post I am going to give my translation of the first six sections, out of ten. My goal is to produce atranslation as near as possible to the Italian original. Since the interrelationships of clauses is important to keep visually clear, the translation sometimes keeps the Italian word order, when not to hard to follow in English.. My short comments, mostly giving the original Italian, will be in brackets. The original is at; it is dated Feb. 12, 2016.

Milanese and Florentine Triumphs - Hypotheses and Comments

1. Introduction

This note can be considered as the continuation of one written month ago 1 on Visconti tarot Modrone or Cary-Yale deck that will be referred to simply by the initials CY. That note had as subject title "elucubrazioni "; Michael S. Howard, who to had contributed that study, translated it with ruminations 2 and that term has led me to recognize that the subject was not digested enough. In fact, the conclusion of the previous note was not really conclusive, especially the uncertainty on the interpretation of the CY deck as a precursor of standard triumph decks or as a variant of such packs already in common use.

I turn to the subject by discussing some additional consideration on the virtues and assumed links with the Florentine minchiate. The CY being examined is and remains the same: no change at all whether you consider it in one way or another; however, its historical significance changes, and that very much: in light also of the teaching of Sylvia Mann, the importance of an original specimen before a standard is incomparably superior to that of an extravagant variation on a theme already known.

2. The Virtues

The era of the introduction of the triumphs coincides with that of the early Renaissance, and among the poetic and pictorial cycles of the time were very popular both the triumphs (with influences of non-immediate derivaton from classical civilization and from Petrarch's poem), and the virtues, often presenting their victory over the corresponding vices. In short, that we find among the triumphal cards some triumphs and some virtue does not occasion any surprise;

possibly a few other correspondences [risconti] are to be found in the tarot.

Everyone knows that the virtues are seven, four cardinal and three theological, but perhaps it is useful to provide some official clarification in this regard; this is how they are defined under the title, In summary, in a catechism of 790 pages 3.

1833 Virtue is a habitual and firm disposition to do good. 1834 The human virtues are stable dispositions of the intellect and the will that govern our acts, order our passions and guide our conduct according to reason and faith. They can be grouped around the four cardinal virtues, prudence, justice, fortitude and temperance. 1835 Prudence disposes the practical reason to discern, in every circumstance, our true good and to choose the means for achieving it. 1836 Justice consists in the firm and constant will to give God and neighbor their due. 1837 Fortitude ensures, in difficulties, steadfastness and constancy in the pursuit of the good. 1838 Temperance moderates the attraction of pleasures of the senses and provides balance in the use of created goods. 1840 The theological virtues dispose Christians to live in relationship with the Holy Trinity. They have God as their origin, motive and object, God known though faith, hoped for and loved for himself. 1841 There are three theological virtues: faith, hope and charity. They inform and give life to all the moral virtues. 1842 By faith, we believe in God and believe all that he has revealed to us and that Holy Church proposes for our belief. 1843 By hope we desire and await from God, with faith, eternal life and the graces to merit it. 1844 By charity, we love God above all things and our neighbor as ourselves for love of God. It is "the bond of perfection" and the form of all the virtues.
I do not know if and how this doctrine of the virtues has already been superseded by more recent versions of the official catechism, but for our purposes it seems to me already more than necessary; if necessary, you should resort to the doctrine of the time, it will not be easy to find in a form similar "official". I add only (Fig. 1) a photo of Faith and Hope, designed by Andrea Pisano a century before the time in question here. Now we know enough to continue our reflections.
3. Catechismo della Chiesa Cattolica, Città del Vaticano 1992, pp.466-467.

[Translator's note: I was not able to get a good reproduction of Franco's photo to put here; you may go to page 7 of the original, but I found one on the Web just as good, at There are also very clear photos of the two separately at and, by Mary Anne Sullivan at


Figure 1. Florence, Baptistry, detail from the South Door

Anyone interested in the tarot cards, in their various forms, structures and orders, certainly will bump into the problem of the virtues. The triumphal cards are ordered so as to see a growing power of the same - a necessary condition for using them in the game with no writing on them directly of their number in the series. Michael Dummett, having discussed at length the various kinds of orders for the major Italian cities, stressed the fact that precisely the three cards that represent three of the seven virtues (indeed, three of the four cardinal virtues) are the most erratic in the "canonical" orders of the triumphal card sequence. On issues like that I can defer to what is recurrently discussed in the literature, at least from 1980 onwards.

The reason why I will deal briefly with the virtues now is that the three theological virtues are in the CY; in this case three are preserved out of three, and the case is presented unusually favorably. As for the four cardinal virtues, only one of them is preserved, fortitude, and, on the other hand, the card of Prudence is absent in almost all tarot packs.

