previous next


DUO´DECIM SCRIPTA (κύβοι, διαγραμμισμός [Pollux, 9.99], in late Greek τάβλα), a game of mixed chance and skill, which must have been substantially the same as our backgammon. The following points of identity may be regarded as established. The game was played on a board of 12 double lines, with 15 white and 15 black men; the throws were counted as we count them; “blots” (ἄζυγες) might be captured; the pieces (whether they started from home or not) had to be brought home; and the winner was he who first cleared off his men. On the other hand, there were three dice instead of two [TESSERA]; and it is impossible to say where the men started, or how blots taken up re-entered; In the initial position the pieces may have stood in three rows of five or five rows of three; and either in the player's own table with a view to the double journey, or in the opponent's table with a view to the journey home. With the three dice the pieces would soon be scattered, and thus a less artificial arrangement than our own may be thought probable. The phrase ὀπισθιδίη ὁδὸς in Agathias may seem to favour the notion that they were played out and home. The board was ἄβαξ ABACUS more generally tabula, or from its raised rim alveus, alveolus; the men ψῆφοι, calculi; the situation at any point of the game, θέσις (Agathias); to move, τιθέναι dare; to retract a move, ἀνατιθέναι, reducere. In a fragment of Cicero (ap. Non. p. 170, s. v. Scripta) we find: “Itaque tibi concedo, quod in duodecim scriptis solemus, ut calculum reducas, si te alicuius dati poenitet.” This privilege is more likely to have been of the nature of odds granted by a superior player, than a regular rule of the game.

The classical Greek writers mostly use κύβοι, κυβεύειν, of games into which skill entered as well as of mere dicing; though Philemon marks the distinction (μεθύειν, διαγραμμίζειν, κυβεύειν, fr. 216 M.). That κυβεία was a game of skill as well as chance is clear from Plato (Rep. 10.604 C, Phaedr. 274 D, and ap. Plut. de Tranq. An. p. 467 A), and from a story told by Plutarch (Plut. Art. 17); cf. Ter. Adelph. 4.7, 21. Ovid alludes to the Duodecim Scripta (A. Am. 3.363-4) among games which lovers are to play together; others are latrunculi (357-8), “spillikins” (361-2), and “go-bang” (365-6, this last couplet has often been misunderstood as though it referred to the placing of the backgammon-men). The lover is, of course, to lose on purpose to his mistress ( “Tu male iactato, tu male iacta dato,” ib. 2.204), either by unlucky throws (perhaps he is to cheat in her favour, cf. 206) or by making bad moves. Martial includes among his modest wants tabulamque calculosque (2.48). The celebrated jurisconsult P. Mucius Scaevola was famed for his skill at Duodecim Scripta (Cic. de Or. 1.5. 0, § 217; V. Max. 8.8.2; Quintil. Inst. 11.2.38). Quintilian further tells the story that Scaevola, after losing a game, recalled the whole of the throws and the way each had been played; pointing out the move where he had made a mistake, and verifying his own recollections by those of his opponent. This is cited as an example of memory and logical sequence (ordo).

