Sunday, June 22, 2008

Proteus -- when is a chess variant, not?

Steve Jackson must have a soft spot for chess variants. Although Steve Jackson Games doesn't really have a very extensive line, and what it does have tends to be stuff that beats a game theme to death (Munchkin, OGRE, Illuminati, GURPS etc.), SJG has no less than three different chess variants for sale.

Or does it?

Tile Chess and Knightmare Chess are clearly chess variants. Both use the rules and pieces of chess and just add some new twists.

Proteus, however, while it appears to be a chess variant, could arguably be something quite different. It's a different sort of abstract game that merely borrows the moves of chess pieces.

The usual rule for judging whether a game is in the chess family is the knight move, which is unique to chess. Proteus does have the knight move, so it can be considered a chess-family game. And it does use the same board as chess. And, of course, it uses many of the same pieces of chess and their symbols, although also adding one new piece. In all these ways it looks like a chess variant.

But the game plays a lot differently than chess, not the least because the object of the game has been changed. There is no King in Proteus and therefore no check or checkmate. Victory comes from scoring the most points. One can also win by forcing a situation where the other side can't move, but the main way to win is by attrition.

This difference in objective, as well as the rotation of pieces between identities, makes for a very unchess-like game. I think it's different enough, in fact, that I'd hesitate to call it a chess variant at all. It seems to be an abstract strategy game that borrowed some moves and nomenclature from chess, instead.


  1. "The usual rule for judging whether a game is in the chess family is the knight move, which is unique to chess."

    Beg Pardon? Where did you pick this one from?

    There are many indication of what makes a chess variant and not just another board game. Royal pieces is one. Non-random move (i.e. no dice), the pawn piece is also a good indication, along with its peculiar moves, and there are other conventions, more or less strict, but the Knight-move isn't one of those.

    Just for an example, try playing chess with the knights being replaced by another piece, and see how "chessy" is that, then try playing a non-chess game with knights.

    Alternatively, you can try to define what is "knight-move"? The leaping isn't compulsory, there are several variants with non-leaping horses. The one-straight-one-diagonal? Is that so important that it makes this piece the cornerstone of chess variancy?

  2. I belive I read it in one of R.C. Bell's books. When I have a chance I'll look for the cite.

    In the meantime, the Wikipedia article about the chess knight notes:

    The knight move is unusual among chess pieces. When it moves, it can move two squares horizontally and one square vertically, or two squares vertically and one square horizontally. The complete move therefore looks like the letter 'L'. Unlike all other standard chess pieces, the knight can 'jump over' all other pawns and pieces (of either colour) to its destination square. It captures an enemy piece by moving into its square.
    The move is one of the longest surviving moves in chess, having remained unchanged since before the seventh century AD. Because of this it also appears in most chess-related national games.

  3. The chinese chess knight move is one square orthogonal to an unoccupied square followed by one square diagonal. That's a non-leaping knight. That is also how R.C.Bell describes it.
    Describing this move as L-shaped is misleading since it ignores the requirement that the intervening square must unoccupied. I'm afraid the wikipedia description is inadequate.
    I suggest you look at some chess variant sites and discussions.

  4. While not leaping, the move of the "knight" in Chinese chess is clearly derived from the chess-family games.

    So far as I can tell, the knight move first appeared in the ancestral chess games in India about AD 620 and has the move has remained remarkbaly stable among all the various national versions of chess that appeared as the game spread during the ensuing centuries.

    It appears to be a distinctive marker for the chess family of games. Can a game be a chess variant without knights? I don't see why not. Gala is one.

    Having the knight move is merely evidence that the game is derived from chess and not some other abstract game. Having the knight move is dispositive that it's a chess variant.

    If it doesn't have a knight move then you might point to other characteristics such as the 8 by 8 board, the "royal" pieces and the non-random move, although I'm not sure those would always be conclusive. The pawn's peculiar move/capture is suggestive, but is different in Chinese Chess for example, even though thta's a clearly a chess-family game. The knight's move in Chinese chess is exactly the same as the western chess, taking into account that one game is played IN the squares and the other is played ON the points.

    I'm not sure why you would say that the "Knight-move isn't one of them." The knight move is more common among all the national chess-family games (and variants) than any other move except the king's move and the rook's. But the king's move of a single space in any direction is hardly distinctive and occurs in any number of games that clearly have nothing to do with chess. The Rook's move all along columns and files is also common to many other games that don't have any relation at all to chess.

    In contrast, I don't know of any game, modern or ancient, that uses a knight move without reference to chess.

  5. you seem to make a much weaker claim than you made before.
    On your Oct. 12 posting you wrote:
    It isn't really even a true chess variant, as no piece in the game has the chess knight move, which, being unique to chess, is considered its signature move."
    and echoed it in the current posting with:"The usual rule for judging whether a game is in the chess family is the knight move "

    That's what I was strongly disagreeing with.

    Now you write:"Can a game be a chess variant without knights? I don't see why not. "
    That was part of my point. So you're agreeing. Good.
    You're still giving too much weight to the knight as a characteristic of chess.

    On Chinese chess you write:
    "The knight's move in Chinese chess is exactly the same as the western chess,"
    No. It isn't. A knight in chinese chess be blockes. In Western chess it can't. Look it up.
    The point/squares difference is an aesthetic difference and has nothing to do with board geometry.

  6. I'll admit that I've reconsidered whether aknight-move is an absolute requirement for a game to be a chess variant, although it seems the strongest evidence when present that a game IS a chess variant.

    It's less clear that a game MISSING the knight move can still be a proper chess variant. In some cases it can, perhaps in other cases not, depending on other characteristics of the game.

    I think you make too much out of the leaping ability as the key characteristic of the knight move. It's not. Other games include "jumps," some with captures and some without.

    I think rather the most distinctive aspect of the knight move is the combination of the orthogonal move and diagonal move, which is unique to the chess family of games. When viewed from that perspective the Chinese Chess knight move is the same move. Even in Western chess the move can be considered to be a orthogonal followed by a diagonal. THe "L" move description is just a convention. Functionally it's identical to the orthogonal/diagonal combo.

  7. I recommend the following sites:

    The former also lists several books about the subject.