3. Direct Construction of the virtues

By direct reconstruction I mean the supposition that the CY is a variation on the theme of the tarot existing in their canonical form, and therefore attempts are made to associate the three "intruders" cards of the theological virtues, which would all be absent there at the beginning, to others replacing precisely those three. No one, to my knowledge, has suggested that the hypothetical complete sequence of triumphal cards in this deck was made up of at least 25 cards: the traditional 22 plus 3 new cards, precisely those of the theological virtues in question. Therefore, is it established that some figures of the tarot were inserted instead of the three theological virtues? Unfortunately an association group to group is not seen, and one should proceed to try different analogies for each individual card. For reports of this kind I can I can use as a basis the famous Encyclopedia of Kaplan 4.
The trump cards Hope and Charity (and the card Faith, which is not shown) do not appear in traditional seventy-eight-card tarocchi decks but are found in minchiate packs, which generally comprise ninety-seven cards. For this reason, some researchers believe the Cary-Yale tarocchi pack is either a minchiate deck or an intermediate game in the development and evolution of either tarot or minchiate. Hope depicts a crowned female figure in profile wearing a long robe, kneeling in prayer, with an anchor tied to her wrists. At the bottom of the card is a hunched figure of a man with a rope around his neck and with the words “Juda traditor” written in white letters on his purple garment. The despairing figure of vice is Judas. It has been suggested that the card of Hope may be a substitute for any one of several traditional Major Arcana cards missing in the Cary-Yale pack – Temperance, or The Hanged Man (suggested by the rope), or The Star with its symbolic meaning of rising new hope. Charity shows a crowned and seated female figure facing front who carries a silver torch in her right hand while supporting a suckling infant with her left arm. Charity is richly robed in an ornate gown with ermine cape. At her feet, beneath the throne at the bottom left of the card, is a crowned king suggesting King Herod. Charity may be a substitute for The Popess, but the image of a woman breast-feeding her child is inconsistent with the traditional imagery of The Popess. The Faith card depicts a female figure with a cross in her left hand; the index finger of her right hand is upraised to ward off evil spirits. Beneath her throne is a crowned king, possibly the figure of Heresy. Faith may be a substitute for The Pope or The Popess.
4. S. R. Kaplan, The Encyclopedia of Tarot. Vol 1. New York 1978, p. 91.

When the same Kaplan puts all the preserved Visconti-Sforza cards in a long table on p. 64 of the same first volume of the Encyclopedia, he adds the three theological virtues above the 21 World, and also, with a question mark following, the Popess-Charity , Pope-Faith, and Hope-Starassociations. But we read in the text copied above that Hope, as well as replacing the Star, could have substituted for Temperance or even the Hanged Man. On the other hand, in place of the Popess could also have been inserted Faith, alternatively Charity.

As a source, the Encyclopedia is not perfect because it does not distinguish sufficiently the associations suggested by the author from those suggested by other experts, who also are mentioned in general and collectively, without being named individually. For our purposes, absolute accuracy is not required, however; it is sufficient to understand if indeed there could be such a substitution. The very fact that recognizable clues were found and interpreted in favor of such associations can be considered a confirmation of that possibility.

4. Inverse Reconstruction of the virtues

By an inverse reconstruction I mean that there obtains in the CY a situation preceding the canonical form of the tarot and that, correspondingly, the three theological virtues were already present in this experimental and pioneering form. What card has subsequently substituted for that which was originally a theological virtue? Our task here becomes easy: we do not have to study the situation again, but we can take advantage of what has already been suggested. One simply has to copy down the similarities found before. Was it true that the existing Popess could have turned into a figure of Charity in the CY? Good; then it can also be said, now, that the pre-existing figure of Charity was then transformed over time (she may have lost the milk) into the canonical one of the Popess. The same applies for the Faith-Pope, Hope-Star couples and other hypothesized associations between a theological virtue and a "canonical" card.

In short, at first sight pairs of associated figures work in both directions. Looking more closely, however, it is not at all certain that the symmetry is really respected: it is possible, even probable, that a likeness suggested in one way proves far less convincing

when viewed in the opposite direction. There is also an asymmetry for the same basic reasons in the replacement of the figures: having a homogeneous group of three figures, how is it possible to "break it up" into three independent figures, or at least into one independent figure and a reduced group of two?

Furthermore, it is naturally more reasonable to find the three theological virtues together in a high position on the list, rather than away from each other or in the front positions. In short, if a replacement there was, it would seem more logical to the group the theological virtues and move them up in the ranking (direct reconstruction), rather than vice versa, their break-up and movement down (inverse reconstruction). On this basis one could conclude that the CY was obtained from a standard deck and not vice versa. However, one might also conclude that it is the very idea of a replacement that does not find sufficient handholds. The situation remains unclear; I will go back over this after pursuing another detour through ... Florence.