None of the above passages shed much light on the details of the game. Our knowledge of them is mostly gained from an epigram of Agathias (Anth. Pal. 9.482; also in Brunck, Anal. 3.60) on a case of special ill-luck which befel the Emperor Zeno (A.D. 474-491). This epigram has been discussed by many scholars (e. g. Salmasius in a long and learned note to the Historia Augusta, ii. pp. 736-761, ed. var. 1671; T. Hyde, the historian of chess; De Pauw, in a special dissertation; Jacobs, in his notes to the Anthology), but until lately was never rendered intelligible. The problem has been solved independently by M. Becq de Fouquières in his Jeux des Anciens, and Dr. H. Jackson in the Journal of Philology: on the few points where they differed, Dr. Jackson has since given in his adherence to M. Becq de Fouquières' conclusions. We reproduce this epigram with a few comments from the sources just named: Εἰς τάβλαν Ζήνωνος τοῦ βασιλέως.
οὐτιδανο, μερόπων, εἰ καὶ μέγα ῥέξομεν ἔργον,
οὔτινος εἰς μνήμην δηρὸν ἐπερχόμεθα:
οἱ δ᾽ἀγαθοί, κἢν μηδέν, ἀναπνεύσωσι δὲ μοῦνον,
ὡς Λίβυς εἶπεν ἀνήρ, τοῦτ̓ ἀδάμαντι μένει.
δήποτε γὰρ Ζήνωνα πολισσοῦχον βασιλῆα,
παίγνιον ἀφράστων ἐκτελέοντα κύβων,
τοίη ποικιλότευκτος ἕλεν θέσις, εὖτ̓ ἀπὸ λευκοῦ,
τοῦ καὶ ὀπισθιδίην εἰς ὁδὸν ἐρχομένου,
ἑπτὰ μὲν ἕκτος ἕχεν, μίαν εἴνατος: αὐτὰρ σοῦμμος
δισσὰς ἀμφιέπων ἶσος ἔην δεκάτῳ:
ὄς τε πέλει μετὰ σοῦμμον ἔχεν δύο, μουνάδα δ᾽ἄλλην
ψῆφον τὴν πυμάτην ἀμφιέπεσκε δίβος:
ἀλλὰ μέλας δισσὰς μὲν ἐν ὀγδοάτῳ λίπε χώρῳ
καὶ τόσσας ἑτέρας ἐς θέσιν ἑνδεκάτην:
ἀμφὶ δυωδέκατον δὲ διέπρεπον εἴκελοι ἄλλαι,
καὶ τρισκαιδεκάτῳ ψῆφος ἔκειτο μία:
δίζυγες Ἀντίγονον διεκόσμεον: ἀλλὰ καὶ αὐτῳ̈
ἶσος ἔμιμνε τύπος πεντεπικαιδεκάτῳ,
ὀκτωκαιδεκάτῳ πανομοίιος: εἰσέτι δ᾽ἄλλας
εἶχεν διχθαδίας τέτρατος ἐκ πυμάτου.
αὐτὰρ ἄναξ λευκοῖο λαχὼν σημήϊα πεσσοῦ
καὶ τὴν ἐσσομένην οὐ νοέων παγίδα,
τριχθαδίας ἀδόκητα βαλὼν ψηφῖδας ἀπ̓ ἠθμοῦ
πύργου δουρατέου κλίμακι κευθομένη
[p. 1.696] δοιὰ καὶ ἓξ καὶ πέντε κατήγαγεν: αὐτίκα δ᾽ὀκτὼ
ἄζυγας εἶχεν ὅλας πρόσθε μεριζομένας.
τάβλην φεύγετε τάντες, ἐπεὶ καὶ κοίρανος ἀτὸς
κείνης τὰς ἀλόγους οὐκ ὑπάλυξε τύχας.


The diagram (from Becq de Fouquières, p. 375) shows the “complicated position” (1. 7) before the emperor makes his throw. Zeno plays the white men, and moves from 1 towards 24. The situation of the numbered points presents no difficulty: thus, White has 7 pieces at 6, 1 at 9, 2 at 10; Black 1 at 13, and 2 each at 8, 11, 12, 15, 18 and 21 (τέτρατος ἐκ πυμάτου). The named points σοῦμμος (summus), δίβος (apparently divus, in spite of the difference of quantity; so Salmas. p. 753 b), and Ἀντίγονος (doubtless a Greek king chosen to balance the Roman emperor) have been ingeniously worked out in accordance with the conditions of the problem. It was natural to assign the σοῦμμος to one of the corner places 1, 12, 13, or 24; but it is now proved to have been at 19, the limit of the highest throw (three sixes) starting from 1. Hence two white men are placed on 19 and two more on 20, “the point next to the σοῦμμος” (1. 11). Similar calculations prove that the Antigonus, where the two remaining black men are situated, was at 14, and the δίβος, occupied by a solitary white man, at 23. In this position the emperor threw 2, 6, and 5; whereupon it appeared that he could not avoid leaving 8 ἄζυγες or “blots,” an extraordinary piece of bad luck. It must be assumed that, if there was any way of playing all the numbers thrown, the player (contrary to the modern rule of backgammon) was bound to adopt it; and Zeno had consequently, as shown on the diagram, to play from 10 to 16, from 19 to 24, and from 20 to 22. Becq de Fouquières imagines that at this point his game was at once lost in some unexplained way: it is only necessary to assume that it was practically ruined by the 8 blots, 5 of which lie just in Black's course and can hardly fail to be hit by him, whatever he throws next. The special hardship of the emperor's case is that, out of 56 possible combinations of the dice, he threw the only one which raised the number of his blots from 2 to 8: this suffices to point the epigrammatist's warning, τάβλην φεύγετε τάντες: the master of the world is as much exposed to the caprices of fortune as any other gamester.

