Re: Dummett and methodology [was Re: The Sun]

Hi Marco,

Sorry for the delay in responding. You know... the holidays. They take their toll.
marco wrote: Isn't the Devil “real but untouchable”?
Proverbially, he is touchable I think. You can meet him, he takes on various disguises, that sort of thing (but so are angels (and Jesus himself is a touchable man) so couldn't an angel go in the sub-celestial section?). But that's not the point - I'm sorry if those descriptions can be misleading - they are not meant to be strictly diagnostic - I didn't make them up and then put the cards where they fit the definitions. Maybe they could use some refining, but they are meant only as broad descriptions of the rationale after we have the threefold analysis in place.

The best way to explain my reasoning for putting the Devil with the other "dark cards" might be to show chronologically how it developed.

First, until about October of this year I took the linkage of Devil and "Tower" (and afterwards for awhile even when I had come to the conclusion that the card's subject was only the Thunderbolt) as axiomatic. It is a seductive idea when you are looking for a story in the sequence. It just seems absolutely right that God would be blasting the Devil, and after that we just start going to heaven.

The other big appeal is the dichotomy of the Devil at one end of the section, with the Angel at the other. Absolute darkness - absolute light. It seems like that section is wrapped up.

But looking back I can see that some doubts in this picture were sown by the early interpreters - neither Imperiali nor the two discorsi make this connection - none of them imply that there is some kind of judgment going on here. In fact Imperiali and the Anonymous put him in the moral section (Imperiali puts him in both, I suppose; he wasn't trying to limit himself to sections, but to make a seamless narrative of the whole sequence). That these 16th century interpreters didn't make the Devil-Lightning connection should have given me pause earlier, but since I went back to them only after I had made the break between these two cards myself, while I can say that I remembered them and appreciate them more now, they were not directly instrumental in changing my views.

The first step came with my growing discomfort with the force of that particular "narrative nugget" in an otherwise less than catechetically clear section. Why such an explicit display of high and serious religiosity in a set of pieces only designed to be easily remembered in sections? Where does it go now? To a Star - is this a symbolic reference to Jesus? But then who are the Moon and the Sun, etc? It just didn't make sense that the designer would be so explicit here but so allusive in the rest. This thought occurred in concert with my increasing methodology to divest the sequence of all narrative traces (the first to fall for me was bringing Love into the Chariot with Virtues section - I had to get rid of the part of my interpretation that said Love is the last card of the first section, and refers to the proverbial expression "Omnia vincit amor"; hence it was a "narrative" that it completed; with Love's narrative role gone, the Devil-Lightning pairing stood out even more painfully).

The Devil was already a "dark" card, along with the other 3 cards Morte, Traditore and Vecchio, so I thought, why not just count him as one? I imagined then that the players could simply say "the dark cards" and think "Death and Devil" or "Death and Hell", a conventional expression, and a logical set.

Since I'm working with the Bolognese tradition, I took it as confirmation that they too, like Imperiali and the Anonymous, saw it this way - in this group - when I noticed that the "bursts" of light in the corners of various cards - across 400 years and many different designers - remained constant, and that the Devil is never shown with these bursts, while the Saetta is always shown with two.

Bolognese "light" and "dark" cards from 16th to 19th century: ... htdark.jpg
(click for a larger image of course)

This last "argument" is not decisive, but it was the "straw that broke the camel's back" for me in divorcing the Devil from the Lightning.

Here is another version explaining myself that I wrote when you and Mike brought up my placement of the Devil in the middle section a few weeks ago -

The Devil

Mike and Marco have noted that I have included the Devil in the middle section, differing therefore from either of Dummett's groupings. Since it is new, I should explain it. It came as a shock to me to do so, and it was the last card to "fall", so to speak, out of the tendency to see a narrative in the sequence. The penultimate for me was the Love card, with the narrative value of "Amor vincit omnia", that I held for quite awhile, as the end of the first section. Now it sits firmly with the Chariot-Virtues section. But the Devil stayed stubbornly attached to the Thunderbolt, with its dramatic prop the tower, the pair indicating a narrative aspect of divine judgment.

As late as September, I was still working under this model for both Love (including a Petrarchan aspect) and Devil, as you can see in one of the schematics I made of "The Conceptual World of the Trump Sequence", where I instruct myself, when making the finished version, to make a "sharper spatial distinction between Death and Devil." ... bleau3.jpg

I was thinking in terms like these (from the notebook at the time): "Designer placed Lightning Bolt after Devil partly for symbolism of the lightning for them, which is God's judgment, the fulmen of Jove. So it symbolizes God keeping the Devil down
- Tullius Hostilis (struck by lightning)
- Quote Valla
- Thomas Cantimpré
(Este ms. too; and Botticelli Dante on Nimrod)"

Within a week or so of that, the "lights" interpretation of the cards Lightning, Star, Moon and Sun (I was still trying to include the World) occurred to me, and I figured that the Devil cannot be considered a "light" card.

In November I wrote: "The final breakthrough came with the Devil, when I realized he was one of the "dark cards", not connected with the Lightning, hence not part of the last group and also removing the last "narrative" component from the interpretation."

This idea was reinforced when I looked at the popular Bolognese tradition, for the 500 or so years that we know it, and saw that, until the 20th century, there are consistently "bursts" of light in the corners of the Lightning card, but never on the Devil, or Death, or the Traitor, or the Vecchio, so that I began to think of them as the "dark" cards. These bursts were consistent across time - over four centuries - and different designers in Bologna, so that, like the equally-persistent distinguishing color-scheme of the court cards in Bologna (Black for Spade, Green for Bastoni, Red for Coppe, Yellow for Danari), I felt justified in thinking they were somehow important to the players in Bologna, perhaps as indices of some sort. Whether indices or not, it was enough to break the relationship of Devil with Saetta, and put him with the "dark" cards instead. It is not difficult to remember his ranking in any case, but I suggest that the proverbial association is the phrase "morte ed inferno", with the Devil as the representative of Hell.

Please note that "dark" and "light" are my own terms of convenience, to show how I separated the Devil from the Lightning. While I think that the designer did in fact think of the "light" cards as "The Lights", I don't think he contrasted them conceptually with the immediately preceding group, which is part of the middle, moral section; rather, I think he wanted players to think of them as "bad" or "fate" (as in fato, with the same connotations as the original sense of our "fatal") in contrast with the "good" section, for easy remembering. So Ruota, the Wheel of Fortune, is one of the "fatal" cards, despite having bursts of light in the corners in the Bolognese tradition.

Re: Dummett and methodology [was Re: The Sun]

Ross wrote:Within a week or so of that, the "lights" interpretation of the cards Lightning, Star, Moon and Sun (I was still trying to include the World) occurred to me, and I figured that the Devil cannot be considered a "light" card.
.... :-) ... what about a light state like "no light" or "black hole in the mid of the galaxy, eating light" (each galaxy is assumed to have a black hole in its middle, as far I understand it) and "Demiurg" or "Demorgogon", who created the world?

If we take 16-19 (Tower-Sun) as a group, then it's embedded between Devil (15) and Angel (20), which looks like a complementary pair, perhaps Devil creating a "nasty negative world" and Angel destroying it.

And we have 1-14 (a greater group) and 21 (= somehow "all and everything") embedding the light sequence between Devil and Angel as the rest of it. .


The most interesting 3-groups-order is in my opinion this one (though more a riddle than fact) ...

6 Love = Petrarca's Love
7 triumphal chariot with female rider = Petrarca's Chastity
8-12 (including Father Time ????)
13 Death = Petrarca's Death
14 Alciato's Fame as Temperance = Petrarcas Fame
17-18-19 = Petrarca's Time ????
20 ?????
21 World = Petrarca's Eternity

"Similarly, all Italians of the 15th century would recognize the vague hierarchy in the trumps: celestial and eternal things are higher than moral allegories, and moral allegories are higher than human stations/types. These are the three divisions recognized by Dummett, and already recognized in the 19th century (and arguably the 16th)."
I don't think, that one can reduce all and everything, what had happened on the different Trionfi decks in a meaningful manner in three parts.

