Are there Chinese xiangqi chess hooligans?

Board Game Studies 5/2002


1 Board Game Studies 5/2002

2 CNWS PUBLICATIONS Board Game Studies CNWS PUBLICATIONS is produced by the Research School of Asian, African, and Amerindian Studies (CNWS), Universiteit Leiden, The Netherlands. Editorial board: M. Baud, R.A.H.D. Effert, M. Forrer, F. Hüsken, K. Jongeling, H. Maier, P. Silva, B. Walraven. All correspondence should be addressed to: Dr. W.J. Vogelsang, editor in chief CNWS Publications, c / o Research School CNWS, Leiden University, PO Box 9515, 2300 RA Leiden, The Netherlands. Tel. +31 (0) / Fax. +31 (0) Board Game Studies, Vol. 5. International Journal for the Study of Board Games - Leiden 2002: Research School of Asian, African, and Amerindian Studies (CNWS). ISSN (CNWS publications, ISSN) ISBN Subject heading: Board games. Board Game Studies: Internet: Cover photograph: Susanne Formanek Typeset by Cymbalum, Paris (France) Cover design: Nelleke Oosten Copyright 2002, Research School CNWS, Leiden University, The Netherlands Copyright reserved. Subject to the exceptions provided for by law, no part of this publication may be reproduced and / or published in print, by photocopying, on microfilm or in any other way without the written consent of the copyright-holder (s); the same applies to whole or partial adaptations. The publisher retains the sole right to collect from third parties fees in respect of copying and / or take legal or other action for this purpose.

3 Board Game International Journal for the Study of Board Games 2002 Studies / 5 c n w s

4 Editorial Board Thierry Depaulis (FRA) Vernon Eagle (USA) Irving Finkel (UK) Ulrich Schädler (GER) Alex de Voogt (NL, Managing Editor) Board Game Studies is an academic journal for historical and systematic research on board games. Its object is to provide a forum for board games research from all academic disciplines in order to further our understanding of the development and distribution of board games within an interdisciplinary academic context. Articles are accepted in English, French, and German and will be refereed by at least two editors under the final responsibility of CNWS, Leiden University. Affiliations The following affiliated institutes underwrite the efforts of this journal and actively exhibit board games material, publish or financially support board games research. International Institute for Asian Studies, Leiden Address: Prof. dr. WHALE. Stokhof, IIAS, P.O. Box 9515, NL RA Leiden (The Netherlands) Russian Chess Museum and Magazine Chess in Russia, Moscow Address: Yuri Averbakh, Gogolevsky Blvd. 14, Moskwa (Russia) British Museum, London Address: Dr I.L. Finkel, London WC 1B 3DG (United Kingdom) Universiteit Maastricht, Department of Computer Science, Maastricht Address: Prof. dr. H.J. van den Herik, P.O. Box 616, NL MD Maastricht (The Netherlands) Corporate Sponsor Game of the Year e.v. Patrons Patrons support the efforts of this journal through continuous financial support. If you wish to become a patron, please contact CNWS by post, fax or. We hereby thank all our sponsors for their generous support: Irving Finkel, Caroline Goodfellow, Niek Neuwahl, Thierry Depaulis, Thomas Thomsen, Spartaco Albertarelli, Jean Retschitzki, Jurgen Stierter.

5 CONTENTS 5 Editorial / Foreword 6 Articles Articles Posts Research Notes Notes de recherche Research reports Book Reviews Comptes rendus Reviews Alessandro Sanvito The Riddle of the Celtic Game 9 Yasuji Shimizu and Shin ichi Miyahara, with Kôichi Masukawa Game boards in the Longmen Caves and the game Fang 25 Susanne Formanek & Sepp Linhart Playing with filial piety some remarks on a 19th-century variety of Japanese pictorial sugoroku games 39 Irving Finkel Pachisi in Arab garb 65 Wolfgang Angerstein The Laska column game: renaissance of an almost forgotten queen variant with connections to chess 79 Thomas Thomsen Chess in Europe in the 5th century? 103 Jean-Marie Lhôte Martin Le Franc et la dame enragée 105 Arie van der Stoep Early Spanish board-games 111 Obituary / Nécrologie / Obituary: Ricardo Calvo (), by Ulrich Schädler 119 Antonio Panaino, La novella degli scacchi e della tavola reale, by Egbert Meisenburg 122 Grigori L. Semenov, Studies on Sogdian Culture on the Silk Road, by Ulrich Schädler 126 Leo van der Heijdt, Face to face with dice: 5000 years of dice and dicing, par Thierry Depaulis 130 Summaries / Résumés / Summaries 133 Instructions to Authors 143

6 6

7 Board Game Articles / Articles / Posts Studies / 5

8 The Riddle of the Celtic Game * / Alessandro Sanvito At the beginning of the 13th century, the game of chess, known in Europe for barely two hundred years, had already achieved a privileged position. Knowledge of the game was one of the seven probitates that distinguished the true knight, and popular poetry had dedicated to chess, especially in France, epics and novels which the bards recited under the vaults of the great halls when the gentlemen played the dice - or assemble a game of chess. This privileged position held by chess in the Middle Ages has been interpreted in different ways and has often been understood in the broader sense of great popularity. Although this is at least partly true, some clarification is required. Harold Murray has compiled an impressive number of chess historical testimonies from the Middle Ages, which led him to state that during the latter part of the Middle Ages, and especially from the thirteenth to the fifteenth century, chess attained to a popularity in Western Europe which has never been excelled, and probably never equalled at any later (Murray 1913: 428). More recently Richard Eales clarified the meaning of this statement by Murray with extremely interesting arguments (Eales 1985: 57), but it is probably Murray himself who, in a manuscript written in 1917 and published posthumously, provides the most correct explanation for his claim (Murray 1963: 32 -33): This association of the nobility with chess was so characteristic that one of lower rank to admit a knowledge of chess was sufficient to raise suspicions as to his identity []. The feudal household played naturally chess, but the game was not confined entirely to the noble and its immediate circle. From 1200 onwards, a succession of town statutes and ordinances prove that the game was also played by the burgess class []. Chess is often mentioned in connexion with taverns, and Wycliffe denounced the English clergy of his days because they frequented taverns, among other things, to play chess. It should only be added that these findings prove, if not the popularity, at least the widespread use of the game of chess in medieval Europe. The so-called chansons de geste made a significant contribution to this spread, especially the more romantic love songs that tell of the adventures of King Arthur and the Knights of the Round Table: The oldest example is the novel de Brut by Wace (1155), which is written on the between Based on the fantastic Historia regum Britanniae by Geoffrey of Monmouth, written in 1136 and 1138. The legends of Tristan and Isolde, the Lai Marias of France and the poems of Chrétien de Troyes and his epigones followed. The frequent recourse to the game of chess in this type of poetry finds its explanation in the characteristics of the game itself, which, going beyond the purely intellectual and in accordance with the customs of the time, were often used as a social metaphor with symbolic, allegorical or ethical content. A complete list of all chess-related passages in these chansons * Translation from the Italian U. Schädler.

