Re: Valets-Slaves Reference

Ross G. R. Caldwell wrote: 14 Nov 2019, 16:50 A slave is certainly in the conceptual range of a page, valet, servant, etc., so it is easy to see how it emerged in some card games. Another manifestation of "ludic logic," as I like to call it, "emergent" features of games, almost folkloric, like making an intrisically weak card like a Jack the strongest or most valuable.
The original passage that sparked off this topic deals precisely with this "conceptual range" and what it might mean where the cards themselves are concerned. Thanks again.

Re: Valets-Slaves Reference

We can rest assured that the Larousse's source was Christian. Belline's deck was unique to him, and the sole copy, assuming with the museum where it is kept that it is authentic, was found in an attic in the 1960s, according to the blurb that accompanies the modern reproduction. While "jack-slave" was a term for a rascal in Shakespeare, as was "slave" itself, only "knave", meaning the same, ever got applied to playing cards. In Spanish the word for the jack is "sota", which also means "hussy, brazen woman, whore", which is more of the same. But I can find no equivalent in French, applied to both the card and a person.

It is my impression that terms used in a certain way in a certain work and almost nowhere else do not belong in a dictionary. I hope that later editions of Larousse corrected their mistake.

Christian was writing an Egyptianizing fantasy in which the tarot played a key role, a fantasy complete with tunnels under the Great Pyramid and even a 22 letter Egyptian alphabet, obviously constructed by varying the Hebrew letters in various ways. It is the same with "sicle", as the "original" of "schekel", and the other terms for suits, for which Christian didn't bother to look up the Hebrew equivalents but just used French terms other than the usual ones. Likewise everyone knew from the story of Joseph that the Egyptians would have used slaves - not rascals, but slaves with suitable servility - in the capacity designated by the French term "valet". Homme Rouge was published as a work of fiction, one of those stories within a story that were popular then. It didn't sell very well, so Christian repackaged it as part of an "Histoire", a term that even in English can apply to a work of fiction, as in The History of the Adventures of Joseph Andrews, by Henry Fielding. The term "history" in that sense means "a narrative relating the events of a character's life", meaning 9 in the Collins Dictionary. In Christian's case, the character is not a person but a practice, that of magic, which is going a bit far with the fictional sense. It may be that the French word has a broader application to fiction than as the English. All the same, Christian presents his story as fact. But so does the title page of Robinson Crusoe, and other novels. It is not a good practice; people are too gullible.

Christian's activity is to be distinguished from what I sometimes like to do, imagine ways the tarot images might have been imagined by educated players in a given place and time. I call that "hypothesis," since it is falsifiable: if, for example, a certain source I present was not available at the time in question, or I have taken a detail in a card in a way no one then is likely to have done. I also argue that the fostering of such imaginings might reasonably have been part of the designer's intent, given the interest in Egypt, Greece, and Rome at the time. (Egypt would have entered in after the promulgation of Horapollo's Hieroglyphica.) That is different from imagining that the cards actually had a certain history, disregarding numerous historical facts known at the time, as Christian did. It is not the same as "historical fiction" either, as I make no effort to turn my speculations into a narrative. I like to think of it as "Informed speculation".

Re: Valets-Slaves Reference

For the sake of clarity, I want to point out that Lévi’s text - a piece of private correspondence - was written in 1860 and circulated by his disciples from the end of the following year according to the colophon. This text on talismanic magic not only includes the aforementioned term ‘Esclaves’ but also uses ’Sceptres’ rather than Batons. The other two suits are named Epees and Coupes as usual.

Pitois’ novel was published in 1863. The reworked “history” was published in 1870. Perhaps the pseudo-academic tone of the latter convinced the author of the dictionary entry of its worth as a serious reference, who knows. Given what we know of the relationship between Pitois and Lévi, it is no great stretch to say that the latter further embellished - as was his wont - what his former master had previously concocted. (I expect that Dummett’s 2 volumes on the occult Tarot will deal with this matter.)