The Latin forms of the technical terms τάβλα, σοῦμμος, &c. in the above epigram, and the absence of any direct mention of the “twelve lines” in Greek authors, point to the conclusion that this precise form of the game was Roman rather than Greek. The twelve lines were doubtless an extension of the original five, so often mentioned in connexion with games of this class [LATRUNCULI; TABULA LUSORIA]; and may have been suggested by the die with its six faces and points from one to six (Becq de Fouquières, pp. 362, 367). It will be seen that the word τάβλα applies to the game itself, not merely to the board on which it is played : this may be paralleled from Old English, where “tables” is a synonym for draughts or, more probably, backgammon; and from Old German, in which Zabel is used for a game in glossaries of the 11th and 12th centuries, and chess is called Schachzabel in some early printed books (Van der Linde, Gesch. des Schachspiels, vol. i. App. pp. 126, 134 ff:). For τάβλα, ταβλίζειν we find in late Byzantine Greek ταῦλα, ταυλίζειν: long before those times αβ and αυ were undistinguishable, both being pronounced as au.

More than a hundred ancient boards, serving for six different games, had been found in Rome alone down to 1877 (Marquardt, Privatl. 838); but only a single example shows the twelve lines. This is of marble, bears a Christian inscription, and is of very rude workmanship and illiterate spelling. It has been engraved by Salmasius (l.c. p. 751), Gruter (Mon. Chr. p. 1091), Becq de Fouquières (p. 364), and in a simplified form, omitting the inscription, by Rich (s. v. Abacus). This is to all intents and purposes a backgammon board, exhibiting the four half tables of six lines each. Mention is made of boards and men of costly materials, or of peculiar construction. In Petronius (100.33) Trimalchio plays on a board of terebinth-wood, with dice of crystal, and with gold and silver denarii for black and white men. Pliny (Plin. Nat. 37.13) has an absurdly rhetorical account of the splendours of an alveus lusorius, in gold and jewels, borne in Pompey's third triumph B.C. 61; in the centre of it was a golden moon of thirty pounds' weight. Becq de Fouquières (pp. 382-3) here amuses himself with the “solar myth,” Suidas (s. v. τάβλα) having recorded a theory that the twelve lines represent the signs of the zodiac, &c. He has already given a common-sense explanation (see above), and, no doubt, is not to be taken seriously here. The Emperor Claudius had his carriage fitted with a board which would not upset, in order to play when travelling (ita essedo alveoque adaptatis, ne lusus confunderetur, Suet. Cl. 33). The tabula lusoria described by Martial was also specially adapted for two different games, probably on opposite sides: “ Hic mihi bis seno numeratur tessera puncto:
Calculus hic gemino discolor hoste perit
” (14.17).

The first line refers to the Duodecim Scripta; the second, modelled on a couplet of Ovid (Ov. Tr. 2.477-8), to the game of draughts (latrunculi), in which the player left with but one man is bound to lose to his opponent who has two. (Becq de Fouquières, Jeux des Anciens, ed. 2, 1873, pp. 357-383; H. Jackson, in Journ. of Philo. 7.236-243; Marquardt, Privatl. 834-838.)


hide Display Preferences
Greek Display:
Arabic Display:
View by Default:
Browse Bar: