Hostname: page-component-5d59c44645-n6p7q Total loading time: 0 Render date: 2024-02-24T01:14:49.440Z Has data issue: false hasContentIssue false

Art. XIII.—The Origin and Early History of Chess

Published online by Cambridge University Press:  15 March 2011


No game occupies so important a position in the history of the world as that of chess. It is not only at the present day, but has been for many centuries, the most cosmopolitan of pastimes; and though one of the oldest known to civilization, it is yet undoubtedly the most intellectual. Long familiar to all the countries of the East, it has also been played for hundreds of years throughout Europe, whence it has spread to the New World, and wherever else European culture has found a footing. A map indicating the diffusion of chess over the habitable globe would therefore show hardly any blanks. Probably no other pastime of any kind can claim so many periodicals devoted exclusively to its discussion; certainly no other has given rise to so extensive a literature. The influence of chess may be traced in the poetry of the Middle Ages, in the idioms of most modern European languages, in the science of arithmetic, and even in the art of heraldry. An investigation as to its origin, development, and early diffusion therefore forms a not unimportant chapter in the history of civilization.

Original Communications
Copyright © The Royal Asiatic Society 1898

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)


Page 117 note 1 Heydebrand's Bibliography of Chess (Wiesbaden, 1896)Google Scholar contains no fewer than 3,358 entries, including about a hundred journals dealing with this game alone.

Page 118 note 1 It already occurs in the Rigveda(X, 92,11) in the sense of ‘four-limbed,’ with reference to the human figure; also in the S'atapatha Brahmaṛna, XII, iii, 2, 2.

Page 118 note 2 The four-membered army is also expressly called hasty-aśva-ratha-pādātam, ‘the aggregate of elephants, horses, chariots, and foot-soldiers,’ in the Ramāyana, Mahābharata, and Amarakośo: cf. Weber, Monatsberichte d. Berliner Akademie, 1872, p. 68, note.

Page 118 note 3 See McCrindle, , “The Invasion of India by Alexander the Great” (London, 1893), p. 102Google Scholar.

Page 118 note 4 According to the results of Professor Jacobi's researches, Rāmāyana” (Bonn, 1893), p. 105Google Scholar, the Rāmāyana in its oldest form goes back to the fifth century b.c.; while Dr. Bühler has shown (“Indian Studies,” ii, p. 26) that the Mahābhārata existed more or less in its present form certainly as early as 500 a.d., and probably much earlier. The word caturanga in the sense of ‘army’ occurs also in the Atharva Veda Pariśiśtas; but Professor Weber (“History of Indian Literature,” English Transl., p. 323) points out that this class of writings must he laterř than 250 a.d.

Page 118 note 5 Published in the Bibliotheea Indica, 1884.

Page 119 note 1 Chapter xix, which contains 62 slokas: cf. Linde, , “Geschichte und Litteratur des Schachspiels ” (Berlin, 1874), vol. i, p. 76Google Scholar.

Page 119 note 2 See below, pp. 126–9, 131, note 1.

Page 119 note 3 See Kielhom, , “Göttinger Nachrichten,” 1885, p. 185 ff.Google Scholar; and Bühler, “Die indischen Inschriften und das Alter der indischen Kunstpoesie,” p. 72.

Page 119 note 4 Kielhorn's edition of the Mahābhāṛya, vol. iii, pp. 362–3; Weber, , “Indische Studien,” vol. xiii, p. 473Google Scholar.

Page 119 note 5 Cf. Weber, , Monatsberichte, 1873, p. 710Google Scholar, note 1. According to ProfessorJacobi, , books i and vii are later additions to the older portion, books ii to vi: see his “Rāmāyana,” p. 65Google Scholar.

Page 119 note 6 suvarṛasūtrdṛṛtapadanibaddhā: cf. Burnouf, , “Lotus de la bonne loi,” p. 363Google Scholar.

Page 119 note 7 Weber, loc. cit.