Furthermore, it is highly likely that the reference to ‘Esclaves’ was based on an as-yet unidentified historical source (perhaps even one I have cited earlier but haven’t looked too closely at). I say this because we know that not only was there an entire suit of Slaves in Ganjifa, but at least one iteration of this game had an explicitly named Slave as the Jack/Valet card.

This concords with the contemporary text I am looking at as well (which gives no reference unfortunately). In fact, I have just noticed that the Wikipedia entry to the reference I posted earlier is deceptive: it refers to a later edition of a work originally published in 1921 (and available on the Internet Archive here: ... lam-In.pdf) The name of the author has been modified, which is why I did not find it before. Pages 335-336 contain what concerns us.

The source of the Larousse entry is undoubtedly Pitois, but my initial query was where the term originated from. We have seen that the French already had knowledge of the game of Ganjifa, spelled in various ways, and knew that there was one suit of cards called Slaves. Now it is only a matter of finding out where they [i.e. Levi & Pitois] got the Valet-slave designation from.

As for the “semantic overlap“, we can only say that this is another point of coincidence, one without any direct historical or linguistic connection.

To return to Pitois, rather than speculation, confabulation is the appropriate term, the mixing of deliberate falsehoods with unfounded speculation and fictional devices, and a dash of truth and probability. Entertaining, after a fashion, perhaps, given the mixture of truth, lies and deliberate red herrings. But ultimately rather annoying for any number of reasons.

Re: Valets-Slaves Reference

Yes, the account you linked to of Ganjifa is quite parallel to Levi, with a suit of slaves in one game and a "major" card called the slave in another game, the lowest after the King and Queen. That makes a good case that Levi was drawing from an article of the time about Persian card games. The only problem is that of finding a source before 1860. Travelers' reports said that the deck was 96 cards in 8 suits. That makes 12 cards per suit. Islam in India has number cards Ace to 10. So we don't know when the "slave" became part of the deck (as opposed to the 10, which might have been added instead).

I was focused on your other idea about a relationship to "jack-slaves" in Shakespearean English. That idea seems dubious to me, because "esclave" did not, as far as I can find, have an equivalent meaning of "rascal" in French, nor was it applied to the jack in playing cards in France before Levi.

In support of your idea (of Levi's having read about Ganjifa) is that Christian's astrological tarot identifies the kings of the tarot deck (he calls them masters) with four "royal stars" in Leo, Taurus, Aquarius and Scorpio (see Histoire de la Magie, pp. 495-496, at ... /f515.item and following). Francois Arago had written about precisely just such "royal stars", attributed to a Persian source, in his Astronomie Populaire, published 1861, which repeats information from another book of 1822, first published 1794 ( ... 9.000.html and preceding). You can read more about the Royal Stars in Wikipedia's article by that name.

What (Islam in India says about trumps in the first game he describes (p. 335) is very interesting: "The major cards in each suit are trumps." That is similar to how I interpret Marziano's game of "deified heroes", c. 1420, and "VIII Imperadori" (of which only the name is known but which I interpret in light of other games by means of "informed speculation"). It also goes 10, 9...Ace in the major suits and Ace, 2...10 in the minor suits, which is similar to Marziano's game. I also see that it was not prohibited by Islam, unlike a purely gambling game.

I wonder how far back these rules go. According to Wikipedia's article on Ganjifa:
The first known reference can be found in a 15th-century Arabic text, written by the Egyptian historian Ibn Taghribirdi (died 1470). In his history of Egypt he mentions how the Sultan Al-Malik Al-Mu'ayyad played kanjafah for money when he was an emir
I cannot imagine that the idea of the major cards (i.e. courts) being trumps would have been a modern result of European influence, because I don't know of any modern games in which this is true. Islam in India also says that "spades were always trumps" in the second game (p. 336). That could be a result of European influence, or vice versa, although I don't know of game with that particular rule, in which a regular suit just like the rest is a permanent trump suit. Given that the Near East and Central Asia had cards before the Europeans did, perhaps they had games with permanent trump suits before them, too.
This game, Islam in India says, was a three-handed game, each person for himself. That is how I think trionfi was, too.