9 10 B OARD G AME S TUDIES 5, 2002 de geste can be found in the work by Fritz Strohmeyer, published more than a century ago (Strohmeyer 1895:). These include such works as Les échecs amoureux, Ogier de Danemarche, La vielle (De Vetula), Flore et Blanceflor, the Roman de Durmat le Gallois, Lancelot, Huon de Bordeaux, the Percival by Chrétien de Troyes, the Tristan von Beroul, the Roland song, the Renaut de Montalban, the Vœux du Paon, the rose novel (Roman de la Rose) and the Queste de Saint Graal. The magical, the great and the wondrous (Le Goff 1999), which is expressed in every line of these works, is of course also clear in the chess-related passages: the chessboard and pieces are always very valuable and extraordinarily beautiful, made of gold or silver and with Set with precious stones. And yet the descriptions are mostly of a fundamental and rather casual nature and simply reflect the position of the game in the general consciousness. There is of course no lack of passages in which the mention of the game of chess becomes interesting, be it from a historical point of view or as a source for the chronological classification of technical details. The intention to astonish the reader, arrested and nevertheless important from a technical point of view, are some passages full of symbols and metaphors that tell of a checkmate in the corner, which indicates the special preferences of the best chess players of the Middle Ages. It would lead too far to take up the discussion of the chronology of these texts, their origins and oral or written sources at this point. Nevertheless, a few explanations are required to introduce the topic of interest here. A much debated question is whether Christian von Troyes was inspired by a story in the Mabinogion, an ancient and legendary Welsh collection, for his percival, which remained unfinished by the poet's death. The name Mabinogion is a creation by Lady Charlotte Guest, who first translated eleven stories from the Red Book of Hergest and the White Book of Rhydderch into English in 1849. These stories were passed down orally until they were written on parchment at the end of the 11th or beginning of the 12th century. During this long period of oral tradition, the connection with Celtic mythology was lost due to the influence of the Christian clergy, so that today we have to be content with fragments, allusions and names. I mention this because the Legend of Peredur contained in this collection, like other Welsh and Irish legends, is the richest source of Celtic mythology in the British Isles in prose form. Although the discussion is heavily clouded by nationalisms and pro-Celtic or anticeltic party affiliations, many researchers agree on the outstanding source value (Loth 1913b: 1-45; Frappier 1969; Nardi 2000). A. Micha writes (Micha 1978:): Le nom de Merlin apparaît pour la première fois dans la Prophetia Merlini de Geoffroy de Monmouth (1134), mais la légende remonte à une tradition popular galloise sur un chef ou un barde nomnée Myrrdin. Gwydd-bwyll In the Mabinogion, Peredur, the son of Evrawg, tells the story of Peredur, who is raised by his mother in the midst of peaceful and unspoiled nature.

10 A. SANVITO, THE PUZZLE OF THE GAME OF K ELTEN 11 He grows up as an innocent child from the country, considers knights like Gwalchmai and Iwein, whom he meets in the forest, to be angels, and when he makes them king of chivalric life and of court Hearts Arthur speak, the desire seizes him to go away and become a knight. In many adventures he slowly grows into his new role and overcomes numerous obstacles. One day he reaches the Castle of Miracles, where he enters a hall in which he sees a board (tabula) for the game gwydd-bwyll, on which the pieces move by themselves. Peredur takes sides with either side, who loses, whereupon the game pieces on the winning side rise and screams of joy as if they were human beings. Annoyed, Peredur takes the game board (tabula) and throws it into the lake. Trying to find the game again provides the framework for Peredur's final adventures. The events in the Castle of Miracles have been varied many times, but in all continental versions the gwydd-bwyll, a game familiar to readers of the time, is replaced by the better-known chess. The best-known examples of this are the Percival by Chrétien (approx. 1150) as well as Wolfram von Eschenbach's Parzival (approx. 1200), which incidentally nourished the not yet proven conviction that both works used the Celtic Peredur as a source. In some variants of this text, even more fantastic additions can be made: In the Dutch version of Gauvain, known as whale wine, the hero enters the castle and sees the chessboard in a hall, on which the pieces only move when they are with a magic ring be touched. In the corresponding French version, Gauvain et l échiquier, the story changes again: Here is the magical game board made of silver and ivory and flies through the halls of King Arthur's court, only to disappear in a miraculous way. Magical chess is also found in the Queste du Saint Graal, and in an English version of the Estoire de Merlin the invention of the magical chessboard is attributed to the magician Guynebans (Murray 1913: 747). Frappier (1978: 550) writes about these adaptations: Toute la machinerie romanesque fonctionne de plus belle: tournois, défis, quêtes, méprises, incognitos; nains, géants, pucelles persécutées, prisons cruelles, fontaines empoisonnées, philtres, longue démence, échiquiers, anneaux magiques, carole enchantée. With reference to the ancient Celtic games, Murray (1952: 34) specified: Board games are frequently mentioned in the earlier Welsh and Irish literature, but few details are given and no game-materials of the older games are known. Dice and lots are never mentioned in connection with these games, so we may infer that they were games of skill alone. What exactly gwydd-bwyll gwydd means wood was used in a game is still unknown. It is mentioned frequently in the Mabinogion and other ancient Celtic sagas, but without any further indication that would allow the more precise nature of the game to be understood. It cannot even be said whether the pieces were differentiated in order to distinguish the two teams. Of course, such an old and unknown game has aroused the interest of chess history research and, more generally, game history research, because although gwydd-bwyll is described as a magical game, it actually existed (Murray 1913: 746, note 20).

11 12 B OARD G AME S TUDIES 5, 2002 The Etymology of the gwydd-bwyll and the fidchell To fully understand the true meaning, we must go back to the learned explanations of a philological nature by J. Loth (Loth 1911:). In his study we learn that the Irish threw sticks for divination purposes. The very expression teulel pren myl wel vyé, throwing wood, was also used in Cornwall, while the Welsh used the expression coelbren for the same ritual, which means something like prediction in wood. According to Loth, an elementary form of writing consisting of straight and oblique lines on wood was used for this, and he comes to the conclusion: On peut en dire autant de l alphabet irlandais dit oghamique. Traces of this Oghamic script have been found in ancient Irish manuscripts. To denote a letter in this script, the Irish grammarians resorted to the word fid, which also means wood, and because wood denoted the letters in general, each letter was identified with the name of a tree. On the other hand, there is the popular belief that trees understand and communicate to whoever loves them. This belief led the ancient Celts to attribute something like intelligence to the trees. The Irish term fidchell is now nothing more than the equivalent of the Welsh gwydd-bwyll; both derive from an ancient Celtic form uidu-queisla, spirit of wood (Loth 1911: 411). Here lies the obvious idea that the stones of the unknown game were equipped with intelligence and could move independently. If we then also consider that moving wood, according to ancient Celtic tradition, meant questioning the future, the entire allegorical framework with its symbolic and magical references fits together in a meaningful way. The search for the traces of this Celtic game has to struggle with difficulties, which often have to do with wrong interpretations of obviously obscure terms, similar to the case of the Roman ludus latrunculorum: Because one inferred from the Latin sources that the game was very demanding without being able to find out the rules of the game, medieval scholars assumed a resemblance to the most demanding game of their time, the game of chess. This led to the custom of translating chess as ludus latrunculorum. This expression was given a twofold meaning: with the humanists it always denotes the game of chess, while with the Roman authors a completely different game is meant (Sanvito 2000: 11-12). With regard to gwydd-bwyll and fidchell, the ambiguity even intensified when the Welsh term was suddenly replaced. In some later translations of French novels of the Arthurian cycle into Welsh, the continental échecs is mentioned under the new Celtic name tawlbwrdd, which makes the confusion worse (Murray 1913: 746). Tawlbwrdd The tawlbwrrd is mentioned several times in the Ancient Laws of Wales, edited in 1841, a work attributed to Howel Dha. It is said to go back to the year 943

12 A. SANVITO, THE RÄTSEL DES K ELTEN-GAME 13 (Forbes 1860: Appendix E, XLVII, 1), while Murray does not date it before 1250 (Murray 1952: 62). At one point (p. 436) it is stated that in this game two groups of pieces face each other, one consisting of 16 pieces and the other of a king and eight simple pieces (Murray 1913: 746). Several respectable studies have provided less clarity than clarity about what is unclear. Forbes published a letter to the editor to The Chess Player's Chronicle in 1860 in Appendix E of his book cited above, Chess among the Welsh.The reader who, under the initials D.P.F. operates, explains that the term tawlbwrrd, which he found in the Glossary of Welsh Terms, should be translated as throw-board. And since throwing is to be understood as throwing dice, the game must be a game with dice, which in any case precludes identification with chess. This obvious conclusion is shared by all professionals today. The reader continues by stating that the group of 16 stones was white and the nine stones on the opposite side were black. In support of his view, the attentive letter-writer draws on a number of documents and interesting quotes, including one which states that this mysterious game was once given by a king and queen to a court judge as an official sign, with the condition that never to sell or part with it. The narrative recorded in the sources is supplemented by information on the value of the individual tiles, which enabled a partial reconstruction of the initial line-up of the game, which results from the clash of 16 versus 9 tiles. In the past, the meaning of throwing board related to the use of dice led some to think of the game as a forerunner of backgammon, and Wotton also represented this hypothesis in his glossary, albeit on a slightly different basis: Tawlbwrdd Mensa lusoria, similis, abaco qui in ludo usurpatur: occurrit inter domestica nobilium utensilia-lib. Iii, cap. 7. Cui autem ludo destinatus erat it abacus incertum (Wotton 1730: 583). At the end of his long letter to the editor, the author concludes that all these sources do not appear to establish as conclusive a distinction between tawlbwrdd and modern Backgammon, as between that enigmatic amusement and Chess. In his comment, Forbes asks himself whether tawlbwrdd could not be a modified version of the Roman ludus latrunculorum and assumes a resemblance to the Persian nard, the direct ancestor of backgammon. Another researcher, the Dutchman Antonius van der Linde, carefully reads through all of these messages and then replies with a tough statement (Van der Linde 1874: 58-59, note 23). This must of course be seen against the background of the time when imaginative interpretations of a nationalistic character tried to prove the existence of chess in their own country before the historically documented point in time. Murray also assumed a close relationship with the ludus latrunculorum, which was played without a dice, but he was referring to the gwydd-bwyll or fidchell and not to the tawlbwrdd. Such a reference would not come as a surprise, as the Romans also brought their games with them to Britain, such as finds from game boards

13 14 B OARD G AME S TUDIES 5, 2002 with 8 8 fields on Hadrian's Wall (for example in Chester, Corbridge, Richborough and Chedworth; see Schädler 1994: 50). But the great English scholar, whose knowledge always guarantees the greatest possible reliability, did not stop at these considerations, but formulated further hypotheses that should be paused for a moment. The Hnefatafl One of his last discoveries after careful research was to consider the Scandinavian game hnefatafl (hnefi means king, tafl comes from the Latin tabula and means table, board, game board). He reconstructed the development of the game from the original tafl to the later hnefatafl, which was known to the peoples of Scandinavia since around 400 AD. It was brought to Britain and Ireland by the Norwegians, from where it also spread to Wales. According to Murray, the old gwydd-bwyll (fidchell) was replaced around 1000 AD in both Ireland and Wales by the Scandinavian hnefatafl, which was particularly known in Wales under the name tawlbwrdd (Murray 1952: 55; Fiske 1905). Archaeological finds make it possible to reconstruct the game board for this game. Two fragments are particularly interesting, one from a grave in Vimose (Denmark), the other from the so-called Gokstad ship excavated near Bergen (Norway) in 1880 (Articus 1983: 92 fig. 3, 94 fig. 9). The former has 18 or 19 fields on one side, while the second has fields. The most important find, however, is still the board that was found in Ballinderry near Moate in the west of Ireland in 1932 (Fig. 1). It is made of wood and has 7 7 holes into which the game pieces were inserted. The central hole is surrounded by a circle, the four holes in the corners are marked with quarter circles. According to Murray (Murray 1952: Fig; Articus 1983: 93 Fig. 7), hnefatafl was played on these boards. In fact, in a Viking Age princely grave of the 10th Fig. 1. Game board from Ballinderry (after: Jack Botermans et al., The World of Games, New York / Oxford 1989, p. 119)

14 A. SANVITO, THE PUZZLE OF THE K ELTEN GAME 15th century in Oldenburg / Holstein (Northern Germany) a helmet-shaped or crown-shaped king stone made of bronze and 36 hemispherical game stones (22 made of walrus ivory, 14 made of whale bones) with holes in the underside found, into which chopsticks could be inserted in order to play with them on such a perforated game board (Gabriel 1985:). There are also written sources that Murray does not forget to mention: An English (or Irish) manuscript written during the reign of King Athelstan () contains a representation of the hnefatafl in the Latin text as alea in the Saxon form, which is very similar to the Has wimose fragment. The text begins with incipit alea euangeli quam Dubinsi, episcopus bennchorensis, detulit a rege anglorum, id est a domu Adalstani, regis anglorum, depicta a quodam francone et a romano sapienti id est Isrl. Although the diagram contains some errors, the manuscript commonly known as Alea Evangelii (Fig. 2) allows the reconstruction of the layout of the game ) Fields (Fig. 3). A manuscript in the National Library of Wales written by Robert ap Ifan in 1587 contains a description of the tawlbwrdd with a drawing of the game board. It has fields of which the second, fourth, sixth and eighth rows are shaded (Murray 1952: 61-63). Murray does not comment, but Bell, who on the basis of linguistic arguments in favor of the use of dice in this game, believes that the tenth row must also have been shaded (Bell 1969b: 44) (Fig. 4). The number of stones is still missing: For the Irish hnefatafl on fields a total of 73 stones are required, namely 48 white and 24 black and a king; Welsh tawlbwrdd has 37 pieces on the board, 12 white with a king and 24 black. 37 stones also came to light in Oldenburg, but there were 22 and 14 simple stones in addition to the king

15 16 B OARD G AME S TUDIES 5, 2002 Fig. 3. Modern reconstruction of the position shown in the Alea Evangelii manuscript (author photo) Fig. 4. Bell's reconstruction of the tawlbwrdd game board described in Robert ap Ifan's manuscript (author photo)

16 A. SANVITO, THE PUZZLE OF THE GAME OF K ELTEN 17 game pieces. The game board from Ballinderry has 49 holes, but of course no starting line-up can be obtained from the board alone. First considerations This somewhat cumbersome representation of the old written sources, which say something about the old Celtic game mentioned in the Mabinogion, mainly served to gain some important chronological clues. The Welsh gwydd-bwyll and Irish fidchell, which turned out to be one and the same game, disappeared around the turn of the 1st and 2nd millennia and were replaced by a little better known game called tawlbwrdd in Wales. Tawlbwrdd, whether it is a general name for a game board on which one played a game of dice, or whether it is the name of a particular board game, is in any case the focus of research interest. Murray advocated equating the game with the Scandinavian hnefatafl, which was introduced in the British Isles at the time, where tawlbwrdd may simply be the Welsh translation. However, if the Welsh term does not refer to a game but only to the board, then, according to Murray, hnefatafl would be the game that was played on that very tawlbwrdd. Of course, it cannot be ruled out that the whole matter is influenced by etymological misunderstandings in the chansons de geste of the Arthurian cycle: if, it seems, Chrétien von Troyes and his successors made use of oral or written Celtic sources, their encounter could be caused some confusion with the old gwydd-bwyll. It has already been proven several times that these poets replaced the Celtic game with chess that was better known to them and more suited to their needs. We also know that these works have been redesigned and translated several times, although some of these revisions, especially retranslations into Welsh, may have been relatively free and thus made understanding even more difficult. Murray himself provides two examples of this by quoting from Bown o Hamtwn and from Y Seint Greal, where the Welsh translation from English does indeed mention a tabula for a war game called tawlbwrdd (Murray 1952: 62). The complete suppression of the gwydd-bwyll by chess in all chansons de geste of the 12th and 13th centuries not only caused confusion, but also influenced the first serious research of the 2nd half of the 19th century, primarily aimed at proving that these ancient games were not identical with chess, and less at an etymological understanding. J. Loth feels compelled to make the following comment (Loth 1913a: 215 note 2): Gwyddbwyll, intelligence de bois ou bois intelligent. C est un jeu celtique, ressemblant beaucoup à nos échecs avec lesquels on aurait cependant tort de le confondre. Be that as it may, the tawlbwrdd is often mentioned in Welsh texts, and in what is most important in our context, we learn that it was a two-party game, one of 16 stones in two rows and the other eight stones around gathered a king. The list leaves no doubt that the

17 18 B OARD G AME S TUDIES 5, 2002 Game board 8 had 8 fields. Not a single game board, neither found nor taken from the written sources, can be reconciled with this arrangement, which says little, however, as there are several variants with 18 18, 13 13, and 7 7 fields, and therefore that one must also be viewed as one variant . Three Chess Manuscripts Now it has not escaped attention that the arrangement of the game as it emerges from the Ancient Laws of Wales surprisingly approximates a non-orthodox chess position. The very improbable identity of the tawlbwrdd with the game of chess should by no means be asserted here. Rather, it is a matter of introducing some analogous and long-known positions that are found in various chess manuscripts but have never been associated with the description of the Welsh text. A 14th century manuscript in the Cleveland, Ohio Public Library contains some Latin texts followed by an incomplete copy of the famous and widespread chess treatise by Jacobus de Cessolis. Immediately behind this, the folia is followed by a collection of 29 chess problems. Murray, who named the manuscript Archinto after a previous owner, puts it between 1370 and 1375 based on calligraphy and certain comparisons. He also noticed that many of these chess positions were taken from earlier Arabic manuscripts (Murray 1913:). Nine of these mansubat, as the Arabs called such endgame positions composed with a special aesthetic, go back to the famous Arab master al-adli, which is why the idea suggests that the Archinto's collection could be nothing other than a later copy of the work al- adlis. In our context, position No. 28 deserves special attention, the initial listing of which is identical to that described in the Ancient Laws of Wales (Fig. 5). It is accompanied by the following explanation: Rubeus primus aget & uincet to semper eundo. Fit mactum in diuersis tractibus. Pedites regine sunt et semper have duos tractus et denique victrices remanent cum victoria. Red attracts and wins by always drawing twice. The checkmate takes place in different ways. The pawns are queens and always have two moves, and then the victorious remain with the victory Fig. 5. Position 28 in the Archinto manuscript in the Cleveland Library (photo by author)

18 A. SANVITO, THE PUZZLE OF THE K ELTEN GAME 19 left. We learn that the white pawns (red in the picture) move twice and win easily, although after their promotion to a queen they can only move one square diagonally in each direction according to the then still valid Arab rules. In the illustration, the stones are designated with the initial letters, as was usually the case at that time, and the two parties are distinguished by red and black ink. Some interesting expressions in the Latin text of this part of the manuscript, and perhaps some diagrams, including the one we are interested in, seem to have been written by an Italian. The Rochus (tower) is designated with R, the Miles (in the text sometimes equus, eques, i.e. horse, rider) with e, the Alfilus (runner), in the text as calvus in the sense of old, wise (as often in medieval European chess texts) is abbreviated with c. For the rex (king) there is a hitherto incomprehensible b, perhaps to avoid confusion with the rochus. The Regina (lady) is denoted by f for ferz and the pawn with p for pedes. The Augusta City Library in Perugia also keeps a 196-page manuscript of the century under the signature MS 775- L.27, in which 65 pages are devoted to chess. Chess positions are presented according to the old rules as well as the new rules that emerged around 1500, according to which the queen and bishop were given their usual moves, which means that the manuscript is dated to the first years of the 16th century. The manuscript treated by Murray (Murray 1913: 733) and some Italian chess researchers (Roncetti 1977a: 9; Roncetti 1977b: 9; Chicco 1984: 10; Pratesi 1996a:; Sanvito 1996:) is particularly relevant because of some very puzzling positions. With the exception of position folio 63, the diagrams are shown without text, but they almost always have a heading and the letter D or f, depending on whether the new rules for the queen or the old rules for the ferz apply, even if the former are in the majority. In the diagrams, the red (white) and black pawns are marked with initials, just like in Archinto. According to Franco Pratesi, the manuscript was written by a single hand, whereas the title Ludus Latrunculorum is from a later hand. The enigmatic positions are masterfully presented. Fig. 6. Position 22 in the manuscript MS 775-L.27 of the Augusta city library in Perugia (photo by author)

19 20 B OARD G AME S TUDIES 5, 2002 Fig. 7. Position fol. 10 in the manuscript of the Malatestiana library in Cesena (author photo) was placed by Adriano Chicco, who wrote: Above the diagrams are headed Ludus Exstensus. The surprise is not due to the material transcription by the unknown copyist (...). The question marks stem from the fact that these twelve diagrams lack any explanation or solution, so figuring out the rules for these oddities seems downright hopeless. Chicco also immediately noticed that position No. 22 corresponds to No. 28 in the Codex Archinto (Fig. 6). To be precise, the colors are reversed, that is, the black stones are on top and the white (red) stones are on the bottom. In addition, the title Exstensus is followed by the letter D, which indicates the new rules. It is clear that under the old rules of the Codex Archinto the promotion of farmers had far less disastrous consequences than under the new rules in the Perugin manuscript. The title Exstensus was chosen by the author in the knowledge that this would only accelerate the result of the old task, but not change it.Finally, Franco Pratesi discovered a third manuscript of 356 pages in the Malatestiana library in Cesena, the first part of which was very similar to the manuscript in Perugia (Pratesi 1996b). The codex that is listed in the library as Ludi varii, idest Ludus rebellionis. Ludus subtilitatis primorum. Partiti de 2 (and subsequently to 13) tracti. Ludus ad capiendum ovines is written mostly in Italian, sometimes in Latin, and sometimes Spanish expressions appear in the text. Pratesi's dating around 1520 is to be agreed, because in other parts of the manuscript positions are described that already appear in older tracts, but also those from Lucena's Repeticiòn de amores: e arte del Ajedrez con CL juegos de partido, which appeared in Salamanca around 1497, as well as Damiano's Libro da imparare giocare à Scachi etc., first published in Rome in 1512. The author seems to be a chess expert and attentive reader of chess literature, whose intention was a complete collection of everything he had already seen, even if the many are empty Make diagrams think of a sudden interruption in work. The unknown author was sure to have had the opportunity to study the Codex in Perugia, since it was devoted to peculiarities

20 A. SANVITO, THE PUZZLE OF THE GAME OF K ELTEN 21 First part from pages 7 to 33 all so-called unorthodox positions with a few exceptions already appear in the Perugin manuscript.The diagram above on page 10 shows the position of the 16 black stones against eight white pawns and a king (also here in red), identical to diagram 22 of the other manuscript (Fig. 7). Only the title is there in Italian (d esteso) to indicate the new rules. What now seems interesting is less the question of the existence of these diagrams in the said manuscripts than their possible connection with the ancient Celtic game. This special position does not show any orthodox chess features, neither in the initial line-up nor in the double move of one side, and yet it was included in the collections of chess problems at that time. It is not without reason that this position is now counted among the heterodox variants according to strict modern classifications (Pritchard 1994: 43 see above Celtic C). Apparently this unique position has attracted the interest of many chess players and there has likely been a discussion as to which of the two parties has the better chance of winning. In any case, centuries later Carrera, one of the greatest experts of the 17th century, was still dealing with this problem (Carrera 1617). In his famous work, the learned priest from Sicily devoted an entire chapter to this question under the title Vantaggio di tutti i pezzi invece di due tratti per ogni colpo and made a very balanced judgment: some believe that he who is hit with all stones except for Pawns and the king, who are in their usual squares, achieve a draw if he can draw twice for each individual move by the opponent; It doesn't seem so to me, because I think that the one who plays with all the pieces has an advantage (vogliono alcuni, che colui, che si toglie dal gioco tutti i Pezzi, rimanendo co i Pedoni, co l Re solamente posti à luoghi proprij , riceua ugual partito, se all incontro possa semper giocar due volte per ogni colpo del nemico; à me non par così, perché intendo, che ne abbia il meglio colui che regge il gioco dei Pezzi). But a little later he concluded: The victory in the game with the pawns will be that some will be promoted to women, and so he will remain unharmed, because in just a few moves they will devastate the entire opposing field (La vittoria del gioco de Pedoni sarà, che alcuni di essi arrivi a Donna, e rimanga illeso, perché allhora in pochi tratti farà strage di tutto il campo nemico). However, there is no indication of the origin of this unusual position, which is simply referred to as heterodox. Conclusions and Implications The Welsh term gwydd-bwyll and the Irish term fidchell have been shown to refer to the same game. There are good reasons for the communis opinio that this game was a modification of the Roman ludus latrunculorum. We know that the Roman game was a strategy game, but that it had a completely different hitting mechanism than the game of chess (Schädler 1994). We know nothing more about the ancient Celtic game, although it is mentioned several times in the Mabinogion. An important testimony is the peredur, where the hero said that

21 22 B OARD G AME S TUDIES 5, 2002 magical game in the castle of miracles. Well-known researchers are of the opinion that the Peredur and other Celtic tales were the source for the chansons de geste des Chrétien of Troyes and his successors. In the place of the unknown Celtic game there is always the more popular chess game on the European mainland. The numerous revisions and translations into different languages ​​of the chansons de geste caused some confusion. We have also seen that in later translations of these texts from Old French into Old Welsh there is no mention of chess or gwydd-bwyll, but the new name tawlbwrdd is introduced. This tawlbwrdd is mentioned repeatedly in the Ancient Laws of Wales, which also shows the starting line-up with 16 versus 9 pieces, which can be found in later chess manuscripts. Research agrees that the tawlbwrdd has nothing to do with chess. The use of dice in that game, which results from the etymology of the word, rules out a relationship between the two games. Not to be underestimated, however, is the possibility that the game board as such could have been used for both kinds of games, a dice board game and a pure strategy game. Murray, on the other hand, believed that tawlbwrdd meant the board on which the Scandinavian hnefatafl was played. So there are some fixed points that allow us to make one or the other plausible thesis. The position described in three chess manuscripts is clearly a chess position. This is borne out by the Latin names of the game pieces in the Archinto manuscript, even if it is just as clear that the position must be more precisely described as heterodox due to the double move of the party with the smaller number of game pieces. The memory of the gwydd-bwyll, which the Celtic storytellers endowed with magical fantasies and which inspired the great medieval poets on the continent, remains limited to the Mabinogion. The period of the return of the game to Wales with a new name cannot be narrowed down to before 1250, a time when the problem collections circulated among chess players. The question arises whether the medieval chess players were influenced by the information in the Ancient Laws of Wales and included that unique position of 16 stones against 9 in their collections or vice versa. The assumption that the chess players, be they Welsh or continental European, whose championship is proven by a wealth of certificates (Sanvito 2002: 48-50), invented this position in order to make their problem collections more attractive, is certainly better justified. It should be remembered, however, that this particular interest in inventing new problems was not based on the beauty of the combinations that sprang from the inventor's imagination, but rather had far more prosaic motives. Large sums of money were wagered on these problems. And the position discussed here, which has been handed down for centuries in the collections of chess problems, was excellently suited for this.

22 A. SANVITO, THE PUZZLE OF THE GAME OF K ELTEN 23 Literature Articus, R From the board game of the Vikings. In: Duisburg and the Vikings. Booklet accompanying the exhibition from January 16 to April 10, 1983, Niederrheinisches Museum der Stadt Duisburg. Duisburg: Austin, R.G Roman Board Games. In: Greece and Rome IV,: and Bell, R.C. 1969a. Board & Table Games from Many Civilizations. Vol. 1, 2nd edition. Oxford. Bell, R.C. 1969b. Board & Table Games from Many Civilizations. Vol. 2. Oxford. Carrera, P Il gioco degli scacchi etc. Militello. Chicco, A I Misteri del Codice Perugino. In: Contromossa, March Eales, R Chess: The History of a Game. London. Fiske, D.W. Chess in Iceland and in Icelandic literature. Florence. Forbes, D History of Chess. London. Frappier, J Étude sur Yvain ou le Chevalier au lion de Chrétien de Troyes. Paris. Frappier, J La naissance et l évolution du roman arthurien en prose. In: Outline of the Romanesque literatures of the Middle Ages, vol. IV: Le Roman jusqu à la fin du XIIIe siècle. Heidelberg. Gabriel, I board game in Oldenburg a thousand years ago. In: Klaus Ehlers et al., 750 years of Oldenburg in Holstein town charter. Oldenburg. Le Goff, J Il meraviglioso e il quotidiano nell Occidente medievale, Roma-Bari. Loth, J Le sort et l écriture chez les anciens Celtes. In: Journal des Savants. Paris. Loth, J. 1913a. Les Mabinogion du Livre Rouge de Hergest avec les variantes du Livre Blanc de Rhydderch, traduit du gallois avec une introduction, un commentaire explicatif et des notes critiques, tome I, Paris. Loth, J. 1913b. Les Mabinogion, tome II. Micha, A L Estoire de Merlin. In: Outline of the Romanesque literatures of the Middle Ages, vol. V: Le Roman jusqu à la fin du XIIIe siècle. Heidelberg. Murray, H. J.R. A History of Chess. Oxford. Murray, H. J.R. A History of Board Games other than Chess. Oxford. Murray, H. J.R. A Short History of Chess. Oxford. Nardi, R Gli scacchi nella letteratura francese del Medioevo, Roma, Università degli Studi di Roma Tre, Facoltà di Lettere e Filosofia, anno accademico 2000/01. Pratesi, F. 1996a. Misterioso, ma oggi un po meno. In: Informazione Scacchi, 4. Bergamo. Pratesi, F. 1996b. Il manoscritto scacchistico di Cesena. Venezia. Pritchard, D.B The Encyclopedia of Chess Variants. Godalming. Roncetti, M. 1977a. Scacchi storici. Il codice perugino. In: La Nazione, ediz. dell Umbria, Perugia. Roncetti, M. 1977b. Scacchi storici. Il codice perugino. In: La Nazione, ediz. dell Umbria, Perugia. Sanvito, A Il manoscritto progressista di Perugia. In: L Italia Scacchistica. Milano. Sanvito, A L Arte degli Scacchi. Milano.

23 24 B OARD G AME S TUDIES 5, 2002 Sanvito, A Il Matto nell angolo. In: Torre & Cavallo, Scacco! Roma. Schadler, Ulrich Latrunculi a lost strategic board game of the Romans. In: Homo Ludens. The playing person IV. Munich-Salzburg. Strohmeyer, F The game of chess in old French. Halle a.d.s. van der Linde, A History and Literature of the Game of Chess. Berlin. Wotton, W Cyfreithjeu Hywel Dda ac eraill, seu Leges Wallicae ecclesiasticae et civiles Hoeli Boni et aliorum Walliae Principum, quas ex variis codicibus manuscriptis eruit. Ed. W. Clarke. London.

