Tafl Family

Sometime around or before 400 CE, a new variety of games started to be played by Scandinavian peoples, which are here (and elsewhere) collectively called Tafl games. There is certainly a vein of similarity in Tafl games to Chess, although this is probably coincidental or representative of a later melding of Chess and Tafl because it is likely that Chess was not yet invented and even less likely to have been witnessed by Scandinavians in those times. Chess and Tafl games are similar in their objective of capturing (or protecting) a king piece, in the movements of the pieces (Tafl pieces move like the rook in Orthochess), and in the speaking out of notification that a player's king has been put into danger. This last item is only known from the last known description of Tafl games being played, dating long after Chess had been introduced into Europe. More likely than the Chess connection is a connection to Hunt games such as Hare & Hounds, Fox & Geese, and Roman Bear Games. These games stretch back into antiquity in Europe and are similar to Tafl in their unequal opposing forces. It is easy to conceptualize the "Fox" (force of a single piece) of ancient hunt games evolving into a king assisted by a small army.

Tafl games were evidently popular in their time and traveled far and wide with Norse explorers and raiders to Iceland and many different parts of the Celtic Isles (Ireland, Wales, and Britain). It was the only board game played by the Saxons. The reign of Tafl, however, was not meant to last and it gradually fell from favor with the introduction of Chess into these parts, being nearly completely lost by the 12th century. Like many customs, however, it managed to stick around for a few more centuries in remote, isolated, and rural areas. The last two known descriptions of play, from Wales in 1587, and from Lapland in 1732, both seem to regard the game as an antiquated curiosity.

The known Tafl boards are all square composed of a grid of smaller areas where the pieces are played. The boards (or depictions or descriptions of boards) discovered have been of six different sizes: 7x7, 9x9, 11x11, 13x13 (rare), 15x15 (very rare) and 19x19: different, but always having an array of an odd number of cells. The central cell, and often the corner cells, bear a distinctive design. Some examples have additional distinctive or decorative squares at the sides that may have had some purpose or rule attached to them, such as not allowing opposing pieces to enter or even being the area the king piece must escape to in some reconstructions. They may have just dictated the initial placement of the counters, provided simple orientation on the board or have just added aesthetically pleasing decoration.

Some of the pieces are peg-shaped and fitted into holes, some were flat-bottomed and played at the vertices of a square tessellation on a board with an even number of squares and others, especially later ones, were played on the cells of a square grid.

The website Hnefatafl, The Game of the Vikings (http://tafl.cyningstan.com/page/238/other-board-layouts) provides a simple formula that most Tafl games use for the number of pieces in relation to the board they are played on: p = 6w - 29, where p is the number of pieces and w is the width of the board, in playing spaces. This gives a ratio of pieces to empty space of approximately 30%. More open space tends to favor the king by giving him and his forces more routes to escape; consequently less open space favors the attackers who will find it easier to make blockades. Usually the attackers to defenders (not including the king itself) ratio is 2:1.

Based on thee variations, it is also likely that rules of the games played were variable through time and geography and it may be that hopes attaining any complete and official set of rules that apply to all the boards and all of the games played on them are unfounded. Conversely, there exists enough information to reconstruct a standard set of rules that can be applied to any Tafl board and are unlikely to deviate in any significant respect from the historically-played games.

The precise rules to all of the medieval Tafl games have been lost to the mists of time. Although several ancient manuscripts and writings contain scant clues as to the setups and methods of play, no complete, unambiguous description of any medieval Tafl game exists. The most information that has, as of yet, been discovered is that of the famous Swedish botanist and father of biological taxonomy, Carl Linnaeus, in a 1732 diary of his expedition to Lapland in the northernmost parts of Sweden, called Lachesis Lapponica. Linnaeus, an astute scientist, was well versed in many fields well outside the domain of botany and paid careful attention to the manners and customs of the Sami people, although he did not speak their language. Amazingly, these Northern people still played a version of the ancient Tafl, called Tablut, at the time of Linnaeus' visit! Linnaeus witnessed and recorded the play of this in one of the most recent and most complete description of any Tafl game. Even Linnaeus' description, however, leaves many unanswered questions and would be far from sufficing as an official rulebook for the play of Tablut.

To provide the rules to any Tafl game, we should first cover the details that are well established and nearly unanimously accepted in the scant evidence that exist for the historically played games. The basic premise of Tafl is actually well understood, and the dirt is in the details as they say. Here, it is the ambiguous details that are poorly understood and left to whimsies of board game reconstructions, which are often little more than guess work.

It is nearly universally accepted that the objective of the "attacker" player, with the larger force, is to capture the king and the objective of the defender is to allow for the king to escape. Even the most definite ideas concerning Tafl games, however, will eventually have some doubt cast on them by a reconstruction. Here, one of the primary questions that arise is: Where does the king escape to? Most reconstructions say that the king has escaped once it has attained any cell at the edge of the board. Sometimes this is stated as escaping "off of the board", but this is just a matter of semantics because a piece that has attained an edge square could not be impeded from further advancement off board. (Unless, of course, the reconstruction being played only allows the king a single orthogonal move per turn, rather than the generally accepted move of the rook in Orthochess. In which case the attacker would have one last chance to capture the king before it escapes off board.) There are, however, many ideas that the king should be made to escape to a corner cell. Evidence for this comes from many existing historical boards which have special markings at the corners. These specially marked squares may also designate a place that a piece may be captured against or be for simple decoration.