5. Comment on minchiate

In traditional Florentine minchiate there are no fewer than 41 triumphal cards; it seems unlikely that this deck, which has been used for centuries, was born with all its 97 cards. In particular, the cards of the four elements and the twelve zodiac signs are presented as a rear insertion within an already standardized sequence ; All historians agree on an interpretation of this sort. Among other things, it is a sequence that has an order recognizable in itself (which turned out not to be sufficient for their use as trump cards in a game, so that to ensure the order the cards were marked with the actual numbers). For example, this is quite reasonable given that the zodiac signs were not added so that one follows the other, in the same way as the corresponding constellations in the heavens succeed one another in the passing months. This is already a sequence a little different from that of Petrarch's triumphs, in which the victory and the triumph were more obvious and corresponded almost to victory in a battle.

What was the initial form of minchiate? No one knows. The only thing that is known is that the game of minchiate in 1477 was done

in multiple ways and that only the one in which cards were won was allowed. It was the difference between the cards taken which determined the final score, but how many all the cards in play were we do not know. Let us assume, just for the sake of argument, that the minchiate pack was originally a deck of 80 cards like the hypothetical CY reconstructed in the study described above, with the help of a possible analogy with minchiate, if only for the theological virtues. One would also be satisfied to have finally seen a minchiate with a "reasonable" structure. However, the discussion cannot end here. How did that CY, which looks like it was invented in the Visconti court, not leave traces in Milan, but leave them some time later in Florence? As if only in Florence had survived an experiment that in Milan would be born only to die very quickly after its birth.

Any reconstruction of the type [genere] ends in leaving us perplexed. Let us try once more to see the situation in reverse: is it possible that a primitive minchiate pack already in use in Florence has generated the Milanese CY? By primitive minchiate pack I mean here the pack of Florentine triumphs purchased by Giusto Giusti in 1440 5, which will be indicated from now on with the initials GG.

6. Two links in a chain

To continue the discussion some preliminary hypotheses. in part already used or under consideration. are necessary: we have already assumed that the CY was originally a deck of 80 cards, 40 pips [numerali, numbered], 24 courts [figurata], 16 triumphal [trionfali]; we admit that the GG Florentine triumphs (whether already called minchiate or not) has existed with its own composition not only before the corresponding standard of 97 cards, but even before the tarot of 78. While holding in the background the appropriate "standard" decks of tarot and minchiate as ultimate goals, we will examine three decks of "experimental " playing cards " as schematically shown in Fig. 2, that of Marziano da Tortona, referred to as the MZ, and also the CY and the GG. Unfortunately, the discussion must be based mainly on assumptions, advanced in succession: none of the three decks is precisely known; of the third, the only one for which we know with
5. Th. Depaulis, Le Tarot révélé. La Tour-de-Peilz 2013, p. 17-18.

certainty the date of production, no card has survived; the only pack of which we have cards is the second, but not all are there, we are not sure how many and which are now lost. The first two packs present themselves as originating in the Milanese court of the Visconti, the first definitely linked, and partly due, to Duke Filippo Maria; perhaps also the second, at least if we accept an early date among those proposed. Already in these first two decks there are uncertainties. The MZ certainly has four kings and sixteen triumphal cards; uncertain is the number of cards of the four suits, including court cards possibly present alongside the king. Especially uncertain is whether this first known deck of triumphs, could really represent the first-ever attempt to create a pack of the type [genere]. with additional cards superior to the others. It also remains uncertain whether it was a totally isolated attempt, with no direct sequel, or if it can be considered as a link in a chain in which successive attempts take into account the prior ones (the identity of place is a clue especially favoring the chain).

The CY also has several uncertain points. The date of 1441 often proposed is not secure. For completing the preserved triumphal cards there are several possible scenarios and no certainty. The hypothetical reconstruction of 16 triumphal cards is based on a possible analogy with chess pieces and (especially in my opinion) on a possible analogy with the MZ - so that would in fact be a previous link on the same chain - with the same number of triumphal cards but already changed in the direction of the standard decks to follow.

The temporal sequence between the two is secure (in the sense that no one has yet proposed a date of the CY prior to the MZ), as is the fact that going from the first to the second we approach the typical tarot form, while not reaching it. Having admitted that the CY triumphal cards were originally 16, the same as those of the MZ, it is easy to assume that the gods or deified heroes are transformed into other triumphant characters, but with a similar structure (at the limit [al limite] also in so far as [per quanto riguarda} the "transformation" of the triumphal series into cards of the four suits, if the allocation proposed into four groups by Michael Howard is convincing).