24 Game Boards in the Longmen Caves and the Game of Fang / Yasuji Shimizu & Shin ichi Miyahara with an appendix by Kōichi Masukawa The Longmen cave-temple, known as one of three major cave-temples in China, is located 13 kilometers south of Luoyang city, in Henan province. On both sides of the river Yi lies a tall mountain range known as the Gate Yi. A large number of big and small cave-temples exist in this area. Most of them were made between the Northern Wei era, when in 493 AD emperor Xiaowendi transferred the nation s capital to Luoyang, and the middle of the Tang dynasty (Longmen 1988). We had an opportunity to visit these world-famous cave-temples in December We observed the many Buddhist sculptures carved on the walls of the cave-temples, but at the same time we looked carefully for material pertaining to amusement pursuits. From previous reports on the Longmen cave-temples it seemed to be certain that no such materials could be found in the cave-temple carvings. However, many game boards have been discovered on the floors or on re-used building stones by Kôichi Masukawa (Masukawa 1995a; Masukawa 1995b). So we searched for similar material in the Longmen cave-temples, and identified one game-board carved on the floor of cave no We took some photographs and brief notes to record this material, but did not have enough time to investigate the other cave- temples. In July 1999 we had the chance to visit Luoyang once more. Again, we had only about two hours to observe the cave-temples. However, we could visit all of the cave-temples which we were allowed to enter at that time, and confirmed the existence of more game-boards in the caves number 20, 543, 1519, and 1628 (fig. 11). All the boards were carved on the floors. We will now introduce each of them following the numbering given on the signboards on the site. The game boards and their designs Cave no.20 (Qian xi si Cave) The cave (fig. 1) was carved in the rock during the Tang dynasty (AD). The main statue is the Amitabha Buddha. The ground plan is roughly square with a width of 9.4 and a depth of 6.72 meters, while the height of the grottoe is 9.7 meters. The game board (figs. 2; 11a) is a square incised in the floor. The length of each side is about 25 centimeters. On one side the carved line is not very clear. Therefore, we can see that the board has six vertical lines, but we cannot determine how many horizontal lines there are. However, six horizontal lines can clearly be identified. Thus the board is likely to be carved as a 6 6 lined board. On the side of the board there is a seventh vertical line, but since the horizontal lines do not touch it, its relevance is difficult to determine. Cave no.543 (Wan fu Cave) The cave (fig. 3) was carved during the Tang dynasty (680 AD). It is square-shaped

25 26 B OARD G AME S TUDIES 5, m Fig. 1. Cave No. 20 (Qian xi si Cave), partly renovated. Fig. 2. Cave No. 20: Game board. C A B 0 4 m Fig. 3. Cave No. 543 (Wan fu Cave), partly renovated.

26 Y. SHIMIZU, S. MIYAHARA, K. MASUKAWA, GAME BOARDS AT LONGMEN AND THE GAME OF FANG 27 Fig. 4. Cave No. 543: The smaller board. Fig. 5. Cave No. 543: The larger board.

27 28 B OARD G AME S TUDIES 5, 2002 Fig. 6. Cave No. 543: The board in the back room. 0 3 m Fig. 7. Cave No (Huo shao Cave), partly renovated.

28 Y. SHIMIZU, S. MIYAHARA, K. MASUKAWA, GAME BOARDS AT LONGMEN AND THE GAME OF FANG 29 Fig. 8. Cave No. 1519: Game board. with a flat ceiling. The main statue is the Amitabha Buddha. The front room is 4.9 meters wide, 4.28 meters deep and 5.3 meters high. The back room measures 5.87 meters in width and 6.85 meters in length, while its height is 5.8 meters. There are two boards carved on the floor of the front room and one on the floor of the back room. I noticed the two boards in the front room when we first visited the site in December The one in the back room we noticed in July 1999 during our second stay. In the front room there are a small and a large board. The smaller one measures centimeters (figs. 3A; 4; 11b). It can be described as consisting of four squares, each one crossed vertically, horizontally and diagonally, thus creating what is generally known as an alquerque board described by King Alfonso X of Castile in his book of The larger board (fig. 3B; 5; 11c) measures centimeters and consists of 11 vertical and 11 horizontal lines. The board in the back room (fig. 3C; 6; 11d) could not be closely observed, but it measures about centimeters and consists of four vertical and four horizontal lines. Since the points of intersection of the lines are not always marked precisely, it seems that the intention was to create 3 3 squares rather than 4 4 intersections. Cave no (Huo shao Cave) This cave (fig. 7) was created in 522, when emperor Xiaomingdi of the Northern Wei dynasty ruled the country. The main statue is the Sakyamuni Buddha. The cave is horseshoe-shaped and its height is about 10 meters, while it measures 9.5 meters in width and 12 meters in length. I found one board in this cave, but unfortunately could not get very close for a better view. The board (figs. 7; 8; 11e) measures approximately centimeters. It consists of a square crossed diagonally. Of the three interior horizontal lines only one is complete while of the other two only the positions are indicated. In comparison with the alquerque board in cave no. 543 one may suppose that the one here is an unfinished board of the same type.

29 30 B OARD G AME S TUDIES 5, 2002 Cave no (Ba zuo si Cave) This cave (fig. 9) was carved during the reign of empress Wu Zetian (AD). The cave is almost square-shaped, has three sides of terraced walls, and a domed ceiling. Its height is 4.42 meters, its width 4.62 meters, and its length 4.5 meters. On the terrace of the front wall, the main statue of the cave-temple stands with two disciples on both sides. The statues are seated figures and they are arranged as a triad. A game board (figs. 9; 10; 11f) is carved on the surface of a piece of concrete used to repair the cave. Therefore, it is certain that this board was carved very recently. There are five parallel lines in one direction, while of the lines in the other direction only three can be identified. Fig. 9. Cave No (Ba zuo si Cave), partly renovated. 0 2 m Fig. 10. Cave No. 1628: Game board.

30 Y. SHIMIZU, S. MIYAHARA, K. MASUKAWA, GAME BOARDS AT LONGMEN AND THE GAME OF FANG 31 Characteristics and distribution of the game boards At first, we could not figure out what these game boards were for, but after we went to Shaanxi province, we could interview many people and come to understand that many of the boards were fang boards. The games generally called fang seem to have slightly different names according to their number of lines. For instance, the four-lined board in the back room of cave no.543 is called fourfang, whereas the one in the front room with 11 lines is called eleven-fang. On the other hand also five-fang and seven-fang exist, needless to say their boards have 5 5 and 7 7 lines respectively. Subsequently we found people playing fang in Wuhan city of Hubei province. Furthermore we could get a lot of information about fang from a person from Hubei province, whom we met in Shaanxi province. This person was from Fangshan prefecture, near Hotei south of Beijing. Five-line fang is commonly played there also. He said that there fang is called cheng fang cheng long (with fang meaning square and long meaning dragon this term indicates that the player who creates a square will be lucky). According to his explanation, there are no oblique lines on this kind of fang board. We a b d e c f Fig. 11. Summary of Longmen game boards. a) Cave No. 20. b-d) Cave No e) Cave No f) Cave No

31 32 B OARD G AME S TUDIES 5, 2002 also received information from Mr. Mai at the Archaeological Institute of Shaanxi Province, who affirmed that during his childhood fang was frequently played in Qinghai province, where he came from. Even during this research trip, we saw people who seemed to play fang in the shade of a tree in front of Henan International Hotel at Zhengzhou city, Henan province. They were using stones and leaves as pieces and played on a board drawn by hand on the ground. It seems therefore that fang games are very popular throughout China. Concerning the board of the alquerque type, the difference from the fang-board discovered by Kôichi Masukawa in Xinjiang Uighur Autonomous Region (see below) is obvious: there the diagonal cross running through the central point of the 7 7-lined board is limited to the four central squares (or three central lines) only. In this instance we would like to point to a similar board of the alquerque type at Samarkand discovered by Kôichi Masukawa, which to our knowledge has not been published yet. On the marble throne in the courtyard of Shah-Jinda s mausoleum a board of the alquerque type, scratched into the marble surface, can be seen (fig. 12). Fig. 12. Samarkand: game board of the alquerque type carved on the imperial throne. Name and date unknown. (Photo K. Masukawa) It is very difficult to determine the date of the carving of the boards in the Longmen cave-temples. However, the one carved in the concrete surface in cave no. 543 must be fairly new to demonstrate that fang was often played in this area, even in modern times. The eleven-lined board in the same cave must predate 1936, since on fig.27 of the report by Seiti Miduno and Tosio Nagahiro, Research of Longmen cave-temples, published in 1941 (Miduno, Nagahiro 1941), taken from inside the Wanfo cave facing the Yi river, the board can be vaguely identified. Unfortunately the five-lined board cannot be identified from the picture. In this paper, we have introduced the six game boards carved on the floors of the