Page 120 note 1 See p. 121: cf. Burnouf, op. oit., p. 466; Weber, op. cit., p. 710; “Indische Studien,” iii, 148, 154.

Page 120 note 2 Dr. H. Luders, of the Indian Institute, has pointed out to me that on the Bharhut stūpa (Cunningham, The Stupa of Bharhut,” London, 1879, plate xlv: cf. Introduction, p. 94Google Scholar; Indian Antiquary, vol. x, p. 119; vol. xxi, p. 229) a board of thirty-six squares, along with what appear to be seven dice or coins, is depicted.

Page 120 note 3 Roth in Gurupājākaumudī, pp. 1–4.

Page 120 note 4 Zimmer, , “Altindisches Leben,” p. 283Google Scholar.

Page 120 note 5 Akṛrājan, kṛrtā, tretā, dvāpara, āskanda; in the TS., III, iii, 1, 2, the five dice are called kṛrta, tretā, dvāpara, āskandū, abhibhū: cf. Zimmner, op. cit., p. 284.

Page 121 note 1 A diceboard, called adhidevana, is already mentioned in the Atharva Veda (V, 31, 6; VI, 70, 1).

Page 121 note 2 Brahmajālasutta and Sāmaññaphalasutta.

Page 121 note 3 Ed. Rhys Davids, vol. i, pp. 6 and 65: cf. Cullavagga, ed. Oldenberg, p. 10.

Page 121 note 4 Cf. p. 120.

Page 121 note 5 A board with 10 X 10 squares.

Page 121 note 6 Ed. Kielhorn, vol. ii, p. 373 (with reference to the formation of ayānayīna, in Pāṛini, V, ii, 9): cf. Weber, , “Indische Studien,” xiii, p. 472Google Scholar.

Page 121 note 7 Cf. Müller, Max, “India, what can it teach us?” p. 310Google Scholar.

Page 122 note 1 Fate as a male.

Page 122 note 2 Fate as a female.

Page 122 note 3 Vairāgya-śataka, 43.

Page 122 note 4 There seems to be no reason to suppose that this had any connection with the draught game of the Romans, much less with the game said by Plutarch to hare been played by Artaxerxea Longiraanus (cf. Hyde, , “Historia Nerdiludii,” pp. 62–3)Google Scholar, or the still more ancient game of the Egyptians (cf. Falkener, , “Games Ancient and Oriental,” 1892, p. 30ff.)Google Scholar.

Page 122 note 5 Who call this pastime the “twice-six game ”: Himly, , Journal of the German Oriental Society, vol. xxxiii, p. 679Google Scholar.

Page 122 note 6 See Hyde, op. cit., p 68 (de Indorum ludo Tchūpur); Falkener, op. cit., p. 267; Tylor, E. B., “On American Lot-Games,” in Internationales Archiv für Ethnographic, suppl. to vol. ix, 1896, pp. 37Google Scholar.

Page 122 note 7 Illustrated in Hyde, Falkener, and Tylor.

Page 122 note 8 “Indische Studien,” vol. viii, p. 193.

Page 123 note 1 Op. cit., p. 230.

Page 123 note 2 Jaina Sūtras, 1895, p. 303, note.

Page 123 note 3 Jacobi, , Journal of the German Oriental Society, 1896, p. 227Google Scholar.

Page 123 note 4 Op. cit., pp. 228–32.

Page 124 note 1 The European successor of the Oriental ‘councillor.’

Page 124 note 2 Sachau's Translation (Trübner's Oriental series), vol. i, p. 183. The opinion of Falkener (op. cit., p. 139) that Albērūnī did not know the game, is due to his not understanding Albērūnī.

Page 124 note 3 See below, p. 137.

Page 125 note 1 Athenæum, July 24, 1897, p. 130.

Page 125 note 2 Cf. Beal, , “Buddhist Records of the Western World,” vol. i, p. 210Google Scholar, note 13.

Page 125 note 3 See “Harṛsacarita,” chapter ii: cf. Beal, op. cit., p. 210, note 18.