The clients of these special packs cannot be overlooked. Especially for the CY, the origin in the ducal court of Milan unfortunately only serves

to explain the extraordinary character. This uniqueness could be of two different types, precisely those that we would like to distinguish, either an elegant variation on the theme of the traditional triumphs or an innovative intermediate structure that will lead to the standard triumphs. To decide, only Einstein with his space-tim ecan help: understanding the place to be that of the court of Milan, the time coordinate will be decisive and the prototype of great historical interest will be all the more likely the further back in time one can push the dating; Already 1441, the most often suggested, is too recent.

To be continued.


This is the conclusion of Franco's 2nd note on the Cary-Yale, at

7. The third hypothetical link in the chain

At this point it may be useful to introduce into the discussion Florentine minchiate. What links minchiate with the above? Exactly nothing, it would be said, and so said all the experts, with the exception of some art historian ... not knowing the history of playing cards. But if we agree to call minchiate the first Florentine triumphs [Ma se accettiamo di chiamare minchiate i primi trionfi fiorentini], here in 1440 Giusto Giusti is ordering a pack, and no one knows how this GG might be constituted, except that it had especially the for us hardly useful coat of arms of Sigismondo Malatesta. There could not be a big difference between the Florentine and Milanese decks if the same Sigismondo Malatesta receives both for his use from Florence in 1440 and from Milan and Cremona some ten years later.

The cards of the Florentine triumphs could be born and developed in a manner completely independent of those of Milan, but this presents itself as a hardly logical reconstruction, and it appears rather likely that the two developments were somehow connected. There must then be considered the hypotheses that the Milanese triumphs gave rise to the Florentine and also, at least in principle, that the Florentine triumphs gave rise to the Milanese ones.

It may seem strange that I do not take into account Ferrara, also having in mind the subtitle of the fundamental book of Dummett, From Ferrara to Salt Lake City. There is something not quite right about that [Qualcosa non torna.] Surely the courts of Milan and Ferrara were in close contact, but when in 1444 those two ducal courts played a little trionfi with special cards, players of the people in Florence were playing in the streets with

common decks. 8 They are precisely those packs of Florentine triumphs that could solve most of our remaining doubts in our historical reconstructions, necessarily [up to now] the result of unverifiable speculation. I am able to imagine "ordinary" packs like that, used by the people, as even [perhaps] produced by Milanese handicraft, in addition to Florentine, although I personally have some difficulty with Ferrara, where early triumphs are actually documented in court circles. [In response to my request for clarification, Franco tells me, "I am not able to imagine in Ferrara an active handicraft capable of producing a big amount of card packs, and people playing in the streets with ordinary cards of local production – as certainly was possible in Florence and maybe! in Milan.] (If anyone thought also of Bologna, or other cities, there might be clues in favor, but they can be overlooked in this discussion.)


(Figure 2. Scheme of packs discussed
[Milano = Milan; Firenza = Florence]

From Florence we have no information [notizia] of decks as old as the MZ, but certainly already in 1440 the GG presented no extraordinary novelty; it was enough to order its production at any of the local manufacturers of playing cards. For the dates attributed to the packs discussed, it does not appear impossible that the GG influenced the CY. However, since it is reasonable to bring the CY back to a local development from the preceding MZ, we would need to make a nice unsupportable somersault [un bel salto mortale] to suppose that the same Marziano had had the idea of his Milan deck from a somewhat similar effort underway in Florence when he was finishing, right there, his university studies. At the state of our current knowledge, this would be an acrobatic act hardly recommendable.

Consequently, we should reconstruct the chain as a derivation of the Florentine triumphs from previous ones in Milan, though not

from the Milanese standard ones but from the intermediate, so to speak, precisely the type of the CY. Then in Milan that deck would have been changed to the standard tarot, while in Florence it would remain as the basis of the next expansion of the traditional minchiate. Necessary conditions for the plausibility and validity of a reconstruction of the type [genere] are the CY being as close as possible to that of the MZ as dated and structured and absolutely not the later and extravagant occasional pack purposefully different from the ones already used at the time.

8. Return to the virtues.

One cannot consistently follow a logical thread between preserved objects and others of an existence, or at least of a form [forma], only hypothesized. In particular, there are secondary observations that, taken individually, may steer in one direction or in another, but that overall perhaps lead to still greater confusion. Of these observations, I would begin with the previously suspended discussion of the virtues.

In the minchiate sequence (the real one respected for centuries, and the other, purely hypothetical, suggested on the same basis as the CY in the previous note) the theological virtues are found in very high position. It may be reasonable, because the theological virtues certainly cannot be put in a hierarchical order below those of the cardinals, or under other subjects that are lower or even with negative characteristics. However it cannot easily be understood why the cardinal virtue of prudence is inserted, unexpectedly, within the group of the three cardinal virtues.