Naturally one can: everything in the world has a begin-middle-end structure. A writer for instance mostly has an opening called "Introduction", a finish called "Conclusion" and a middle part ... but a deciding message of the work is also, how many chapters he had chosen to present the middle part, may it be one point or maybe 50 or 500.

The Mantegna Tarocchi has a 5x10-structure. That's the basic message.

The Ferrarese Tarocchi order seems to fill a 1-5-5-5-5-1-scheme ... another favor for the number 5.

0 Fool
1-5 Persons
6-10 Good things
11-15 Bad things
16-20 Heaven
21 World

The Minchiate operates with 5 Papi and 5 Aries. If one follows the game ideas expressed in the verzicole, then there's a general interest to honor partitions of the number 40. Only few verzicole (the choice of 28 and 13) associate the idea to part the 40 cards in 3 groups.


Sola-Busca and Boiardo Tarocchi show an interest to part the 20 "inside cards" (1-20-1 scheme) in 10 pairs (1-2, 3-4, 5-6 etc).

The general Tarot has 5 suits, not 3.
4 normal suits and a trump suit.

Re: Dummett and methodology [was Re: The Sun]

Hello Ross,
thank you very much for the detailed discussion about the Devil and "Fire from Heaven". Your point is clearer now. I would like to comment more, but I could not found the time for reasoning and documenting myself on the subject.

Something I found particularly interesting: I had not thought of the "bursts" of light in the corners of Bolognese cards as indexes. Those details always interested me, but I never tried to make sense of them. I see your point about their presence in the four "lights", but I wonder what could be their meaning in general. They seem to appear on a number of cards, so they do not denote only "light", they must have some other meaning too. For instance, why do they appear on Justice and Temperance and not on Strength? ... cture1.jpg

I am sorry, but my time is really limited at the moment. I hope to come back to the subject in the next days / weeks.

Re: Dummett and methodology [was Re: The Sun]

Marco, I wouldn't have thought twice about them if I hadn't seen them persist through time, and had a parallel "indice" system with color-codes for the Kings (which disappears earlier - both things are gone completely in 20th century decks). It's a strong presence for such incidental details, across all known designers, for 400 years (note that in the Alla Torre 17th century deck, the bursts in the Sun are painted there, but not engraved in the wood).

I've tried to deal a few hands and see if they make any difference when you fan the cards with only a little bit of the corners showing, but since I don't have the whole deck it is hard to make a good experiment.

I had a theory that a double burst generally introduces a new "section" - but this doesn't work, unless Love belongs to the Papi and the "lights" are ignored. Remember that we aren't free to re-order the cards by theory or some arrangement of these "indices" - the Bolognese order, even more than the designs, is inviolable.

Force and Sun are devoid of bursts, and they are the highest of their groups (taking the two counting cards World and Angel as the high ludic group and not as "lights", although all together these subjects constitute the third, "heavenly things" group), but the Sun doesn't really need bursts to be seen in a hand (it's bigger than the rest), even though it might help knowing Virtue from Virtue when you only have a hint to look at (Force's pillar would tell you, as Justice's sword; so the burst only serves to distinguish the upper left corner of Temperance from the other two).

The bursts of the papi would make one pope the lowest (like a "popess"), if the "double burst" phenomenon introduces a new section, but we know that wasn't the case in Bologna (and that leaves the problem of Love, which has to come below the Chariot). The two emperors have left bursts, the two popes right bursts, and even though one pope's left burst seems fainter than the other, I still see clear traces of woodcut black ink underneath, so, unlike the Sun card, that burst was meant. Unfortunately we have no other set of Bolognese papi to compare (and Mitelli doesn't care). If the "new section" idea is used, then the best order for the papi would be double-burst pope, pope, (blank matching) emperor, emperor.

If they mean anything, really and consistently, I haven't found the logic without violating the traditional sequence, which would be ridiculous for such a long-lasting tradition with a well-known order.

Thus I gave up on a "semantic" meaning for them - as I tried to say, it was valuable, but fortuitous, for me to separate the Devil from the Lightning, but perhaps nothing more.

So, for the moment, I assume a miscellaneous use, and only for play:

If you see nothing in the corner, you have a counting trump or the Sun (the "dark" cards have enough of their subjects in the corners to make them out quickly, as does Force; and the Sun is pretty easy too).

See a burst - left or right, I guess you could fan either way, I don't know of any explicit instruction on a standard practice -, and you have to look a little more closely to see what you have.

Not much use, it seems, but maybe enough for rapid habitual play, especially for knowing counting cards from empty cards.

This engraving, from the opening of Pisarri's Istruzioni necessarie per che volesse imparare il giuoco dilettevole delli Tarocchini di Bologna (1754) shows the players fanning the cards to see them from the upper left corner - ... ersmed.jpg ... detail.jpg

Re: Dummett and methodology [was Re: The Sun]

Addendum to Devil (including general questions of methodology and the three groups)

It doesn't matter whether we place him in the moral section or the third section (however you want to qualify or define it) - he is still in the same place in the sequence in relation to the Thunderbolt in every deck ever made. It may redefine his meaning if he is one or the other section, in the same way that Justice's meaning is changed when she is moved from the moral to the heavenly part (such movements prove that they did indeed read meaning into the sequence, or parts of it). I say "may" because those who made the C order, with Temperance intervening between Death and the Devil, effectively seem to have ejected the Devil from the moral section; but did they thereby intend to indicate that there is, or to create, an explicit relationship between the Devil and the Thunderbolt? Or was Temperance put there to lessen the severity of the Devil following immediately upon Death, just as Temperance appears to have been put between the Pope and Love in the B orders in order to break any implied relationship between those two cards? In this case, then, the Devil's being apparently in the last section in C (I suppose someone could argue for his place in the middle section, although it would disrupt the symmetry of this section in the Tarot de Marseille order specifically), makes him merely a casualty of someone's interpretation of the middle section, and does not necessarily imply that there was thereby to be inferred a direct relationship between the Devil and Thunderbolt in the final section.

But all the question of the Devil's sectional assignment in A or C really accomplishes is to bring us back to the position Dummett stated - that to know the original meaning, we have to know the original arrangement. It'd be too much to say that we can know what the original arrangement was, but I believe that the strongest argument by far can be made for A, and particularly the Bolognese A.

It doesn't matter if you agree with me or like my explanations or the words I use to describe them, or the insights I think they generate. What I mean to say is that it doesn't matter if I convince you with my arguments about the meaning for why this or that "section" or "subsection" exists. What is important, and what I will insist on, is that the following methodological question is sound and crucial - how did the players learn the sequence, which is, by extension, how the designer intended it to be learned?

The way the players learned the unnumbered sequence reflects the intent of the designer.

When a new player sat down at the table, never having seen the game before, they learned the trump sequence first by groups, then by habit or memorization.

The groups will reveal the intent of the designer. The specific order of the cards within the groups will be the designer's choice, but may not mean anything beyond his own convenience or whims. These may be unknowable.
It seems to me highly unlikely that the new player sat down and was shown the cards and told "Angelo Mondo Sole Luna Stella Saetta Diavolo Morte Traditore Vecchio Ruota Forza Justizia Tempra Carro Amore Quattro Papi Bagattino Matto - okay, your deal". Nor was he sent to a corner and told "come back when you have it memorized, and then we'll tell you the rules".

In Bologna, the players learned at the table, for the first time, for over 300 years. There must have been a method of groupings to learn the trumps. It is the only way to learn it quickly enough to join in common play, when no written rules existed and there were no numbers on the cards. However, since the groupings or mnemonic did not make it into the earliest written or printed accounts of the rules, they must have been easy enough, and banal enough, that they were as quickly learned as forgotten.

First prediction - somebody of all those players over 300 years must have jotted down an account of their first game of Tarocchi, including an account of the mnemonic. This will reflect the designer's intent.

Second prediction - such a jotting will probably survive in a diary, letter, or a single page inserted into a book somewhere. It will probably, but not necessarily, be in Bologna.