Page 125 note 4 Beal, op. cit., i, p. 215 ff.

Page 125 note 5 Bãna mentioṛs the artāpada in another passage of the Harṛacarita (see Cowell and Thomas' Translation, pp. 6 and 266) as well as in his Kādambari” (Nirṛaya Sāgara Press, 1890, pp. 180, 376)Google Scholar. It is natural to suppose that in both these passages he meant the same thing as in the present passage, and not ‘diceboard,’ as the smaller St. Petersburg Dictionary interprets it in the Kādambarī.

Page 125 note 6 Cowell and Thomas, p. 65.

Page 126 note 1 See the references in Nüldek's, “Persische Studien”: Sitzungsberichte d. Wiener Akad. d. Wissenschaften, 1892, pp. 23–4Google Scholar.

Page 126 note 2 English Translation by Sprenger, 1841, pp. 171–5; Text and French Transl. by Barbier de Meynard, Paris, 1861, pp. 55–61: cf. Linde, op. cit., vol. i, pp. 1–3.

Page 126 note 3 The title of the Arabic translation of the Pañcatantra: cf. Noldeke, op. cit., p. 22.

Page 126 note 4 Mas'ūdī also attributes the invention of Nard to an ancient Indian king, but adds that according to others it was invented under the Persian king Ardashīr Babakan: cf. Linde, i, pp. 2–3. The twelve squares of the board on which it was played he explains as the twelve months, and the thirty pieces as the thirty days of the month. The decision of the dice he explains as the dependence of man on fate. This interpretation of the game is practically the same as that of Yaq'ūbī: see Nōldeke, op. cit., pp. 22–3.

Page 126 note 5 Corroborated by Arabic scholars at the Oriental Congress held at Paris last September: see Athenæum, Sept. 18, p. 387.

Page 127 note 1 Cf. Schroeder, L. v., “Indiens Litteratur und Kultur,” pp. 718, 723Google Scholar.

Page 127 note 2 Schroeder, L. v., op. cit., pp. 723–4Google Scholar.

Page 127 note 3 Cf. the arithmetical progression attributed to Āryabhata by ṛaṛguruśiṛya, ed. Macdonell, p. 180.

Page 127 note 1 Biographical Dictionary, translated by de Slane, MacGuckin (Oriental Translation Fund), Paris, 1845, pp. 70–5Google Scholar.

Page 128 note 1 References in Linde, i, p. 7.

Page 128 note 2 Edited with Gujaratī and English translations by Dastūr Peshotan, Bombay, 1885; text and German translation by Salemann, , “Mittelpersische Studien,” Mélanges Asiatiques tirés du Bulletin de l'Acad. des Sciences de St. Pétersbourg (St. Petersburg, 1887), pp. 222–30Google Scholar; discussed by Nöldeke, , “Persische Studien,” Sitzungsberichte der Wiener Akademie, 1892, pp. 20–6Google Scholar: cf. “Grundriss d. iranischen Philologie,” vol. ii, p. 145; “West, ibid., p. 119, § 103.

Page 128 note 3 “Persische Studien,” p. 26.

Page 128 note 4 This omission must be accidental, for the elephant and chariot must have been included in the total of sixteen, the regular number of pieces on each side. The passage is important as containing the earliest mention of the number of pieces in the game.

Page 128 note 5 Yaq'ūbī and Mas'ūdī both speak of chess as dependent on skill in contrast with Nard: cf. Nöldeke, , “Persische Studien,” p. 24Google Scholar. In Nīlakaṛṛha's “Nītimayūkha” (seventeenth century) chess is described as a game dependent on force of intellect (krīṛā buddhibalāśritā): see Weber, , Monatsberichte, 1873, p. 711Google Scholar.

Page 129 note 1 Cf. West, , “Grundriss,” ii, p. 119Google Scholar; Nöldeke, op. cit., pp. 20–1.