Questions may also arise as to how the king is captured. It is generally accepted that the king is captured exactly as other counters are captured, via double custodianship. Also known as sandwiching, this just means that if a counter has two opposing counters on either side of it, so that all three of them are now in an orthogonal line with the odd counter in the center, the central odd counter is then captured and removed from the board for the rest of the game. Again, however, some dissent arises. This dissent, however, may only arise from misinterpretations of Linnaeus' account of Tablut or from misunderstandings by Linnaeus himself. It is easy to find 20th century descriptions (Murray, Bell, et al.) of the Sami Tafl game, Tablut, stating that the king was captured by quadruple custodianship, or being surrounded on all four sides by opposing pieces. It is easy to see, however, that this can make the objective of the attackers very difficult to nearly impossible. It is increasingly common to interpret Tablut rules and Linnaeus' description of them as meaning that a king is captured by quadruple custodianship only when it is still at the central square (throne or "konakis") and also by triple custodianship, being surrounded by three opposing counters where the fourth vacant side is the konakis. Interestingly, this also suggests that a king can not re-enter the throne or konakis once it has left. I would personally suggest using simple double custodianship capture for all pieces in all Tafl games. This not only seems logical, but is also simple.

Another (almost) universally accepted Tafl rule concerns the movement of the pieces. All pieces move any distance orthogonally (like the rook in Orthochess), provided that all cells passed over are vacant. This is nearly universal with only a few exceptions: a challenging and slower-paced game can be played, in particular on the smaller 7x7 board, by only allowing single orthogonal moves, one cell in any orthogonal direction to a neighboring and vacant cell. Sometimes this slower movement is applied to the king only. Other exceptions to this rule consider the central square off limits to the attackers or all counters, even including the king once it has left its throne. Others consider the corner squares or other specially marked squares to be off-limits to defender counters, and sometimes they are places against which a defender counter can be captured.

Capture by double custodianship as a rule has seen very little disagreement. Most of the exceptions to this have already been discussed above, i.e. the king being captured by quadruple custodianship at the konakis and triple custodianship by the konakis. Reconstructions have, at times, not allowed the king any capturing moves. The questions that arise that would need to be addressed in a Tafl Official Rulebook, concern compulsory capture, and capture at the edges and corners of the board. My personal suggestion is to allow L-shaped double custodianship capture at the corners and not to allow capture by single opposing counter for pieces at the edges. I would also not suggest a compulsory capture rule, for either side.

Other questions arise considering custodianship capture. Can more than one opposing counter, when they are in-line orthogonally be captured (multiple capture custodianship)? Presumably, a double capture is allowed as a total of two single captures, although theoretically, a rule only allowing the capture of a single opposing counter per turn could be implemented. Also, can a counter safely move and come to stop between two opposing counters (intervention)?

Although historical sources leave much ambiguity concerning the precise rules of custodial capture it is obviously prerequisite to decide on a set of rules before commencing a Tafl game. Although some methods described above will speed up or slow down the pace of the game, and some will offer more advantage for one side or the other, any of them will work to play. My personal suggestion is to always allow intervention between two opposing pieces without threat of capture after the turn. In other words, a capture can only be made by enacting the capture yourself and the opposing piece(s) are to be removed at the end of your turn. Also, a counter may enact up to three regular custodianship captures in one turn and would thereby capture 1, 2, or 3 counters, but multiple capture custodianship is never allowed. Multiple capture custodianship, illegal by my reconstruction and others, is where one counter moves to the end of a straight line of two, three, or more opposing counters that have a friendly counter at the other end of that straight line and then all of the middle opposing counters are captured.

Linnaeus described a rule stating that if the king should ever have a path of escape, the defender player must call out "raichi", to notify his opponent. If the king has two routes of escape, the defender wins the game as there is no way for the opponent to block both routes. In this case, the defender calls out "tuichu". Although this may be helpful for balancing the play for better odds to the attacker, it seems likely that this rule represents a later borrowing from Chess and may not have been used historically (compare to "check" and "checkmate"). This calling out rule is necessarily left in game play for Tablut but is typically left out of reconstructions for other Tafl games.

Deciding which player has the first move is one of the first questions to resolve. Historical sources provide little or no information as to this, but most reconstructions seem to favor the attacker having the first move. After all, he is the attacker.

The known Tafl games are described here under the names Ard-Ri, Brandubh, Ficheall, Tablut, Hnefatafl, Tawlbwrdd, and Alea Evangelii. Ficheall, however, is likely to not be a Tafl game, as there are historical documents stating that the opposing sides in this game are of equal number, something unheard of in a Tafl game. "Ficheall" is today simply used as a generic term for "board game" and may have been a term applying to different kinds of games in the past.

Sources