The second prediction is based on conservative calculations that of 10 people per year learning the game, 1000 in a century, at least 1 per cent will have written something about the experience, that is 10. Therefore 30 in 300 years, of which it seems probable that at least a handful will survive somewhere. These numbers are surely too conservative, but the term "handful", suggesting less than 10, is surely probable. It is therefore worthwhile to think of ways to search for unpublished diaries and letters that may contain such an account, in Bologna. Of course, it will more probably be an accident that brings it to light, but I believe - predict - that it will be there, somewhere. If the Bolognese game is the original game, even if invented in Florence, and if the account is early enough (or is otherwise convincingly reflective of the earliest game), then we will be as close as predictively possible to the designer's notes for the game. If the game were invented in Florence and such a record for the period before Germini became standard is found, then the same may be predicted.

I've had time to prepare for utter rejection from everybody, even people whose opinion matters, and it still doesn't matter if my wordy explanations convince or not. It doesn't even matter if they are fully right or not. My position, my methodology, is inassailable, mainly because it is defending so little. It boils down to the fact that the cards were made to play with, the sequence to be memorized quickly, and the system had to be easy. It had to be fast, and banal enough that it was done so quickly that it was just as quickly forgotten as it was unnecessary, which is why no mention of it has been found yet. It had to be banal. It had to be groups. All we have to do is figure out what the groups were, assign some snappy names to them, and start to play. The order of the particular cards in them is easy once the groups are down. Why the designer chose those groups is explained by the three classes of subject matter, which I have called for convenience - snappy - earthly, moral, and heavenly. They learned the highest and lowest group first, not an iconographic group, but the only group that mattered in practical terms because it was the only group that counted, literally. The iconography of this ludic group - highest and lowest - then matches where they go in relation to the rest of the cards in their part of the threefold iconographic divisions. This means low people and big ideas.

Why the inventor chose the rest of the particular subjects, or rather why those subjects were obviously groups for the original audience, is the more difficult business, which has to be argued carefully (and usually clumsily by me), from art history and documentary history. I think I've got a good explanation for all the sections, and why each subject would have made clear sense in the group and fallen immediately into its (sometimes arbitrary) place in the group for the original audience, but I know my rhetorical skills are frequently not up to the task of convincing my own audience. Now that I'm standing on firm ground, however, I'm not afraid to keep trying.

Re: Dummett and methodology [was Re: The Sun]

I need some clarification, Ross. When you say:
They learned the highest and lowest group first, not an iconographic group, but the only group that mattered in practical terms because it was the only group that counted, literally.
Do you mean the bottom five and the top five? That's what counted in the scoring, in all the accounts of rules that I've seen (considerably after the 15th century, unfortunately; and even then, there was also the Fool, and the Kings, and combinations). If so, memorizing them would indeed be easy, especially if the papas weren't in any predefined order. Other than that, you'd have the bagat, who has his place in the order right in his name. On the other end, you have the hierarchy of lights, plus judgment and world. All you have to keep straight is which comes first of the last two, which indeed looks rather arbitrary in the Bolognese deck. If not, say again why a guy with wings on his hat and shoes should come before a trumpeter.

Then, yes, you have a good design for play by the masses, either an improvement over the original or a good original from the first. You still don't have an ur-tarot, because you don't know if the original designer thought of your point, or gave it your prominence. He or she might have been thinking of other things besides ease of memorization. It seems to me that if ease of play had been the only objective, the designer would have thought to put numbers on the cards.

On the other hand, if the game was already out there, and someone got the idea, let's still not have numbers, but let's make it a little easier, with the cards that count the most in the scoring, then he could make some money out of such a deck as the Bolognese, at least in places where a different deck and order wasn't already in people's heads (or they had no other choice, due to political pressures).

But perhaps you weren't after an ur-tarot. Maybe you were just after one reason why the "four papi" rule and the "three lights" before the last two might have been introduced in particular markets, setting aside the question of whether it was the ur-tarot or not. If so, there's no problem, a very minimal claim. And a good methodological point. Please clarify.

Re: Dummett and methodology [was Re: The Sun]

It might well be, that producers didn't use numbers on the cards, cause they were aware of different rows used by the players.
Everybody, who needed numbers to remember a hierarchical row, could add the numbers himself.

For the producer it was less work.
For the players it meant more freedom. They could buy Bolognese cards, but play a Ferrarese game, if they were used to the Ferrarese row.

"Doppelkopf" - a German game with possibly 200 years history, developed from Schafkopf - has a lot of variants, which might be easily more than 100 or even 200.
When players sit down to play, and they played never together before, they have to set up the rules, according which they want to play. That might be a matter of a few minutes.
Recently - maybe 20 years ago - a Doppelkopf-Verband has built which also plays tournaments, championships etc.. Naturally they needed fixed rules, which were accepted by all players, who participated in this organization.

This didn't change the usual habits of old friends, who gather occasionally or once in a week. I myself frequented a pub, where it was played nearly all days, 1-3 tables. Once in a year a greater tournament. They played with rules, which I've never seen elsewhere.

Doppelkopf has features, which are similar to the rules in the Austrian Königrufen, which is a Tarock variant. John McLeod expressed the opinion, that Königrufen would be perhaps the most interesting Tarot game.

Some of the possible variations are so strong, that they indeed change the whole game and its strategies.

It's difficult to make conclusions on ideas "how players will behave". One has to gather a lot of personal experience and there's a big group of different games and one will never know all.

Re: Dummett and methodology [was Re: The Sun]

mikeh wrote: Do you mean the bottom five and the top five? That's what counted in the scoring, in all the accounts of rules that I've seen (considerably after the 15th century, unfortunately; and even then, there was also the Fool, and the Kings, and combinations).
I don't know any game at all which scores exactly like you describe, Mike. It sounds like an interpretation of the rules given by Kaplan I, p. 295, which comes from Mathers' The Tarot (1888), via Singer Researches into the History of Playing Cards (1816) pp. 238-239, whose source is Court de Gébelin, Monde Primitif t. VIII (1781), pp. 381-383, which he says he was taught by "M L'A R.", who was probably "Monsieur L'Abbé Rive", i.e. Jean-Joseph Rive (1730-1791), author of Eclaircissements historiques et critiques sur l’invention des cartes à jouer (Paris, 1780), and a subscriber of Court de Gébelin's Monde Primitif.

Court de Gébelin's version of the game is described by Dummett in Game of Tarot (GT), pp. 232-233, and Dummett and McLeod in A History of Games Played with the Tarot Pack (HGT), pp. 39-41.

Court de Gébelin's game is hardly standard or representative of all Tarot games, so if that is all you know about, I think you need to get back to the books.

Start with GT chapter 8 (esp. p. 199r);
Tarot: jeu et magie p. 19;
Twelve Tarot Games pp. 16, 26, 37, 59, 79, 102, 116, 151, 177, and 196;
Il Mondo e l'Angelo chapter VI (esp. pp. 149-150);
David Parlett, A History of Card Games chapter 18 (esp. p. 241);
and HGT chapter 1 (esp. pp. 6-7).
Or, start with John McLeod's introductory page on Tarot games at -

The standard counting trumps are the World, Bagat and Fool. In Bologna, it is Angel, World, Bagat and Fool. Then the 16 court cards, with Kings valuing the same as the counting trumps. The standard valuations are:

5 or 4 points for each of the counting trumps and Kings;
4 or 3 points for the Queens;
3 or 2 points for the Knights;
2 or 1 point for the Valets.

In Minchiate and Sicilian Tarot, some other trumps are worth points as well.

In Minchiate (GT p. 344, HGT p. 330) the valuations are:
10 points for each of the arie (5 highest);
5 points for XXXV to XXX (Gemini to Cancer);
5 points for XXVIII (Capricorn), XX (Fire (element)), XIII (Death) and X (Chariot);
3 points for the papi V to II (Love and the three rulers);
5 for papa uno (Bagat) and Matto;
5 for the Kings.