Page 129 note 2 Cf. Nöldeke, , “Grundriss,” ii, pp. 145, 169 ffGoogle Scholar.

Page 129 note 3 The modern Kanauj, Sanskrit Kānyakubja.

Page 129 note 4 See Mohl's, Translation, vol. vi, pp. 306–12Google Scholar.

Page 129 note 5 Ibid., pp. 353–6. These are two different stories: cf. Mohl, , preface to vol. vi, p. 5Google Scholar.

Page 129 note 6 Ibid., pp. 312–6.

Page 129 note 7 Shatranj being as close an adaptation of catrang as Arabic phonetics will admit.

Page 130 note 1 It is interesting to note by the way that though this Pahlavi translation is lost, two versions of it are still in existence. The Syriac version made about 670 A D. and entitled “Kalilag wa Damnag” (from the two jackals Karataka and Damanaka in the Pañcatantra) was only found in 1870, the story of its recovery forming one of the most interesting chapters in the romance of literary history. The Arabic rendering of the Pahlavi translation made in the eighth century is a work of prime importance, because from it flowed other versions of these fables of Bidpai or Pilpay (a corruption of the Sanskrit vidyāpati, ‘chief Pandit’), which exercised great influence on the literature of the Middle Ages in Europe. For the bibliography of this subject see Lanman, , “Sanskrit Reader,” p. 313Google Scholar.

Page 130 note 2 Cf. Noldeke, , “Grundriss,” ii, pp. 144–5Google Scholar. Fīrdausi also relates the Kalīla and Dimna was brought from Kanauj to Persia under King Kisrā (Mohl, vi, 356–65).

Page 131 note 1 See Himly, , “Das Schachspiel in China”: Journal of the German Oriental Society, vol. xxiv, pp. 172–5Google Scholar; xxvii, p. 121 ff. Though the Chinese game shows several innovations, such as the introduction of artillery and of a river dividing the two halves of the board, it is essentially the same as the Indian. Thus it is played on a board of sixty-four squares with sixteen pieces on each side; and the order of the pieces from the corners is: chariot, horse, elephant, with the general (=king) in the middle. The Chinese game is clearly a war-game, and is described as such by a Chinese writer of about the eleventh century (see Liude, vol. i, p. 87, note 24). But the presence of the elephant in it was so striking that the Chinese call it the “game of the elephant” (Himly, I.e., p. 175). Professor Douglas tells me that elephants were numerous in China in the old days, and that the commentator Tso (who lived within a century after Confucius) says they were employed in battle between the states of Wu and Ts'u (b.c. 512).

Page 131 note 2 Himly, , Journal of the German Oriental Society, vol. xxxiii, p. 672Google Scholar.

Page 131 note 3 Cf. Linde, op. cit., vol. i, p. 136.

Page 131 note 4 Linde, op. cit., vol. i, p. 137.

Page 131 note 5 Ibid., p. 139.

Page 132 note 1 Linde, op. cit., vol. i, p. 141.

Page 132 note 2 Ibid., p. 142.

Page 132 note 3 Ibid., p. 144, cf. 24.

Page 132 note 4 Ibid., p. 143 ff.

Page 133 note 1 A peculiarity of the chess described by Nīlakaṛṛha in the Nītimayūkha (Monatsber. d. Berliner Akad., 1873, p. 707) is that the elephant occupies the corner, but has the move of the rook. Professor Weber here observes that this is the original position of the elephants in the Indian army, a statement probably based on Kāmandaki (xix, 37), who says that the array is the most formidable in which the elephants are on the flanks. There is, however, no historical evidence that the elephant ever occupied the corner in ancient chess. In Nllakaṛṛha's game the camel (probably under Persian influence) occupies the square (the third from the corner) of the old elephant. This is perhaps why the elephant has here been substituted for the unintelligible rook (as in Yida's game: see below, p. 136.

Page 133 note 2 So already in the Pahlavi chess-book: see above, p. 128.