Once this strange order has been accepted, departing from the GG (assuming that it already respected the sequence of minchiate) and from the CY (if it really was in conformity with the suggested reconstruction), we understand that there may have been interactions between the two decks, Florentine and Milanese. If the theological virtues, all three together, had been inserted into a deck that had none, thus getting in Florence the minchiate (or sooner the GG), one would expect to find them inserted as a compact group, no Prudence in the middle, as indeed is found in the same minchiate for the four elements or twelve zodiac signs.

In conclusion, if one can speak of an “error” in the positioning of Prudence, this would be explained

preferably as present first in Florence, and then eventually repeated in Milan, rather than born only later in Florence, at the time of the "new" theological virtues. To be convinced of a passage like this, however, it would be necessary to acknowledge either that the MZ had been an isolated experiment, not followed in Milan, or that the MZ and GG cooperated together at the birth of the CY; however the task appears daunting.

9. The court cards.

Probably the most original characteristic of the CY is the presence of six court cards in each of the four suits, for a total of 24 cards of which only 17 are preserved. One interpretation, actually suggested by some experts - which is not completely convincing, but yet does not seem absurd - is that that particular pack was intended for a lady of the court, and had the intention of enhancing alongside the knights and military leaders [condottiere] also the corresponding female figures who lived in the ducal court: therefore also ladies in waiting [dame di compagnia] and maids [cameriere], next to the queen in the cards. A similar explanation can be advanced for the pack described to the card historians [agli istorici delle carte] by [da] Ross Caldwell 7: that deck being dedicated to a noble woman, why not insert a superior card as Emperor-Empress, by identifying her precisely with the same lady, after the other cards are attributed to the personages [personaggi] who live next door?

A different explanation for the court cards and pips may be based on analogy with chess pieces, with a queen’s side alongside the king’s, easily converted into female characters on one side and male on the other. This hypothesis was advanced independently in the previous note, but was already suggested earlier by Lothar Teikemeier in 2003 8.

These are hypotheses, and other might be proposed; but let us see the case as that ring of the chain that was mentioned. For the MZ and GG we do not have enough information, but in none of the known tarots is there such multiplicity. The deck in part, but only in a small part, is still approaching that of minchiate,

because in it there are at least the two male and the two female pages. Some hints could be received of possible interaction between the GG and CY; However, if we limit our attention to the court cards, it appears easier to explain the CY as a quirky variation on existing successes in a more traditional form.

10. Conclusions

An earlier discussion of the Cary-Yale tarot, or Visconti di Modrone, has been continued, commenting on various assumptions on the cards of the theological virtues and, more generally, the potential reciprocal influences between Milan and Florence. For Florence there is not the information given comparable with that of the Milanese triumph pack [mazzo dei trionfi] of Marziano da Tortona, the only known triumphs definitely of the first quarter of the fifteenth century, and this is a strong argument in favor of a Milanese priority. However, at the dates of the precious Visconti-Sforza packs, Florentine triumphs were already circulating among ordinary people, and it would be precisely those that should be a priority to find and study. For the Cary-Yale deck it still remains uncertain whether it could be an historically important prototype, but the indications against seem more significant than those in favor. The discussion cannot certainly be considered exhausted, but on these issues, and especially on the Milanese tarot, there is felt more the lack of further documentation than (the lack) of an umpteenth contribution to the debate on the little that has been preserved.

Franco Pratesi – 12.02.2016


I have a few criticisms of Franco, affecting only sections 8 and 9 of his note; I have no no problems until then. It seems to me that in section 9 he is ignoring some of the assumptions up to that point, and in section 9 he is ignoring some data. My intent is to go over the same points as Franco, but in a way that accepts and takes into account all the assumptionst. I want to emphasize that I am not trying to substitute a different set of assumptions that are somehow more reasonable, or more grounded in fact. The main object of these assumptions, as I see it anyway, is to reconstruct how Marziano's game might relate to the later tarot and minchiate.

1. Hypothetical and real orders of triumphal cards.

In section 8 Franco says:
In the minchiate sequence (the real one respected for centuries, and the other, purely hypothetical, suggested on the same basis as the CY in the previous note) the theological virtues are found in very high position. It may be reasonable, because the theological virtues certainly cannot be put in a hierarchical order below those of the cardinals, or under other subjects that are lower or even with negative characteristics.
The problem I see is that up until section 8 there were two hypothetical constructions for the CY, Franco's and mine. They both need to be examined. For reference, here they are again.


The subjects in italics are hypothetical reconstructions. Mine, admittedly, does not have the theological virtues in a high position. I was trying to make a link between the Marziano and the Cary-Yale by means of a division of 16 triumphal cards into four groups, using the schema that was passed on by the Cary Family and filling in the blanks as best I could. There are various ways of ordering the virtues. That is part of an answer to Dummett's question of why the virtues appear in such different places. Does faith trump prudence, or fortitude trump temperance? A rationale can be provided, but so can rationales for other orders. If you look at the line-ups of the seven virtues as they appear in various frescoes (i.e. Giotto c. 1305, miniatures, and cassone (e.g. Pesselino 1460s, dal Ponte 1430s, among others), sometimes Charity is most prominent (taking the center position) sometimes Prudence, sometimes Justice. The others are similarly variable.