Thus: 10 for the 5 highest trumps, 5 for the other counting cards, and 3 for papi V to II.

In Sicilian Tarot (GT p. 380, Twelve Tarot Games p. 102, HGT p. 377), the valuations are:

10 points for trumps 20 (Giove), 1, the Picciotti ("little one", equals French "petit"), and the Fuggitivo (Fool).
5 points for trumps 19-16, and the Kings.
4 points for the Queens.
3 points for the Knights.
2 points for the Ladies.

The Three Worlds Model

Hi, Ross,

Happy New Year!

Thanks for the posts -- a nice present for the holidays. Like Marco, I really wanted to comment earlier but, you know... holidays. Plus, once I got started rambling... well, you know about that too. So apologies for the delay, the length, and the rambling.


As with Howard's post, your posts are a great opportunity to reiterate, ruminate, and perhaps clarify some of my favorite topics. In particular, this seems like a good time to emphasize the distinction between Dummett's findings and my own conclusions. Because attacking Dummett with strawman arguments is a long-standing Tarot tradition, I have tried to avoid giving people more opportunity for that. Because virtually no one agrees with my conclusions whereas his findings are factual, he should be defended against such conflation. A second big section is more to the point of your post. Yes, the Three Worlds schema (using John Shephard's term) is common, and recognizing that division in Tarot is the essential first step to understanding the cycle as a whole. However, you seem to be going a bit overboard with it, as if announcing the arrival of a new Universal Monomyth. It's a common structure, but not universal nor always well defined. So... basically, I'm just quibbling about emphasis and elaborating a bit. The main elaboration concerns Shephard's version of this and his terminology, as I have used them for a decade now. I'll talk about the Devil as one of the middle trumps in a separate post.

Most of the points to be commented upon are naturally disagreements, so it is helpful to keep in mind that we are, in most ways, substantially in agreement. Regarding Dummett's "three distinct segments" and Shephard's Three World model, (and my own 6/9/7 model), the discussion here is vastly more salient than most of what passes for Tarot history, here or elsewhere. I've spent many years attempting to get this topic front and center, with essentially zero success. I'll pick up on that below, but the point here is that significant questions are being debated. Dummett's findings (which were admittedly very limited, as discussed below) on the three sections and my own much broader conclusions about a 6/9/7 design to the trump cycle, are being considered (albeit mostly rejected) rather than universally ignored. Huzzah!

Regarding disagreements between you and I, these tend toward the trivial when taken in the context of 21st-century Tarot discussions generally. We're talking about historical Tarot rather than talking about Egyptian symbolism, Lombard pawnbrokers, alchemy, dowries, numerology, Albigensian heresies, Chess, or the Battle of Anghiari. For many years, both of us have taken the position that Tarot was, first and foremost, a card game. The content of the trumps had a two-fold purpose, as reasonably memorable and immediately recognizable tokens for cardplay, and as an inspirational subject matter to elevate the game. This is in keeping with other moralized games. Today, most people here would give grudging assent to that, before moving on to yet another revision of traditional occultist/New Age folklore or their most recent confection of historical fiction, weaving real Tarot history into a tapestry with famous people and events.

We have both attempted to construct a less fanciful, more historically sound reading of the trump cycle. Again, most folks here would claim to be doing sober analyses, both historically and iconographically, even though most of the discussions are laughably unrelated to Tarot or based on recycled esoteric folklore. Both of us are strongly inclined to respect the findings and consensus opinions that make up what you have called the Standard Model of Tarot history, dissenting only when there is a particularly strong and clear argument against some detail of it. When we do indulge in leaps of speculation, we are usually pretty clear about it and try to avoid building towers in the air, speculation based upon earlier speculation which is mistaken for factual foundation. Speculation one step beyond the facts, for the purpose of explaining those facts or answering another legitimate question, is very different than speculation which is two or three steps removed, piled up into a whole world of fantasy.

So, compared with the 99.9% of Tarot enthusiasts who do not share our common view of this material, we are substantially in agreement. For the half dozen or so (maybe a dozen?) who share that Standard Model view and commitment to less-speculative readings of the trumps, we are using the same approach but reaching different conclusions.

Ross G. R. Caldwell wrote:A depiction of heaven and earth will place heaven above earth; everyone intuitively understands this. The things of heaven are higher than the things of earth. So, in a vertical hierarchy, where would you then put depictions of concepts and ideas, or personifications of moral concepts like allegories? Since they are not people you can meet or things you can bump into, nor are they real but untouchable like the heavenly bodies and supreme realities like God on his throne, such symbols will be placed higher than earth and Man but lower than heaven and God, in the middle space between the two. This is what the tarot trump sequence does, and so do countless other works showing formally similar hierarchical orders. Only the specific iconographic content will differ, depending on the context and function of the work. This is the threefold scheme of the trump sequence.
...we can say that the sequence of the remaining trumps falls into three distinct segments, an initial one, a middle one, and a final one, all variation occurring only within these different segments.
Game of Tarot, p. 398.
The threefold architecture of the trump sequence that Dummett discovered...
We need to clarify this, lest we libel the dead by attributing to him ideas he most certainly did not share and explicitly argued against. In a sense he found that design, stumbled across it, but failed to see it. Dummett drew attention to the two dividing points, but he did not suggest, much less argue for, three groups of trumps in a meaningful hierarchy. He was analyzing the various sequences for taxonomic purposes, to help understand the diaspora of early Tarot decks. That is, Dummett was doing exactly what Howard claimed he did not do, using the different designs to understand the history of Tarot's spread. The variations in ordering are a bizarre fact of Tarot history, obviously at odds with the usual practice of conservative game play and requiring some explanation.
Michael Dummett wrote:If play is to be possible, the ranking of the trumps must be apparent to all players and subject to no dispute. How, then, are we to explain the variations that we find in the order of the trumps in Italy from the fifteenth to the seventeenth century?
(Page 396.)
Dummett then considers various arguments, not relevant here, and concludes that the variations reflect an extreme localization of Tarot game play. This is the passage which leads up to the quote above, and explains his interest in the analysis.
Michael Dummett wrote:The observable variations in the order must therefore be due, not to the absence of a fixed order, but to that phenomenon evident throughout the entire history of the game of Tarot: the extreme localisation of specific modes of play. Again and again we find that the players in one city or town play only amongst themselves and do not know those of a neighbouring town; the detailed rules, and sometimes the whole type of game played, diverge from locality to locality, the players in one circle being quite unaware of the manner of play of those in another, and, often, of their very existence. The different orders for the trumps that we find in Italy must represent different practices adopted in different cities, presumably at a stage earlier than that at which numerals came regularly to be inscribed on the trump cards. Evidently, quite a short time after the game of Tarot had first been invented, players in various cities or regions developed local peculiarities in their modes of play, which, in Italy, extended to the conventional order of the trumps; this must have happened before it became usual anywhere to inscribe numerals on the trump cards, and hence before the end of the fifteenth century. Modern players might feel that it would be impossible to memorise the order of twenty-one trump subjects so accurately as to be at once aware, without the need for reflection, which card was superior to which; but, as we know from the Bolognese game, this doubt is quite misplaced. The different orders of the trumps testify, not to a reliance on the numerals alone, but to the existence, at an early date, of wide local variation in the manner of play.
Note that this is an observation of fact rather than an explanation. Dummett does not offer an explanation for the deck and game being altered in almost every locale, other than a passing comment about the absence of indices. They developed "local peculiarities", in every locale, for no apparent reason. Outside of Italy, this level of variation in decks did not happen.
Michael Dummett wrote:When we look closely at the various orders, we find that there was far from being total chaos. A first impression is of a good deal of regularity which, however, is hard to specify. Now the cards which wander most unrestrainedly within the sequence, from one ordering to another, are the three Virtues. If we remove these three cards, and consider the sequence formed by the remaining eighteen trump cards, it becomes very easy to state those features of their arrangement which remain constant in all the orderings. Ignoring the Virtues, we can say that the sequence of the remaining trumps falls into three distinct segments, an initial one, a middle one and a final one, all variation in order occurring only within these different segments.
Immediately after this analysis of the variations, he uses it as the basis for his further analysis of three regional traditions. This is the context of his division of the trump cycle into three segments. He was not doing iconography, at least not intentionally.