Page 134 note 1 Mohl, vol. vi, p. 311; Linde, op. cit., vol. i, p. 67 ff.

Page 134 note 2 Mohl, vol. vi, p. 355; Linde, vol. i, p. 68. See above, p. 121 (dasapada in the Pāli sūtras).

Page 134 note 3 A few examples will illustrate this. The word has come to mean ‘rebuff’ from the warning ‘check’ ! meaning ‘(mind your) king,’ where the original sense has been so far forgotten that we even say ‘check to your queen’ ! The word also signifies a banker's draft (spelt cheque), from the earlier sense of ‘counterfoil of a bank bill.’ It further designates a ‘;square pattern’ like that of the chessboard. In Murray's English Dictionary nineteen meanings are distinguished in the noun ‘check’ and seventeen in the verb ‘to check,’ to say nothing of numerous subordinate senses. Then we have among derivatives the word chequer, with sixteen meanings as a noun and seven as a verb. Another is exchequer which originally meant ‘chessboard’ and then ‘treasury department of the state.’ This seemingly heterogeneous sense it acquired from the fact that it previously referred to the table covered with a cloth divided into squares, on which the accounts of the revenue were kept by means of counters. What an amount of history is compressed into this one word!

Page 135 note 1 See Linde, op. cit., vol. ii, p. 181.

Page 135 note 2 See Linde, vol. i, pp. 146 and 154, note 12.

Page 136 note 1 In the Chinese game the corner piece still retains the name of ‘chariot’ (Himly, Journ. of the Germ. Or. Soc, vol. xxiv, p. 173). On the possible phonetic connection between the Sanskrit ratha and the Persian ruhh, cf. Weber, Monatsberichte, 1873, p. 707, note.

Page 136 note 2 Cf. p. 133, note 1. Some interesting representations of the old chessmen of various nations may be found in the work on chess entitled Historia shahiludii (pp. 132–7) published in 1694 by the great Orientalist, Thomas Hyde, who was Professor of Arabic and Bodley's Librarian at Oxford two centuries ago. In one of these the rukh is depicted as a camel. It is interesting to notice that in dealing with Indian chess, Hyde gives twelve Sanskrit words for elephant engraved in Devanāgarī characters (evidently reproduced from the writing of a Pandit). These names are transliterated, sometimes incorrectly (e.g., as dvírada and as gudge). The meanings of most of these words are explained, in several cases wrongly (cf. especially that of ). In his account of the game of Nard (p. 68), Hyde also gives five Sanskrit words in Devanāgarī (e.g. ucksh). This is, I believe, the earliest instance of Sanskrit words in Devanāgarī appearing in any printed book.

Page 137 note 1 Sachau's, Translation, i, pp. 183–4Google Scholar.

Page 137 note 2 See Mohl, , Transl., vol. vi, p. 355Google Scholar: cf. Linde, op. cit., vol. i, p. 68.

Page 137 note 3 Mohl, loc. cit.

Page 138 note 1 Sachau, i, 184.

Page 138 note 2 Like its successor the bishop, with a limitation.

Page 139 note 1 Represented in Weber, , Monatsberichte, 1872, p. 67Google Scholar; Sachau, op. cit., i, 183 (reversed by mistake in the process of printing); coloured in Linde, op. cit., appendix to vol. i, p. 1, and in Tylor, op. eit. Falkener, op. cit., gives a photograph of the board and figures (a boat here occupying the corner).

Page 139 note 2 Weber, , Monatsberichte, 1872, p. 63 ff.Google Scholar, gives the Sanskrit text with German translation: cf. Linde, app. to vol. i.

Page 139 note 2 The game in the Pahlavi chess-book is clearly double chess. See above, p. 128.

Page 140 note 1 “Asiatic Researches,” vol. ii; quoted by Linde, op. cit., vol. i, p. 70.

Page 140 note 2 Op. cit., i, 68–9.

Page 140 note 1 Cf. Jacobi, , Journal of the German Or. Soc., vol. I, p. 233Google Scholar.