In the CY order I propose, the cardinals are ordered in the way that they appear in a 15th century Bolognese illumination which for some indeterminate time has been in Milan ( The theologicals are ordered in St. Paul's order. What is needed is a memorable order. Allegories can be constructed to help one's memory. It is not hard to remember that in the face of a bad turn of the Wheel, one must have not only fortitude but faith, the basis for hope; then Charity comes next, following St. Paul. This seems to me not illogical nor hard to remember. It is an adequate basis for a sequence. With it, most importantly, one can play whatever the game was that Marziano designed, one that required the four groups.

What I want to suggest is that the ordering of the CY-type and the GG-type might have been different. The GG-type can be the proto-minchiate order and the CY-type can be my suggested order, based on the structure of the MZ and the assignments passed on by the Cary family.

2. Return to the virtues, in terms of "direct" as opposed to "inverse" constructions.

Franco continues:
If the theological virtues, all three together, had been inserted into a deck that had none, thus getting in Florence the minchiate (or sooner the GG), one would expect to find them inserted as a compact group, no Prudence in the middle, as indeed is found in the same minchiate for the four elements or twelve zodiac signs.

This seems to be meant as a refutation of the "direct" construction of the theologicals, from different previous cards. But no matter whether these cards were replacements or put there originally, even as part of the original ur-tarot, it is still unexpected. The oddness is not something that one can draw any conclusions from, regarding whether it resulted from replacing other cards. Presumably it was done for a reason, whenever it was done. I would guess that people had a sense that prudence was very high in the virtues and it should be put with other high virtues.

What needs to be done is just what Franco did earlier: look also at the inverse construction of the virtues, and then compare the two. So if the theologicals were there already, what cards would they be replaced by? For Franco, this perhaps would be a nonsense question, because they never were replaced in minchiate. But there were two permitted games in Florence, at least in 1477, one called "triumphs" and the other called "minchiate". They seem also have had two different decks as well. The Charles VI, Florentine if only because of its Medici "palle" on the Chariot card, would seem to be a triumph deck. It could be that someone removed all the minchiate-only cards, but that seems unlikely with beautiful works of art. The Charles VI has a Moon card and a Sun card, so likely a Star card as well. The Star, Moon, and Sun form a group, without any subtlety required. There is also a lightning-truck tower in the Charles VI, a card called ""Lightning-bolt" (Sagitta, literally "Arrow") or "Fuoco" (Fire) in the early lists. And if you look, Fire, Star, Moon, and Sun are in exactly the same part of the sequence as the three virtues plus prudence in the Minchiate. It would seem to me reasonable that the game called "triumphs" replaced the 3 theologicals plus prudence with Fire, Star, Moon, and Sun. Just as the four in one group are all virtues, the four in the other are all lights, of increasing strength. Then another deck was for people who didn't want the theologicals removed; at some point it was called minchiate.

It seems to me that the inverse construction makes more sense than the direct, because it has one coherent group replacing another in exactly the same places in the sequence. It is like the replacement of the Popess by Juno and the Pope by Jupiter in the Besancon, where Juno took the Popess's position and Jupiter the Pope's.

3. Interrelationships between the two proposed orders, the CY-type (my proposal) and the GG-type (Franco's proposal).

Clearly there is a relationship between the CY-type (Milan) and the GG-type (Florence). If nothing else, they have the same cards, at least on the hypotheses that Franco and I were considering, mine and his. But what is the relationship?

There are four possibilities, excluding for the sake of argument the influence of other cities:

(1) the card-selection was generated in Milan on the basis of the MZ and then passed to Florence, which gave them a different order that they found more "logical".

(2) The cards were generated in Florence in an order that made sense to them, passed on to Milan and rearranged there on the basis of the MZ.

(3) The cards were generated in Milan but then rearrnaged due to influence from Florence.

(4) The cards were generated in Florence and rearranged due to influence from Milan.

On hypothesis 1, the 16 cards, organized according to the "four groups" order I proposed, came to Florence. Either the card makers didn't know the Milanese order or they thought that the sequence there was illogical. They didn't know about Marziano and his game, and how spreading out the cardinal virtues fits the idea of the four groups. They only knew about other trick-taking games, and the 4 + 3 groupings of the virtues. The theologicals should be near the top, they reasoned. And time should be before death, because that is when it counts. They think that Petrarch, with his desire for Fame, was an elitist. The average person isn't going to be famous no matter what he does. To him it doesn't matter that Time destroys Fame; Time is what happens before Death. The result in Florence is then Franco's order of the 16 triumphal cards.