We also need to keep in mind that Dummett did not include the Fool, nor the three Moral Virtues, but only "the remaining eighteen trump cards". This is not a problem for him, as he was not attempting to explain the trump selection or arrangement. More importantly, he did not explain his findings, not even in the FMR article. Why are there these "three distinct segments"? This is a place where speculation, specifically about the design and intended meaning of the trump cycle, can be used to explain the facts. As usual, however, Dummett declines to speculate.

That is Dummett's version of a "threefold architecture". He did not deal with the lowest trumps as a group, at all, and his vague characterizations of the middle and highest trumps were essentially correct but largely unhelpful. No one -- not one person -- ever followed up on his findings and characterizations. (Excepting me, of course. BTW, if you know of anyone who pursued this analysis between 1980 and 2000, I would very much like to know about it, to be able to cite them on it.) Dummett's "three segments" analysis was not an iconographic explanation of the trump cycle, but a finding of facts which remained in need of an iconographic explanation.

It was a couple decades after Dummett, when I learned about his analysis (indirectly, via a summary by Depaulis which Christian Joachim Hartmann posted (June 11, 2000) to TarotL), that I pointed out that it was consistent with my own iconographic explanation of the meaningful design of Tarot. His analysis and my own both identified the dividing points of the Pope (and lower cards) and the Devil (and higher cards). My iconographic analysis explained his historical findings, while his historical findings were consistent with, and thus tended to support, my iconographic analysis. These groupings are another example of Ong's schematic relationships.

It has been pointed out that Dummett's second division, at Death, seems arbitrary in terms of the variations in ordering. That is, in terms of his stated methodology, Dummett's findings are even more limited than they initially seem. Because the highest trumps were in the same order, from the Devil through the Sun, in all decks, he could have chosen a different boundary. Dummett saw no design to the lowest trumps as a group and even excluded the Fool, while his methodology did not actually provide a dividing point for the highest trumps. Dummett had to take meaning into account, at least implicitly: Death is kind of a big deal, the end of this life and the beginning of the next. So he recognized that Death can reasonably go in either segment, as the end of one or the beginning of the next. Because he was just trying to do taxonomy, as prelude to understanding the early dispersion of Tarot, he could profitably ignore these "problems" -- they weren't his problems. His analysis has essentially no explanatory power, and I cite him because his analysis does tend to support something which occultists and their contemporary apologists cannot conceive: divisions other than 7/7/7.

Conversely, because I was explaining the entire 22-card cycle of images, my analysis was into groups of 6/9/7. Although Dummett did seem to glimpse some of the iconographic significance in his FMR article, he offered no such interpretation. For better or worse, this explanation of the design of Tarot is entirely my own. Here's the point: although IMO his analysis is consistent with my own and tends to provide historical evidence and support for it, there is no reason to believe that he would have agreed, and therefore he should not be blamed for my views. Given the fact that virtually no one accepts my analysis (Marco and Hendley being the only two I can think of), exonerating Dummett seems important. I was so concerned about misrepresenting him that, before I first posted The Riddle of Tarot online, I asked his friend and co-author John McLeod whether he thought that citing Dummett in this manner was misleading, or whether my presentation misrepresented Dummett's position. I also included this disclaimer in the first footnote:
Footnote A wrote:It should go without saying that although this presentation makes many references to Dummett’s analyses in The Game of Tarot, it seems unlikely that he would endorse such a (mis)use of his work. In quoting him at length, I do not intend to imply that he would agree with any of my own conclusions, any more than I agree with his suggestion that the images are a kind of sampler of triumphal images. He refers to Gertrude Moakley’s “brilliant suggestion” that the name trionfi derives from the subject matter of the trump cycle, but admits that it is difficult to discern what the underlying concept of the triumphal sequence might be. (GT 87.) Elsewhere, he writes that the designer selected “a number of subjects, most of them entirely familiar, that would naturally come to the mind of someone at a fifteenth-century Italian court”, and that “it is rather a random selection”. (GT 387.)

Moreover, most of my interpretation was developed before I even learned of his analysis of the early decks. I arrived at my interpretation based solely on the iconography and sequence of the cards as described and illustrated in Kaplan’s Encyclopedia. I have chosen to present my interpretation in the context of Dummett’s analyses because he has written the most comprehensive and reliable book on Tarot, which establishes many historical constraints for any study of early Tarot; he has framed the null hypothesis of a triumphal sampler and demonstrated its sufficiency for most historical purposes; he has pointed the way toward a study of sequential meaning, and done all the hard work of discovering and analyzing the early trump orders; and in my opinion, his findings support my conclusions... but that certainly doesn’t mean that he would agree.
At the same time, I wanted to credit him and use his finding to support my own analysis. One entire section of The Riddle of Tarot is devoted to "Grouping within the Sequence", and it was written around his analysis. And there was this additional caveat:
Footnote C wrote:Although the present interpretation is quite different from any earlier attempt, both in its approach and its results, some aspects of it have precursors. Most notably, Michael Dummett analyzed the trumps into three groups, which are precisely the same as the groupings used here. The iconographic interpretation into three types of subject matter (the “Three Worlds motif”) is confirmed by the actions of fifteenth Italians who re-ordered the trumps, as shown by Dummett’s analysis: in each case, they maintained the division into three groups, reflecting three types of subject matter which they considered distinct and kept separate.
I have been promoting this as a sine qua non first step in understanding the trump cycle for many years. The observation that there are three different types of subject matter answers a host of common questions such as, "if Tarot isn't heretical, then why does _______ trump the Pope?" (Merely displaying a Triumph of Death image or Dance of Death cycle also answers that question, but some people are just too fucking stupid to understand any kind of answer.) That is why I love to see others considering the Three Worlds model, in one way or another, rather than endlessly recycling traditional occult and New Age folklore or imposing topical irrelevancies (like Lombard pawnbrokers or the Battle of Anghiari) on the trumps.

To this day there are very few who agree with me on the 6/9/7 analysis of the trump cycle into major sections. Marco and R.A. Hendley seem to be the only two who have repeatedly gone on record as saying this is a fundamental part of the design of the trump cycle, and presenting their own versions of it. Hendley, in a variety of brightly illustrated posts, has (largely) followed my own analysis and, on at least one occasion, cited my earlier posts. Marco seems to have a number of ideas very close to my own, as well as some differences. In particular, both have presented the same 6/9/7 division of the trumps into three sections, the macro analysis into three types of subject matter which I've been promoting since 2000. In my first detailed presentation of the idea I described the three sections as 1) Religion Triumphs Over Society, 2) Virtue Triumphs Over Circumstance, and 3) God Triumphs Over Death and the Devil.

Looking back at some 2009 threads, you, Robert, and Kwaw (at least) preferred a 4-group system, 1/5/9/7. This is more in keeping with Dummett, who also left out the Fool, (and the three Virtues), and failed to categorize the trumps as any sort of unified whole. Now you have concluded that the Devil is not part of the highest section. As noted at the top, you are attempting something similar to what I have tried to do but as a less detailed interpretation rather than a card-by-card programme. Like Dummett, you don't seem to think such a specific explanation is appropriate -- the trump cycle wasn't that precisely structured.

Just to be very clear: IMO the two viable candidates for an Ur Tarot ordering are the Bolognese ordering (first column, GT 399) and the Tarot de Marseille ordering (fourth column, GT 401). You prefer the former, while I prefer the latter, but both are vastly more defensible than any of the dozen other orderings. Likewise, the two best interpretations of the trumps are yours, a somewhat more structured version of Dummett's triumphal sampler or vague hierarchy of common subjects, and mine, a well-defined and perfectly structured moral allegory. Moakley comes in third, perhaps tied with Paul Huson's 2004 discussion of medieval drama and the Four Last Things. So again, although I disagree with your conclusions, this is not in any way comparable to my "disagreements" with the occult apologists or New Age neo-Jungian simps.