Thanks very much for the link to Islam in India.

Added a few minutes later: I changed the first paragraph slightly after rereading the Wikipedia article more closely.

Re: Valets-Slaves Reference

The question of finding a source before 1860 bothers me a little more. Re-reading Wikipeida, I see in 1895 a quotation that "From travellers to Persia in the seventeenth century we know that a set of ganjifeh consisted of ninety or ninety-six cards in eight suits or colors". I assume that 90 cards would be 9 or 10 per suit, or just a mistaken impression. 96 would be 12 cards per suit, not the 13 of the more modern game using the international deck with three picture cards and ten number cards per suit, 52 cards total, which is the one with the "slave", merely the Indian name for the card the English called "jack". I assume that Levi had read about the modern terms, which he took to be ancient, and which Christian foisted on the Egyptians. This game is no earlier than the late 18th century, I would guess. But who knows? Perhaps Levi knew of something else. According to Dummett, it was the Mamluks who added the third court card, so in India there would still be two, as in Persia. There is also the question of how old the other game is, with the suit of slaves. It would be good to know Levi's source.

Relating to my concern about trumps, I notice that Wikipedia uses a different word, saying that "the lead suit is always spades." I am not sure what a "lead suit" is. Other than trumps, all I can think of is that the first trick must be led with spades. Yet Shurreef clearly says "trumps": "No trump is turned up, because spades are always trumps, and the holder of the ace of spades leads." Well, it doesn't matter, since it is a modern game, probably taken from the Europeans. What is more interesting is the other game, with 96 cards, 12 per suit, 2 of them courts, 4 major suits and 4 minor ones. Then he says "The major cards of a suit are trumps. What does that mean? What are the "major cards of a suit". I assume the king and vizier, but it is not said. Does he mean that they always lead the others, even though the ranking of the numbers vary? But surely he, or at least his translator, knows what trumps are., as evidenced by the next page. This is a revised edition. It might be useful to see what the unrevised text said.

Re: Valets-Slaves Reference

Mike, you're conflating the 2 Persian/Indian games with the "English" deck, mentioned beforehand: the deck in which the valet-slave appears is the "Akbar version" on the top of page 336 in Sharif's book, which contains 13 cards per suit; 1-10 plus Slave/knave, Queen, King.

Re: Valets-Slaves Reference

I wondered about that. After describing the 96 card game he has a footnote to "BG". Then he says "The game played by Akbar was more elaborate" with a footnote to Ain. Then he says, without a new paragraph, "In north India the suits are:" followed by the four suits, in English and Hindi, with 10 number cards, a jack (which they call by the name for "slave"), a queen, and a king. This sounds like the English deck, and it certainly isn't more elaborate than the other. And at the end of the account is another footnote, this time to an English-language source. So I take the Akbar game to be a more elaborate version of the 96 card game, and the one with the "slave" as jack as one played in north India with an English deck.

Re: Valets-Slaves Reference

mikeh wrote: 18 Nov 2019, 11:50 I wondered about that. After describing the 96 card game he has a footnote to "BG". Then he says "The game played by Akbar was more elaborate" with a footnote to Ain. Then he says, without a new paragraph, "In north India the suits are:" followed by the four suits, in English and Hindi, with 10 number cards, a jack (which they call by the name for "slave"), a queen, and a king. This sounds like the English deck, and it certainly isn't more elaborate than the other. And at the end of the account is another footnote, this time to an English-language source. So I take the Akbar game to be a more elaborate version of the 96 card game, and the one with the "slave" as jack as one played in north India with an English deck.
Yes, I think you must be right: he names the English game first, then the Mughali, but then proceeds to deal with them in inverse order. The elaborate comment refers to the preceding game, played by Akbar, i.e. the Mughal emperor. My mistake.