On hypothesis 2, the cards are generated in Florence, using the same reasoning I have just gone through for that city. Then they go to Milan, where Filippo sees the cards from the standpoint of the Marziano game and re-orders them.

On hypothesis 3, Milan adjusts its cards after seeing what Florence has done. For example, it puts prudence between hope and faith, puts the virtues into two groups, and raises the three theologicals plus the one intellectual virtue to a higher level.

On hypothesis 4, Florence adjusts its cards' order after seeing what Milan has done.

I see no justification for hypothesis 3, at least during Filippo's lifetime. Nothing suggests that prudence was ever between hope and charity there. It is true that we can do an inverse construction, based on the 16th century lists, that puts these four virtues in the same place as in the proto-minchiate. But such replacement could have been done at any time, and is more likely after 1450, when there was a new duke of Milan, one in alliance with Florence and friendly with the Medici, and when tarot in Milan was more likely not the preserve of the court but of the populace.

Hypothesis 4 is unlikely because the proto-minchiate order is in fact Florentine, and there is nothing about that order that suggests Milan's influence on it in particular.

Hypothesis 2 has the problem of explaining why exactly 16 cards were chosen, and not only that, but ones that fit the Marziano structure so well. First, why 16 in particular? Why was it necessary to borrow a triumph from Boccaccio, that of Fortune. And if one, why only one? Boccaccio had other triumphs, namely Riches and Wisdom. (Wisdom was seen by some as higher than Prudence, while others made no distinction.) Then in particular, the four cardinal virtues fit Marziano's structure well. If the cards were generated in Florence, without Marziano's influence (he hadn't been in Florence since 1407), it is a striking coincidence that it fits Marziano so well.

So I would redo Franco's flow chart. Given the assumptions Franco and I developed, the triumphal cards most likely went one way in the 1430s, from Milan to Florence. Then, if the four "replacements" --perhaps three, if Prudence was simply dropped, and Fire added considerably later--happened in Florence, these changes went back to Milan, when the city was under new management. Or, if the "replacements" came from somewhere else, such as Ferarra, the same process occurred but from a different city. Less likely, I think, it could have happened in Milan itself, and spread to other cities from there. In any case, the game was now for the masses in Milan as well as Florence, and Marziano's square array of four groups and four cardinal virtues a thing of the distant past.

Below is my proposed flow chart. It ignores most intermediate steps, if any. It also ignores influence from other cities. In Florence, I see that as particularly important in the development of its non-minchiate "triumphs", both before and after the GG of 1440, and in changing the triumphs in Milan after Filippo's death. It may also have been important in forming the CY-type in the first place. I make a point of emphasizing that is the order of a CY type that is being hypothesized, not just of the physical object that has been preserved. I assume that is what Franco means by "CY' in his chart, but it is not totally clear. The same is true of the "GG type" (even if it may be that the GG itself is not of this type, as I will explain later). By “GG” in the chart I mean “Franco’s GG-type, and possibly the GG itself”.


And for comparison here is Franco's again.


4. The court cards.

There is the question of whether, as far as the court cards, the CY is "a quirky variation on successes in a more traditional form", as Franco concludes (stravagante variazione sul tema di trionfi già esistenti in una forma più tradizionale) , or, alternatively a quirky early form that became eclipsed by later successes. I am not sure exactly what "traditional forms" he means, but presumably one at least without female knights, and probably with half as many female pages.

But again, there is the magic number 16 to consider. Decks in fact did vary in the number of suits, but as far as we know, all the pre-triumph decks in which the suits had equal priority had the same number of cards in each suit (whether Marziano is an exception is unknown; there it is more likely that including the triumphal cards in each suit, the suits had the same number as the triumphal cards alone). That is one reason for thinking that the number of cards per suit would have equaled the number of cards in the triumphal cards, grouped as a suit. Another reason is the various bits of odd documentation suggestive of such a hypothesis. In 1422 Ferrara there are "13 new playing cards" with backs the same as the old (; on Jan. 1 1441 there are "14 figures"; in 1457 there are orders for decks of "70 cards", which could either be 14 x 5 or 4 x 12 + 22; late in the century Franco finds that the cost of a triumphs deck is in a 5:4 ratio to the cost of an ordinary deck. None of this proves anything, to be sure.

Dummett's hypothesis, that the ratio of triumphs to cards per suit was 3:2, has going for it the various lists of 21 triumphs (the fool not being a triumph) from a variety of sources and cities. But these are late in the 15th century at their earliest. Can we really infer from generalizations made some distance after creation to a time near the invention? If so, then refrigerators never had unsightly bulbs on top, and automobiles never ran on steam.