Ross G. R. Caldwell wrote:The threefold architecture of the trump sequence that Dummett discovered is not just a quirk of Tarot - it is a basic principle of spatial organization in iconographic vertical hierarchies - the basic moral valuation of hierarchical space, the low, middle, and high places; the center, and the sides....
Yes, but it seems as if you may have gotten a bit carried away in the presentation. This is reminiscent of Joseph Campbell's reification of his monomyth. It became something universal, which it could not be, and therefore the complexities and uniqueness of particular narratives had to be ignored or distorted so as to fit his Procrustean over-simplification. One size does not fit all, and there are crucial differences which tend to be ignored or falsified in the quest for universals.

Such generalities are good to note, useful conventions like many others. The real world and its art tend to be messy. Some works clearly display such organization, others don't.

John Shephard's Three Worlds model (The Tarot Trumps: Cosmos in Miniature, 1985) is very closely related to your construct. He uses it to describe the same aspect of the trump cycle, and like you he also emphasizes the parallels with the E-Series prints. He introduced it, appropriately enough considering his theories, via the cosmological spheres of that cycle. The four sub-lunar spheres, lowest in the Aristotelian cosmology, he termed the Realm of Man. One of those, the sphere of Fire, he also called "a middle term, a middle world in the chain of being between the aethereal world of Heaven and the mundane world of Earth". (Clearly, the cosmological parallel is being abused here -- it just didn't fit as he wanted it to.) In art, specifically the E-Series, this middle section "consists of figures all of which are allegorical. It deals with the spirit in man, the soul from its entrance into the human body", and he calls it the Realm of the Soul. The highest level "consists of figures of a theological or celestial nature. It shows the life beyond Death, with the Devil and Tower representing the fate of the imprudent soul consigned to Hell, while Judgement and The World show the Resurrection and the proper reward in Heaven for the virtuous. It corresponds broadly to set A of the Mantegna series, the Heavens. I call this set the Realm of Eternity." Despite a lot of confusion and occultist preconceptions, Shephard came close to getting this part of his structural analysis correct.
Shephard, blinded by traditional numerological preconceptions, excludes the allegory of Love from the rest of the allegorical figures, and wants to rename Time/Hermit as Prudence, an allegorical figure, but also to rearrange the sequence to exclude him from his proper place with the other allegorical figures. He wants very much to force the trumps into septenary groupings, as had generations of occultists before him. Still, he came closer to sorting out the macro structure of Tarot than anyone else, and his Three Worlds terminology and descriptions are excellent. Like Dummett, he advanced the discussion a bit -- we should look for and acknowledge such progress. Also like Dummett, no one attempted to advance Shephard's analysis further, although endless variations on the occultist septenary analysis continue to this day. We are not the first to look at this stuff, and both these guys from the 1980s should be remembered by those who are now discussing this topic, three decades later.

Here is a 2001 TarotL post which makes the same point about three realms or worlds, each with a different type of subject matter, connecting it with Dummett/Depaulis and once again using the E-Series prints as the primary cognate example.

The Virtues of Tarot de Marseille (March 29, 2001) ... sages/9532
Ross G. R. Caldwell wrote:Court cards not numbered; rank is implied. Similarly the court cards don't tell a story, although they are a meaningful hierarchy. Similarly for Chess figures or any game with symbolic figures representing the hierarchy.
Good point -- another schematic grouping with hierarchy. Pips were not exactly numbered either. That is, they did not have numerical indices, (contrary to some things one might read on this list, from those who have not read Dummett). Their rank was illustrated by picturing a number of items. Other things were significantly grouped, as well. The card player knew that in the round/female suits the greater the number of items depicted, the lower the rank, while in the linear/male suits the larger the number of items the greater the rank. Just as pawns are of a different kind than the second row of chess men, so pip cards are different than court cards. These are more examples of Ong's schematic groupings which are generally recognized without comment.
Ross G. R. Caldwell wrote:"It would be absurd to suggest that the order of the court cards was invented to teach how a court was organized. It would be absurd to suggest that the Chess pieces teach how a kingdom is organized. Both can do that, in a vague way, but that it not their intention. The understanding of the hierarchy is implicit, expected of the audience. Similarly, all Italians of the 15th century would recognize the vague hierarchy in the trumps: celestial and eternal things are higher than moral allegories, and moral allegories are higher than human stations/types. These are the three divisions recognized by Dummett, and already recognized in the 19th century (and arguably the 16th).
"The nature of the differences among the various trump orders shows that there was a broad understanding of these three divisions, and the overall hierarchy."
We should tread gently on the "recognized by Dummett" part, even though I've put it in similar terms in the past. As noted above, he did not appear to recognize the overall design toward which he had pointed. "The question is whether the sequence as a sequence has any special symbolic meaning. I am inclined to think that it did not." When I am just a lone voice in the wilderness, I tend to take polemical and even hyperbolic positions, but if there are others with a similar view, it becomes more important to state things a little more precisely. In general, however, that's always been my position with regard to Dummett's findings.
Riddle of Tarot wrote:The iconographic interpretation into three types of subject matter (the “Three Worlds motif”) is confirmed by the actions of fifteenth Italians who re-ordered the trumps, as shown by Dummett’s analysis: in each case, they maintained the division into three groups, reflecting three types of subject matter which they considered distinct and kept separate.
Ross G. R. Caldwell wrote:"Threefold scheme of art; planets (children), Schifanoia, pictures of people having visions, etc. Pseudo-Mantegna perhaps most relevant scheme:

1. Ranks
______ __Muses (poetry, art)
2. Ideas --- Liberal Arts (science, intellect) _________Virtues (morality)
3. Celestial
Yes, the 3-part analysis is one of the most basic. More basic still is the 2-part division into this world and the next. Representatives of Mankind and conventional personifications need not be distinguished; Everyman is himself an allegorical personification. Likewise Emperor and Pope, when they represent all secular and clerical people. All the changeable sub-lunar parts of the cosmos, or the circumstances of this life, including representatives of Mankind constitute the lower, mundane world, while the rest of the cosmos and judgment by God are of the higher, divine world. In a cosmograph, the Moon is the obvious dividing point, given that it is changeable like the sub-lunar world even though it is part of the eternal/celestial realm. In a vita humana with personifications, Death is the obvious dividing point, leaving one world and entering the next. The distinction between the Realm of Man and the Realm of the Soul, while appropriate to some works, is routinely ignored. The most prominent example is the Dance of Death where the Ranks of Man is completely joined with the personification(s) of Death.

If present, the lowest level may be illustrated in various ways. Not only are most Ranks of Man motifs unique, but there are other forms this level may take. Essentially, Shephard's Realm of Man depicts some representation of Mankind or Everyman. For example, the lowest section may show Adam in some form. Etymologically and in many contexts, Adam=Mankind, and his sin condemned all. Showing the Creation or, more commonly, the Fall of Man via Adam and Eve, indicates that the overall story is about Mankind and specifically post-lapsarian Man. Rather than a ranks or stations/estates motif, Mankind as a whole may also be indicated by showing representative ages. This shows Mankind and also his life cycle, just as a Ranks shows Mankind and his various places in society. Both are sketchy, schematic abstractions which employ something like synecdoche, representing the whole via characteristic parts. The two most extreme versions of that are showing just one figure, Everyman, and showing just two figures, lowest and highest, in the figure of speech called merism. Most often, women are implied rather than being explicitly depicted but, again, things vary and are sometimes quite messy.

The middle group may also take various forms. Commonly, a grouping of Deadly Sins could be used to symbolize the things of this life. We suck, hence the need for Grace. The opposite, Virtues, could be used just as well, as could a combined Vice/Virtue schema. The four Passions of the Soul is another schematic motif to illustrate vita humanae, as is their subjugation via a Stoic apathea or Christian transcendence, mystical triumph in this life. These and many others are allegorical versions of the circumstances of life. In every case, a selection is made from the countless possible choices.

