July 2024 - Xiangqi: Chess With Rivers and Elephants

When I learned to play Shogi a year and a half ago, I was vaguely aware that other variants of Chess had been flying around the place, historically. One specific one that I've had described to me at that point was the Chinese variant of Chess, called Xiangqi (象棋). It's spawned its own Korean subvariant that uses the same pieces, though with a slightly different board and movement rules.

I've been told, as I've had the game described to me many years ago, that it's a very difficult game, and with the very limited experience that I could gather, I can definitely second that. The pieces themselves are similarly easy to tell apart, if you're not willing to play it in what I would call "wingdings mode", which I'm not. My Hanzi is good enough to sort most of the pieces into what they're supposed to be, though in most versions of the games, the pieces on opposite sides (nowadays usually of different colours as well), will have different Hanzi, since apparently colour was expensive back when this game entered the regular citizenship in Imperial China.

This time, it's probably best to start with the board. It has two very striking features added to the standard grid-construction of a Chess-like gameboard. Both players have a 2x2 (we'll come back to those dimensions) box in the heart of the starting position. This is the "palace" of either player. Then the gameboard is bisected horizontally by a river. This feature will become very annoying later.

The pieces, unlike in other variants, move not on the grid spaces, but rather on the grid intersections. It's not much of a mechanical difference, though it does explain the markings in the palaces.

The General is once again the analogue of the king. This time, however, it's not allowed to leave the castle, and it can move along the grid (ignoring the diagonals used to mark the castle.

There is an additional form of checking that the generals can do, called the "flying general rule", whereby two generals may not directly face one another on the same grid-line. In this case it'd be allowed for either general, on their turn to capture the opposing one, making this effectively a check. The flying general rule is thus never actually played, but is relevant in that it'd make a player move into check.

To either side of the Generals is one advisor, which are the saddest stand-ins for bishops I have ever seen. They can't leave the castle either, and are only allowed to move one step diagonally. They're mostly there to block against chariots, and perhaps put themselves in the way of a cannon instead of the general. Because they live on the same diagonal, they're mostly going to block another, until one of them gets taken.

Then come the Elephants, which can move two steps diagonally, and can't jump over pieces. They also can't cross the river, so they'll stick to the half of the board that they start on.

The Horses are unique in the way, in that they can be blocked. While they can technically reach the same pieces as the knight piece in Western Chess, horses specifically move one step on the grid, then one step diagonally, and they can't jump over pieces. That means that they can be blocked by pieces standing directly adjacent on either of the four positions. This is perhaps the rule that's the most difficult to keep in mind when coming from Chess or Shogi.

All the way in the corners of the board, are the chariots, which are really just rooks by another name. They are arguably the most powerful piece, perhaps rivaled by the cannons, depending on the proficiencies of the player.

The Cannons are placed two spaces in front of the Horses. They also move like rooks, but can only take pieces by jumping over any other piece. The colour of the jumped piece does not matter. This makes "defensive lines" not so much of an option in this game.

Three rows in front of the backmost line, every second position starts with a pawn, which move like normal pawns until they cross the river, after which they may also move sideways a single step.

One might immediately note two very interesting things while learning the movement rules. There's no real analogue to the bishop in Western Chess, so the game heavily really relies on the Elephant and Chariot pieces. Doubly so, since the Horse's movement is so hampered by the no-jump condition, which makes it very easily blockable.

Promotion is far less prominent of a feature insofar that the only the pawns get to promote, and don't really get much more versatile. Their primary function has always been positioning for blocking horses, and facilitating taking pieces with elephants, and their promotion feature helps exactly with that function and very little else.

This already has a tendency to make this game slower paced than Western Chess, and Shogi. To make the late-game less so, some additional loss-options are introduced. "Chasing" pieces and perpetually checking the Generals are ruled as losing moves. The same is true for move-repetitions, which will be relevant in late-games, in which not enough pieces remain to checkmate the General.

I've not had the opportunity to play against a person, and playing against a computer usually makes me feel a bit stupid. It certainly doesn't help, that I haven't managed to win a single game yet, no matter how badly I tell the computer to play. While part of the difficulty in getting a good grasp of the game is certainly the new movement, I also just don't think this is compatible with the way I tend to play for control of the board. Xiangqi's approach to controlling the board is mostly contingent on positioning with the cannons. The best way I might compare the experience to Western Chess is positioning your pawns in such a way, that one can exert control only using the queen. This is a style of play that one really needs to key into to move the game onto the opponent's side of the board.

I'm also very much wondering about the point of the Elephant, especially since it features so heavily in the name of the game. All elephants live on the same diagonals, but can't cross the river, so they can make two moves forward, before they have to retreat. The way they move and are positioned just doesn't seem like it would make them suited for defensive play, and since they're confined to the home-rows of the board, they're not much use for board control either. I tried using them for positioning, but I'm not good enough yet for it to make a difference, and I've not seen them utilized by the computer much either.

The reduced number of pawns makes them somewhat more valuable as material for positioning, especially when it comes to using the cannons. A cluster (promoted) pawns can really open they option for something like a firing line - provided they survive long enough for the cannon to get into position.

Beyond these specific observations, I've also found that it's significantly harder to adjust my own playing style. For example, since the flying general rule also makes it possible to checkmate with only a single chariot left, so taking those out of the equation is perhaps the best general approach. I could not for the life of me figure out how to do that though.

In conclusion, I’m going to try my best to learn to play the game better - at least to the point to where I can beat any computer. Chances are, that I won’t love it as much as I love Western Chess and Shogi, but it’ll certainly be more fun, if I feel like I have a chance.

Next
Next

June 2024 - Migrating off Wix