So there was a very good reason for females in all ranks: to make the number in a suit equal 16. There was already a precedent for females in all ranks in the Stuttgart Playing Cards, c. 1430, where two of the four suits were all-female and two all-male. Admittedly, none of the courts were on horseback. But for that there are the Courtly Hunt Cards, of the 1430s: the queens in herons and falcons are on horseback (( That this theme reached the common level is suggested by some French 1490s woodcut cards (their suit signs not yet stenciled on) reproduced in Hind's Introduction to a History of the Woodcut, vol. 1 (

There is every reason to think that the innovation would have been well received by its intended audience. In Milan and nearby the game was a social occasion that included and perhaps was presided over by women, at least among the nobility and the rich (which is all we have any evidence of). The Borromeo fresco, c. 1445-1450, had three women and two men, with the dominant figure being the center woman ( There is an extant description of the lost frescoes at Pavia done for Galeazzo Maria Sforza, which depicted "Elisabetta [Maria Sforza] and damsels playing cards and other games" (I am quoting Lubkin, A Renaissance court, p. 309, citing Welch, "Galeazzo and the Castello di Pavia", p. 373). There is the fresco at Malpaga Castle, on the border between Lombardy and the Veneto, shown in an essay by Andrea Vitali,, near the end; the card players are all women. There is also Viti's letter and illustrations to a lady of the court in Urbino about the Boiardo deck (

On my favored hypothesis 1 about the triumphs, the deck that would have gone from Milan to Florence would have been such a deck, before the Florentines adjusted it to their own taste. This is where Fernando de la Torres' verses and ekphrases (descriptions of imagined paintings) to Countess Casteneda in Spain of c. 1450 are relevant (Ross Caldwell at They did more than create an all-trumping female Emperor (never called Imperatriz) to extol his patroness and her virtues. He made most of the 48 other cards females, as can be seen by looking at the individual verses and descriptions, even while retaining the traditional male titles of rey, cavallero, and sota for the courts. That is precisely the kind of gender-bending that the female knights of the CY have. If, as Ross surmises, Torre may have got the idea of a trump card from his stay in Florence in 1434-1435, he likely would have got the idea of feminizing male cards from the same source.

There is not the same evidence for female dominance in triumph-playing in Florence as in Milan. What there is, is of men playing with men: two men arrested for playing "trionfi" in 1443, Pulci and Lorenzo playing "minchiate" in 1466. There may be social reasons for this. Men didn't marry until 28 or 30. So fathers locked up their daughters and young men developed the habit of socializing with each other (see references 56 and 57 at; their section on Florence is also helpful). So it is harder to imagine female courts of all ranks arising in Florence than Milan. Milan-style suit cards in Florence probably changed fairly soon, I'd guess by 1440, because even in Milan of the mid 1440s the Brera-Brambilla has the standard four courts. And with this change went the need for precisely 16 triumphs and the link to the suits. Whether the GG of 1440 was a proto-triumph or a proto-minchiate, I would not hazard a guess.

So I would say that the CY-type, as far as its courts, was most likely a quirky beginning, or early phase, first in Milan and then Florence, that was soon eclipsed, sooner in Florence than in Milan, by later successes.

5. Conclusion.

I started with two hypothetical orders of the same 16 cards, presumably derived from Petrarch, Boccaccio, and the 7 medieval virtues, to fit the structue of Marziano's deck in Milan. Various possible ways the triumphs could have evolved between Florence and Milan were examined. One order of triumphs, suggested by documents passed on by the Cary Family, is closest to Marziano's structure and therefore associated with Milan. The other is associated with the later minchiate and therefore Florence. Most likely the cards would have gone from Milan to Florence, where, not knowing about Marziano, cardmakers rearranged them on a different basis, in particular moving the theological virtues up in the order. At the same time the 16-card suits would have been adopted for a time, as evidenced by Torre's use of feminized suit cards in Spain; but at some point Florence would have dropped female knights and relegated female pages to just two of the suits. The result is then the basis for the game later identified as "minchiate". Further changes, replacing prudence and the theological virtues with other subjects, perhaps already used in other centers, would result in a deck at some point identified as "triumphs". What these terms actually referred to in Florence at the times of their various early uses has not been assigned a probability. Also unassigned are influences from other tarot centers that might have impacted the process, and Florence's and Milan's influence on other centers. All of this is contingent upon at least one of the hypothetical structures, mine or Franco's, having been at some time before 1440 actually existing.

Some people might be wondering about the legitimacy of this kind of speculation, about hypothetical changes not only in the order of cards, but also their very subjects, in a world in which people from one region were continually visiting those of another. and what the basis is for going back from the actual to a hypothetical with not only a different order but different cards. For such people I will have another post.