The subject matter of the E-Series obviously is quite different than the subject matter in Tarot. In addition to being commonplace subjects needed by artists in their compositions, (the actual historical function of the project), the Muses, Arts, and Virtues are an aspirational catalog of strengths. To have inspiration, skill, and virtue is to be great in this world and prepared for the next. The prominence given to the Pope, Apollo (a "type" of Christ), Theology, and the three Pauline Virtues, emphasize this humanistic duality. This can be the middle register of a three-level work just as can the Seven Deadly Sins. One design is hopeful and encouraging while the other is a dire warning. Neither is like an Ages of Man design, an actual vita humana, or like the contemptu mundi cycle of the middle trumps, which schematically illustrates the vicissitudes of Fortune and the role of Virtue in the life of Man. Even within Tarot the design varies. In the Bolognese sequence, Virtue is shown with the Triumphs, as something desired but which cannot prevent reversal and downfall. In the Tarot de Marseille sequence, each of the three Moral Virtues is shown triumphing over one of the three turns of the Wheel.

Likewise, a third section is transcendent, something beyond this life, but it may still take various forms. This may reflect an individual judgment (as in Costa's allegory in Bentivoglio Chapel), the Last Judgment as in countless works, or something as vaguely allusive as the cosmograph of the E-Series. It could be a mystical transcendence in this life rather than post mortem, again failing to conform to a neat, distinct Three Worlds design.

In various ways, the elements may be combined rather than separated into groups. For example, in Petrarch's Trionfi there is often no clear-cut, simplistic analysis. Fame is of this world, yet triumphs over Death. Time is usually what leads to Death, yet Petrarch used it to wash away Fame, an ubi sunt trope. Time and Fame are conventional allegories, very popular circumstances of life, and yet in the Trionfi they appear after Death. Messy. As another example, in illustrations for Petrarch's cycle there is sometimes mixing and repetition. The Triumph of Death, in particular, is the third of six in the cycle but often contains an explicit Ranks of Man, (with the characteristic emperor and pope), the allegory of Death itself, (always the most significant point in Man's life), and representations of both Heaven and Hell, or souls being carried by angels and demons. One might expect the Ranks of Man to appear in the first Triumph, and often it does. Again, art is messy and often less schematic than things like the E-Series.

So yes, different categories of subject matter are being illustrated, and that the simplest, most clear-cut diagrammatic version of this may be loosely characterized as mankind, allegories of life, and things which transcend this life. But we should keep in mind that this is a gross simplification which is not always directly applicable. It is the context which makes a detailed explanation of the trump cycle possible, but it has little explanatory power in its own right.

I would argue that the most generic description of this 3-part design, at least in the Western European tradition, is "Triumph of Death". Death is the one thing that unites all Mankind and which occupies the same position in all human life -- its end. Death is the great transition between the two realms, and while it ends this life it begins the next. Death is the greatest theme of Christian allegorical art, with endless works of vanitas, memento mori, ubi sunt, and related genre. From Vado Mori and the Three Living and Three Dead through the pervasive Dance of Death and Triumph of Death per se, with many unique allegorical works like Petrarch's Triumphs or Michault's Dance of the Blind.

This is partly because death is that natural end of Everyman, but also because death is at the heart of Christian mythology. In Genesis, Adam/Man sins and his punishment is death. In the Gospels, Christ's death is the sacrifice to atone for Adam's sin, and his own resurrection, triumph over death, is the first fruits of the general resurrection. In Revelation 20, we have the final triumph over the Devil/Sin and death. Tarot is another in this vast tradition of Christian allegories of death. The trump cycle is itself an elaborated Triumph of Death, elaborated in terms of the middle trumps which show the life cycle leading up to death.
Ross G. R. Caldwell wrote:Looking for any authority who described "threefoldness" in vertical hierarchies proved difficult; it became apparent that it was such a fundamental, underlying and completely natural basis for the spatial organization of iconographic information, that people describing such art assumed it rather than explained it as a principle.
First, thanks for the pointer to Ong's 1959 article, "From Allegory to Diagram". He may not have given you the quote you were hoping for, but Ong's discussion of the "schematic relationships" among the figures, the diagrammatic composition, is directly on point for Tarot. This is the structural aspect, what I've called the sequential context, which constrains and clarifies the intended meaning of the trumps. This is what the three sections mean for Tarot: they are the macro structure of the trump cycle. The micro structure consists of the smaller groups, the eight duos and two trios that I've called affine groups. The arrangement of these three major groups and ten lesser groups tells us what the trump cycle was intended to convey, the intentio operis. Ong was making the generic point, and Tarot -- properly understood in terms of those groupings -- is just one more example illustrating his point. These structures are what enable us to dismiss most orderings and interpretations and get closer to the intended meaning.

Your attempt to find the best analysis of this schematic tendency in a general sense is interesting. Finding a good quote, or creating an pithy and insightful description yourself, seems worthwhile. As noted above, this is an over-simplification like Campbell's 3-part cycle. His Separation, Initiation, and Return can even be stretched to cover the Three World analysis I've offered for Tarot, the three realms you are discussing here, if one is so inclined. Our individual identities as suggested by a Ranks of Man can be taken as our embodiment in this life, this world, separated from the Divine by Adam's Fall. This life is naturally our initiation into the next, and the De Casibus or Wheel of Fortune cycle both initiates us and leads to death. Death returns us to God for our Last Judgment. The question is to what extent this sort of simplification is helpful. If you say, "Campbell's Monomyth" to a Tarot enthusiast, you fall immediately into a swamp of vacuous impositions and empty analogies, taking you away from an historical analysis of the specifics of a particular Tarot deck or decks.

You used the term prolegomena, preliminary remarks for the study of the groups, subjects, and sequence of the Tarot trumps. That is precisely how I used such a discussion in various presentations, most notably in The Riddle of Tarot. Although the emphasis was on analysis of the various trump cycles themselves, rather than other works, numerous examples were cited of composite works uniting different types of subject matter. Over the years many more have been presented, and I have even used this tripartite analysis to define Triumph of Death in a general sense. As such a basic conceptual aspect, it needs to be presented (i.e., it is absolutely necessary, which is why I continue to return to the subject) but it explains very little about the trump cycle details.
Ross G. R. Caldwell wrote:Triumphal Arch of Alfonso V in Castel Nuovo, Naples

1. Triumph of Alfonso (terrestrial)
2. Virtues (moral)
3. Archangel Michael flanked by saints Anthony and Sebastian (celestial)
Nice example.
Ross G. R. Caldwell wrote:Lorenzo Costa, Cappella Bentivoglio Triumph of Death (starts with allegory, introduces concept of "literal" as a subset of "terrestrial" or "real" (actually vice-versa))

1. Triumphs of Fame (and Fortune) and Death (and Chastity)
2. Exempla (Fame), souls ascending (Death)
3. Souls ascending, angels, God the Father, Jesus, Mary (in Death triumph only; the two compositions are a diptych)
Here we see some problems that often arise when attempting to force-fit a particular work into a "universal" schema. The truth of the work is that it's just messy. Contemporaneous figures are used in the allegory along with historical/legendary ones, which is to say that it does not fit the oversimplified model. I won't go into your elaborate speculation (in which you suggest that there may be a specific topical reading to virtually every detail) except to say that, in the absence of documentary support, you may be reading too much into it.

Regarding the messiness, it is worth noting that the composition's diagrammatic flow begins with scenes of the Fall of Man from Genesis. These are in the center of the upper-left quadrant. Surrounding them, Mankind's fallen state is exemplified by eight legendary incidents, exemplars arranged in a Wheel of Fortune. Historical stories turned into a Wheel of Fortune, with the Fall of Man at the center -- where Time or Fortuna would ordinarily be placed, turning the Wheel. It is unique, and conceptually messy. As a grouping, these historical images are far removed in time from the topical elements in the lower parts of the paintings. They are also different in kind from the allegorical elements in the lower part of the panels, and from the soul rising to judgment in the upper-right quadrant. So on the one hand it is both unique and seemingly messy, at least if one wishes to oversimplify.