32 Y. SHIMIZU, S. MIYAHARA, K. MASUKAWA, GAME BOARDS AT LONGMEN AND THE GAME OF FANG 33 Longmen cave-temples. According to various information gathered in Shaanxi, it can be assumed that at least four of these boards are fang boards. Although there is still much to investigate about the date of the boards, it is ascertained that more than 60 years have passed since they were carved. Talking to the people we met in China, we were told that today fang is not limited to a small area, but is popular all over China. My informants also said they used to play a lot during their childhood, but today had few occasions to play. They added that young people today only rarely play catch, since they prefer new forms of amusements. Appendix Fang an unrecognized Chinese board game / Kōichi Masukawa (1) The discovery of the game Fang Asia was the home of some of the world s oldest civilizations. Particularly in the zones of the basins of the Yangzi and Huanghe various cultures developed from ancient times. Naturally pastimes and amusements were an integral part of peoples lives. We know of ancient Chinese sports, ball games, shooting matches and dice games, and there were also a few board games, for example liubo, the oldest of all which can be traced back to the 6th century B.C. at least. Weiqi was already played in the first centuries AD, shuangliu (Chinese backgammon) dates to the 5th century, xiangqi (Chinese chess) to the 11th century, dama (a race game with dice on a xiangqi board) to the 12th century and sheng guan tu to the 16th century, etc. However, China is a very large land with a variety of cultures, each with its own traditions and customs, so we can assume that there were many unknown old board games in China (compare Eagle 1998). One such game was played in Xinjiang Uighur Autonomous Region, among the Uighur people from the islamic sphere in northwest China (fig. 13). When visiting Xinjiang in early September 1992, I could observe an unknown game in the small village of Dabanchen, a little to the southeast of Turfan. Two young workers were playing fang I learned the name five years later, the board being drawn on the ground, and one player using pieces of broken pottery, the other using pebbles (fig. 14, 15). Fortunately, I was able to take photographs. In the first game there were only two players, and in the second game somebody also watched (fig. 16). Each game lasted about 15 minutes, including thinking time, and they played for small stakes, as is the case with Xiangqi in Turfan. This was the curious board game with 7 7 lines and a diagonal cross through the cen- Fig. 13. Fang board found in Xinjiang Uighur Autonomous Region.

33 34 B OARD G AME S TUDIES 5, 2002 Fig. 14. Catch players. Fig. 15. A catch game in progress.

34 Y. SHIMIZU, S. MIYAHARA, K. MASUKAWA, GAME BOARDS AT LONGMEN AND THE GAME OF FANG 35 Fig. 16. A Fang game in progress. tral point that hitherto I had not seen in Asia, including Japan. After my journey I studied many Chinese books and journals (Xu Jia Liang 1991; Shi Liangzhao 1992; Guo Shuanglin, Xiao Meihua 1995; Ma Guojun 1996; Zhan Fengming 1997; Wen Wu 2, 1992), but could find out nothing about the game . Nor was there any information in other European or American books on games (Culin 1895; Murray 1952; Bell 1979). About this time I delivered a lecture with photographs at a seminar held by our society of games history and I wrote in our journal (Masukawa 1993: 22-23, 25; Masukawa 1998: 65-66). But at that time I knew neither the name nor the rules of the game, which to judge from the photographs was a form of Go. When I again visited an area on the central reaches of the Yangzi in China in August 1997, on my return trip I unexpectedly met a student, Mr. Wan Rai, a native of Xinjiang Uighur Region. He matriculated in the Department of Asian History at Osaka University of Education, and told me that he had often played fang during his childhood in Xinjiang Uighur Autonomous Region, and that his grandfather had told him that fang was the traditional board-game of the area, although I could not verify this. Later I met him in Osaka and exchanged letters about fang, and learned how fang was played. The general rules of fang The game consists of two parts. The first is a game of envelopment (or position) like Go, in which each player occupies points, and the second half a war game. The win-

35 36 B OARD G AME S TUDIES 5, 2002 ner is the player who captures all of his opponent s stones. Normally the player with the black stones moves first. Playing on a board with 7 7 lines and therefore 49 intersections each player has 24 pieces. 1) The first stone must be placed in the center-point. 2) The players take turns in placing stones on the points of intersection of the lines, with the aim of creating a square around a field, while at the same time preventing their opponent from doing so. 3) When all the pieces are placed on the board, the game enters its second phase. 4) The players take turns in taking a stone of the opponent which is not part of a square off the board and then moving one stone vertically or horizontally from a neighboring point to the point left vacant by the removed stone. When a player manages to create a square of four pieces of his own he is allowed to remove an opponent s piece from the board. This is repeated again and again. So a player aims to create squares himself, as squares of four stones are safe and no stone can be taken from it. 5) Each player must take his turn to move, so that if necessary he must break up a square. However, he can rebuild the square with his next move, unless one of the stones has been removed during the opponent s move. 6) The players win by removing all of their opponent's stones. Bibliography Bell. Robert C Board and Table Games from Many Civilizations. Revised ed. New York. Culin, Stewart Korean games, with Notes on the Corresponding Games of China and Japan. Philadelphia. [Reprinted as Korean Games. New York, 1991.] Eagle, Vernon A On a phylogenetic classification of Mancala games, with some newly recorded games from the Southern Silk Road, Yunnan Province, China, In: Board Games Studies, 1: Guo Shuanglin, Xiao Meihua Zhongguo dubo shi {A History of Gambling in China}. Taibei (Zhongguo wenhuashi congshu, vol. 42). Longmen literature custody, division of archeology Cave-temples in China. Longmen cave temples. Beijing. Ma Guojun, Ma Shuyun Zhonghua chuantong youxi daquan {Complete Collection of Traditional Games in China}. 6th edition, Beijing. Masukawa, K Go. Tokyo. Masukawa, K Report of board games in China. In: Yugishi-Kenkyu. The Study of Games History, 5: Masukawa, K. 1995a. Sugoroku I. Tokyo. Masukawa, K. 1995b. Sugoroku II. Tokyo. Masukawa, K New research on board games in Japan and other countries, In: Yugishi-Kenkyu. The Study of Games History, 10: Miduno, S., Nagahiro, T A study of the Buddhist Cave-Temples at Longmen, Honan. Tokyo. Murray, H.J.R. A History of Board Games other than Chess. Oxford.

36 Y. SHIMIZU, S. MIYAHARA, K. MASUKAWA, GAME BOARDS AT LONGMEN AND THE GAME OF FANG 37 Shi Liangzhao Boyi youxi rensheng. Hong Kong. Shimizu, Y., Miyahara, S The Game Boards Engraved to the Longmen Grottoes in China, In: Yugishi-Kenkyu. The Study of Game s History, 12: 58-63 (in Japanese with English summary). Wen Wu {Cultural Relics}, vol. 2. Xu Jialiang Zhongguo gudai qiyi {Ancient Board Games of China}, 2nd edition, Taibei (Zhongguo wenhuashi zhishi congshu, vol. 25). Zhan Fengming Zhongguo tongyou shi {A History of Games in China}. Shanghai. Note 1. I am indebted to David Wigg, Frankfurt, for editing the English text.

