Date redacted: August 1996, revised August 22, 1996
Redactor: Justin du Coeur
Sources: Francis Willughby's Volume of Plaies, c1665. First attestation, Skelton's Why not to Court, 1522 (Singman, OED). John Florio's 1598 Italian-English dictionary says that it was known in Italy as Calabrache.
Willughby writes that "There is no other game at cards that is any thing a kin to this"; today we would generally classify it as a "fishing" game, although it appears to be the earliest such game in the West. (Although Florio does list a game called "Andar a pisciare".) This is a pleasant little game, lightweight and easy to learn. It is not described in any other gaming sources that I know, but is mentioned by name in several period sources; I presume that this idiosyncratic name is only being used for one specific game.
The general purpose of the game is to make pairs and mournivals (fours of a kind), by matching with cards that are face up on the table. (Note that a "pair royal" means three of a kind.)
Note that the term "mate", which I use occasionally herein, is my own term to signify matching a card with another of its rank; Willughby has no convenient term for this.
French-suited deck of 52 cards.
Before beginning, agree upon a stake. It appears that you will win or lose a couple of stakes in a typical game. Everyone stakes 2, except the dealer, who stakes 3; the dealer collects the stakes in a pot. This is the money that the winners will collect from at the end, although (if you do particularly poorly) you may have to stake more at the end.
Willughby is not explicit about how to pick the dealer; I would assume that you lift for dealing, as with most games. I would generally recommend that high card deals. I believe that the dealer has a slight advantage, but that needs testing.
Deal eight cards to each player, one card at a time. Spread the remaining 12 cards on the table, face up. If there are any mournivals in these 12 cards (unlikely but possible), the dealer takes them all immediately, and places them by him.
Starting with Eldest, take turns pairing up with the cards on the table. Find a card in your hand that matches one or three on the table. Take the one(s) on the table, mate them with the one in your hand, and place them face-up by you. Note that this mating is by pairs -- you must put down either two or four at a time, not three. (Willughby doesn't state this clearly, but I read it between the lines.) This is how you score, by taking cards up from the table and mating them with ones in your hand. If there is a pair royal on the table, and you have the fourth, be sure to take all three that are on the table.
There are a few special rules for laying down cards, which collectively boil down to, "If you have a pair in your hand, and you can prove that it is impossible to ever mate those cards (since their mates will never be on the table), you can lay them down." Specifically:
Be sure to do these things as soon as possible; if you don't do them, and someone notices, they may be able to claim the pair instead.
When a player can not make any mates with the table, they must "lie down"; that is, they must place all of the cards in their hand in the middle of the table for others to mate with. Willughby attributes the name of this game to the way people laugh at someone who must lie down. At this point, obviously, you can no longer make any more mates in this round, but you are not out; the cards you have laid by you will count when scoring.
When only one player is left (that is, all the other have laid down), the hand is over. The one who is left is not allowed to make any more mates. However, the one who is left immediately wins five stakes out of the pot.
(It is easy to misread Willughby on this last point (as indeed I did misread him at first), so it deserves a little clarification. He says that the winner of the last "trick" wins 6. But if you examine his math, at the end of his description, you find that he means winning 6 *instead* of winning 1 -- in general, a "trick" (that is, a pair) is worth one stake. So you actually win 5 *extra* stakes for taking the last trick, which makes the math work out.)
All remaining cards, both in players' hands and in the middle of the table, go to the dealer. (This is why the dealer appears to have an advantage.)
Scoring is based on how many cards you gained or lost in play. If you have eight cards by you, you have broken even, and neither gain nor lose. Otherwise, gain or lose one stake for each two cards. That is, if you have ten cards (two extra), you gain one stake; if you have six cards (two few), you lose one stake, and so on. Since you can only place pairs by you, there should never be odd cards to worry about.
All stakes are paid to and from the pot that the dealer maintains; place any losses in the pot, and take winnings from it. Willughby points out that everything should come out even. 11 stakes were wagered upfront (3 from the dealer, 2 from each other player). Eight cards from each player (that is, 40 cards total) are essentially counted as "breaking even", so the remaining 12 cards, or six stakes, will "win". Add to that the five stakes taken by the person who won the last hand, and you get the 11 that had been wagered. If anyone won excessively, it means that someone else lost (that is, had fewer than 8 cards at the end), so they will cancel out.
Willughby recommends that this be played as a fast-paced game, and notes that mistakes are common. In general, if you catch someone else's mistake, you can benefit from it.
If the dealer overlooks a mournival on the table, the person who notices it first can take it. (Give the dealer a chance to notice it, though.)
If a pair royal is on the table, and a player takes only one of it, the player who notices can take the other pair.
If you have a pair royal in your hand, and only lay down a pair of it by you, someone who notices that when your cards are laid down can take that remaining pair.
Willughby notes a number of strategy points. The general thread is that if you have a certain mate, leave it for last -- you want to get as many chances to mate as possible.
Note also one important point that Willughby only implies -- winning the last trick is worth a lot. Besides implying that you will have won a number of tricks, it also gains five stakes extra. This is an even bigger reason to make sure that you can stay in as long as possible: to win that last-trick pot.
Willughby describes a possible variant for four players. Deal ten cards to each player; count 4 stakes for taking the last "trick" (that is, being the last player able to play), which actually means that the winner of the last trick will win 3 stakes extra. To make the math work out, everyone scores based on an break-even of *ten* cards at the end; that is, win one stake if you have 12, lose one stake if you have 8. This will result in a pot that contains 9 stakes; six go to those who take the extra 12 cards, and three to the winner of the last trick.
He says that, with some tweaking, one should be be able to play with a variety of numbers of players. We can follow this reasoning with a few possibilities, particularly if we are willing to vary the number of cards in the middle:
Between the two available variables (the number of cards per player and the amount wagered), you can adapt the game to many circumstances. All that said, it isn't clear to me that the game was actually played with these variations in period. Willughby was fond of this sort of mathematical extrapolation, and these variations may just be his own fancy...