On the other hand, the pieces are neatly illustrative in their own right and they are united in the overall composition to tell a coherent story: starting with the Fall of Man, then ancient examples of Fortune, contemporaneous examples of Fame and Death, and finally the triumph over death. It fails miserably to conform to either neat groupings of like character or to the simplistic bottom-to-top reading. Spatially, the narrative thread begins at the top, goes down, sideways, then up. As such, it serves as a great warning about the one-size-fits-all, monomyth approach to iconography.
Ross G. R. Caldwell wrote:Tarot trump sequence
1. Human types (highest and lowest; highest complete ranking of court cards)
2. Virtù and Fato
3. Heavenly order

The tarot trumps are neither monument, nor painting, nor quasi-encyclopedic model-book; they are the pieces of a game. Their context is ludic, and play-function is therefore one of the constraints upon their design.
It's good to recognize, and never forget, that Tarot was a game and the trump cycle was intended as both a moral allegory, to make the game inspirational, and as an easily learned and readily usable hierarchy of trumps. Beyond that, however, looking at the moral allegory itself, there seems to be little about the game which would constrain either the selection or arrangement of the subjects -- they just need to make some sense, which is generally true in other contexts as well. It is a cycle which could be easily adapted for a chapel or other architectural space, like other unique cycles we've seen, or to a series of prints.
Ross G. R. Caldwell wrote:This explains why the various sequences share the threefold sensibility, and why Dummett was able to reduce it to three families. It also explains why the designer chose these three types of subject matter, vertically arranged in three divisions, to illustrate the pieces of his game.
I'm not sure what you mean here. It seems as if you are saying that the commonplace division into three types of subject matter in various works of art explains the specific division into three types of subject matter in Tarot. If so, then it seems simpler to just note that complex works of art, grouping and/or mixing different types of subject matter into a unified composition, were commonplace. If it is stated that way, then perhaps the absence of a more detailed analysis in the iconography literature is not surprising.
Ross G. R. Caldwell wrote:It finally explains why it was easy for the players to understand the logic of the subjects, why they were where they were. The threefold structure is a necessary, but not sufficient, explanation for the choice of subjects and their number - different subjects could occupy the lower, middle and highest levels...
Yes, the three types of subject matter are necessary to understand the logic of the design of Tarot, the arrangement of the subjects into groups in archetypal decks. But a different design could easily have been chosen. If you want to keep roughly the same number of trumps, one can take the middle three decades of the E-Series, keep the Muses (9), the Liberal Arts (7), and the Virtues (7) -- voila! Twenty-three trumps in three clearly defined groups but lacking the Three Worlds structure. If the orderings of each group were taken from literary sources, it might seem to be very didactic and mnemonic set of trumps. Even in terms of Tarot, however, the Three Worlds aspect does not seem to help at all in understanding the choice of subjects or their number. These seem to have been based on other considerations, specifically, a detailed iconographic program.

Well, that was more fun -- thanks again. I started on a reply to the Devil posts... maybe tomorrow.

Best regards,
We are either dwarfs standing on the shoulders of giants, or we are just dwarfs.

Re: The Three Worlds Model

Happy New Year Michael!

Thanks for your response. Naturally I'll have to respond ... bit by bit. Some bits bigger than others.
mjhurst wrote:Now you have concluded that the Devil is not part of the highest section. As noted at the top, you are attempting something similar to what I have tried to do but as a less detailed interpretation rather than a card-by-card programme. Like Dummett, you don't seem to think such a specific explanation is appropriate -- the trump cycle wasn't that precisely structured.
That's right. I think the designer wanted to make a trick-taking game with a permanent set of trumps. He had some ideas, and worked out the math.

Among those "ideas" were the symbolic subjects of the trump sequence. It seems there were already some games with extra cards, either wild cards or "power cards" like Emperors, if Fernando de la Torre's can be taken as indirect evidence of their existence in Florence in the early 1430s, or if the "Emperors" of the name Imperatori actually means that some Emperors were added to the deck. You can do whatever you want when you make decks by hand; but to be popular, you have to take a risk and make the effort to carve a plate to print. But one or a few extra cards in a deck are just gimmicks; people take them or leave them, they don't stamp a game and make it popular. The trumps in Tarot aren't gimmicks or just a few extra cards that you can take or leave; they make the game.

Whatever "gimmick" cards might have been around in decks, at the very least the word "Emperors" shows the conceptual world from which such gimmicks would have emerged - the "higher" cards in a rational sequence had to start with whatever was higher than Kings. Naturally this is Emperors and Popes. So I think this was the beginning of the designer's iconographic programme - to finish or complete the worldly sequence. This is the "terrestrial" part of the 3-part analysis, and it does the rest - with the math of the game providing some constraints and necessities.

As I noted, the earliest written rules - about 1600 in Bologna (manuscript), and those printed in France in 1637 - don't contain any mnemonic or any other system for learning the trumps. Bologna just lists them in descending order, and the French rules don't even name them, just giving the counting trumps (Monde, Bagat and Math).

I said I can't believe that anyone, even them with their presumably better memories, could learn such a list precisely in the few minutes given them before the first game. The written rules presume the leisure to memorize it at leisure. For the Bolognese writer, the audience could do it any way they wanted (for the French cards, they must have already had numbers and probably titles printed on by 1637). But I read the rules for clues.

The first is the descending order. Most early lists give the cards in this way - the Strambotto, Alciato, Lollio, Garzoni. To me this indicates that the power of the cards - the ludic power - is what counted. Whatever system underlay the quick and dirty at-the-table method of learning therefore probably gave the cards in their ludic values first.

This seems confirmed in the French rules' mention of only the counting trumps, and in the Bolognese rules' discussion of the sequence after the mere listing of the trumps (including the special rule for the papi in the discussion of general ranking), which mentions first only the group which counts, namely the Angelo, Mondo, Bagattino and Matto (see , chapter VI). If either of these is any indication, this second "clue" from the written rules shows that players were first taught what counted in the trumps, the counting trumps, the highest and lowest cards. They are my "ludic group", which I think must be the designer's first concern in his overall conception of the subjects of the sequence. This is why the symbolic subjects that he chose to illustrate them with don't have to have any necessary symbolic relationship, group or logical sequential relationship, with the iconography of the rest of the group they are necessarily appended to. They depict the same general type of subject matter, because that is what their place in the threefold structure demands - earthly things, and Bagatto and Matto show low human types; and heavenly things, of which the World and the Angel (Resurrection, Last Judgment, End of Time) are the biggest conceivable things. But the Matto and Bagattino don't have to be interpreted as if related to the papi in a unified symbolic or moral whole, and the World and Angel don't have to be interpreted as "lights" somehow. The two funnymen are suitably low human types, and the World and Angel are two suitably grander subjects than the Sun.

So, putting to the side the mathematical question of how many the game was designed for (and which determined the number of trumps) for the moment, the symbolic constraint was first the counting cards, which were to be the highest and lowest in the sequence. Popes and Emperors follow, as the highest human types. What is higher than Popes and Emperors? Moral things, which are personifications or allegories. He chooses conventional allegories that the audience would recognize from what they saw during the Saint John the Baptist processions and the Magi one, namely a triumphal chariot with love and virtues (and which seems to have suggested the name of the game to him), probably to be thought of as the "good things in life" (that's how people tend to memorize it, at least, when they are explained to be a group and told to distinguish them from before and after, especially in relation to what comes after), followed by necessity, the fateful or "bad things". Then, of course, looking up, you know the rest.

The symbolic groups are determined by the general threefold structure of complex works, the "Three Worlds", in their proper order, while the total number was worked out in the math of a game for four players where all the cards are dealt out. The moral subjects are conventional, and their specific order would be easy to memorize once their groupings were seen. So, for practical purposes, there are four groups: people, good things, bad things, and the heavens.

So to me it seems that the designer had a symbolic sequence in mind that would use symbolic subjects like the rest of the deck, particularly the court cards, but higher subjects. He worked out the math, and then chose subjects grouped together by theme, few enough to be memorized quickly (four).