37 38 B OARD G AME S TUDIES 5, 2002

38 Playing with filial piety some remarks on a 19th-century variety of Japanese pictorial sugoroku games / Susanne Formanek and Sepp Linhart In Japan there exist two board games, both called sugoroku, which are nonetheless different in nature. Ban sugoroku, meaning board sugoroku, is the East-Asian equivalent of the well-known backgammon or trick-track. It is played by two people and was brought to Japan via China in the 6th century at the latest. This has already been described by Masukawa Kôichi in an earlier article (Masukawa 2000) and does not concern us here. What we are dealing with is the second type, e-sugoroku or pictorial sugoroku, a game that strongly resembles the European goose game. (1) 1. On sugoroku in general The history of e-sugoroku is said to go back to the 15th century, but the earliest known surviving examples date from the 17th century. The first games of this kind in Japan seem to have had no pictures, but only words. Since the Buddhist religion never succeeded in transforming its terminology into pure Japanese, this terminology is very difficult for the Japanese to understand, and the first sugoroku are said to have served as a means for teaching the necessary Buddhist vocabulary. From these Buppô sugoroku, as they were called, the pictorial sugoroku gradually developed, and the earliest examples of this game extant today are also closely related to Buddhism: these are the so-called Pure Land or Paradise sugoroku, jôdo sugoroku, in which the players wander through the ten worlds of the Buddhist cosmos, that is, the six realms into which, according to Buddhist lore, living beings are continuously reborn unless they attain enlightenment, in addition to the four realms of enlightened beings, with Buddhahood or the Pure Land on top of the other three. With a starting square that usually represents the world of human beings, subsequent moves in these jôdo sugoroku can, on one hand, bring the players with finality to the winning square, that is, Pure Land the ultimate goal the faithful might want to achieve, or, with equal finality, to the worst of all realms, the most atrocious of Buddhist hells, the never-ending Hell Without Bottom. This square eventually constitutes a kind of negative counterpole to the winning square; reaching it meant that the player was out of the game. (2) It seems that, towards the end of the 17th century, these Jôdo sugoroku had become a popular entertainment not only for the Buddhist acolyte, but also for the common people. From this time onward, sugoroku games enjoyed increasing popularity and evolved into a variety of different genres. We can assume that this increasing popularity of the game and its diversification from the end of the 17th century onwards were related to the rapid development of the printing culture, first in Kyôto and in Ôsaka, and later, during the 18th century, in Edo, the present Tôkyô. E-sugoroku are also called kami-sugoroku, meaning paper sugoroku, because they consisted of one woodblockprinted piece of paper. Until the invention of the multicolored woodblock print tech-

39 40 B OARD G AME S TUDIES 5, 2002 nique around 1765, such prints were either prints in monochrome (mostly black on white), or prints that were hand-colored after printing. After the establishment of a standard print size, the ôban size, roughly 24 x 36 cm, a viable size for printing, sugoroku game boards were often assembled from several prints: two, four, or even nine sheets joined together, so that the sizes of the finished products vary, up to a size of about 90 x 70 cm. They were sold in folded form (rather like modern maps), in envelopes or wrappers into which they could be easily inserted again after use. These covers often featured pictures related to the game; they, however, are only rarely preserved today. It seems that these games were not backed with stiffer paper, so that they could be more easily stored away when not in use. Figure 1: Go-hôbi (The gift), detail of Kôfukô furiwake sugoroku (Figure 4). Two children at a play of sugoroku. The right one is throwing a die, while the left one holds a piece of paper as a marker in its right hand. Behind the game lies the illustrated envelope of the game.lauthors collection.

40 S. FORMANEK, S. LINHART, PLAYING WITH FILIAL PIETY 41 As a part of the townspeople s culture, these e-sugoroku fall, of course, within the manifold category of the Japanese woodblock prints commonly called ukiyoe; the same artists who drew the famous Japanese woodblock prints, such as Hiroshige and Kuniyoshi, were also responsible for designing the paper sugoroku games. (3) In order to play, players needed a die and markers or figures with which to play. In the first Buddhist sugoroku, a six character formula of a Buddhist sutra was used to write six Chinese characters on the six sides of the die, while the later, non-religious games used a die with 1 to 6 marks on the six sides, more or less like the ones we use today. Dice preserved in museums tell us that they were made from many different materials ranging from wood to ivory. The pieces representing the players seem to have been made of small pieces of paper, probably with the names of the players written on them (Figure 1). All sugoroku games have a starting square usually called furidashi and a goal called agari. If we analyze the kinds of paper sugoroku according to the rules of each game, we can discover four basic categories: 1. In the mawari sugoroku or roundabout sugoroku (Figure 2) the player advances for the number of sections stipulated by the number thrown with the die. The nickname doubtless arose from the fact that, since the game usually starts in one Figure 2: Nezu Sendagi tsukuri kiku hitori annai (Guide to the chrysanthemum exhibition at Nezu and Sendagi). A simple black and white mawari sugoroku, and at the same time a program to the chrysanthemum exhibition held in Edo in the autumn of 1845, by an anonymous artist. The players (and visitors) start in the lower right corner and turn in a spiral until they reach the goal, the chrysanthemums in the center. 34.6 x 43.5 cm. Authors collection.

41 42 B OARD G AME S TUDIES 5, 2002 Figure 3: Honchô nijûshikô sugoroku, drawn by Ichiyôsai Toyokuni III, texts by unknown writer, and issued probably at the end of Authors collection.

42 S. FORMANEK, S. LINHART, PLAYING WITH FILIAL PIETY 43 corner, advancing square to square along a Monopoly-like spiral course towards the goal in the center of the sheet, the players progress in a roundabout fashion. 2. Next comes tobi sugoroku or jumping sugoroku. Each field or square in the game gives instructions as to which square players must proceed to, when a given number is thrown. Sometimes, fewer than 6 numbers are given; In this case, players who fail to come up with one of the stipulated numbers have to wait for the next round in order to make a move (Figure 3). 3. The tobi mawari sugoroku represents a combination of the above two types. In structure, the game resembles mawari sugoroku, but some of the squares may contain instructions requiring players to make different moves, that is, to jump. 4. Lastly, we come to the furiwake sugoroku, parted or divided sugoroku, containing two (or more) different paths that advance towards the winning square. Frequently, these occur as specified paths for men and women, or for good and not so good people (Figure 4). The two sugoroku that we shall attempt to analyze here belong to category 2 and category 4. Attempts have also been made to classify sugoroku games according to their presumptive consumer categories, such as: sugoroku for grown-ups, sugoroku for the whole family, sugoroku for children, sugoroku intended for use in the entertainment quarters (usually with female entertainers), but in many cases it is difficult to place a particular game into a certain category. Our two games seem to have more in common with the first two categories rather than with the third and fourth ones; They were surely not intended primarily for children, but might have been introduced to children for educational purposes. Pictorial sugoroku are often classified thematically, that is, according to the nature of their contents, but this classification is not yet generally established. In his Nihon e-sugoroku shûsei, a standard collection of sugoroku found in many libraries, Takahashi Junji (1980) places all sugoroku into one of the following 16 categories: 1. buppô sugoroku, jôdo sugoroku, or Buddhist or Paradise sugoroku 2. dôchû sugoroku, or travel sugoroku 3. shibai sugoroku, or theater sugoroku 4. shusse sugoroku, or career and success-story sugoroku 5. rekishi sugoroku, or historical sugoroku 6. meisho sugoroku, or sugoroku of famous places 7. yûgei sugoroku, or entertainment sugoroku 8. bungei sugoroku, or literary sugoroku 9. kaika kyôiku sugoroku, or late 19th-century sugoroku for education on modernization 10. sensô sugoroku, or war sugoroku 11. refueling ryokô sugoroku, or sugoroku of discovery and foreign travel

43 44 B OARD G AME S TUDIES 5, 2002 Figure 4: Kô fukô furiwake sugoroku. Anonymous illustrator, texts selected by Ryokutei Senryû. Published between 1837 and Authors collection.