Chess and I: click here for the homepage. A closely related subject: Sensible names for fairy chess pieces.
§1 The general principle: prefer self-explanatory and unambiguous symbols
Unless it's about decorative living-room chessboards or showing off in glamorous social media posts, I think that chess pieces and chess piece symbols, including heterodox chess, should always look like what they are -- or, at least, that must be one of the options, in case several sets are available. No trick, no deception, no riddle.
That's important. Chess is a rational and scientific game, and heterodox chess should not be less rigorous and serious. Plenty of us like blitz or bullet clock settings -- we must concentrate on chess-specific reasoning, and mostly have no interest (while playing!) for graphic riddles and confusing cosmetic, aesthetic, thematic, cultural symbol choices.
Typical or average standard chess players, of course, should be one of the main targets for the spread of heterodox chess. Most of them won't be willing to learn idiosyncratic and discouraging conventions, if simple and obvious alternatives are at hand. I think we should keep things as friendly as possibile.
In any heterodox variant involving hybrid/compound pieces, made up of standard chess pieces, the default symbols should always be self-explanatory and unambiguous. Very good hybrid symbols are widespead on chess platforms and applications, although they encounter considerable resistance for some specific variants. (A practical suggestion for Chess Dot Com variant players: in the menu Settings > Themes & Colors > Piece theme, choose "Classic" or "Alpha". They are not the most beautiful sets, but [still?] have the best unambiguous hybrid symbols.)
In section §2 of my web page Pictures (The kind of fairy chess piece symbols I like) you can find a wide visual exemplification of what I mean, starting from basic cases and then extending to more specific types of fairy pieces. I'm not good at drawing, but I make myself understood.
The same principle can apply to fairy piece figurines/miniatures, for playing chess variants over the board. (I'm talking about standard-style chess sets, not thematic ones.) Some people, for example, may think that a miniature looking like a bishop with the addition of an unmistakable king element can be a great solution for the knight-bishop hybrid, aka the archbishop. Of course it isn't! That can be, of course, a bishop-king hybrid -- which is also a legitimate and interesting fairy piece. For simple and affordable solutions in this regard (variants with fairy pieces over the board), you may take a look at section §1 of my page Pictures: My no-cost handmade fairy pieces.
§2 Cultural neutrality and alternate sets of pieces
Cultural neutrality is a complex theme, and I'm not going to address it in my little web space on heterodox chess. Anyway, I stay convinced that typical or average standard chess players should be taken in utmost consideration for the spread of heterodox chess. Such players may come from any country and culture, but all of them can play chess and are used to its symbols and conventions.
It's very good that, for some PyChess variants, more than one set of pieces are available. My idea is that at least one of them should always be as self-explanatory and unambiguous as possible, so that an average chess player has only very little new information to memorize -- and will try and maybe enjoy that variant, not being discouraged from the beginning.
Providing that at least one such set is available, we'll certainly be glad to see that users can also select other sets, with different aesthetic, thematic, cultural features and orientations.
§3 Royalness, and a note on the king's finial (that is the top element or crowning ornament)
In standard chess, kings are the only royal pieces, and no other piece is royal. Some heterodox variants have different situations in this respect: that's why we need a royalness sign. On Chess Dot Com, a fleur-de-lis sign has been used since 2021. Previously, it was a glow (halo, aura) around the piece. I think the fleur-de-lis is a good convention for royalness. Good enough, at least.
In some variants, each colour has more than one royal piece. In that case, different signs could be used, also depending on specific rulesets. For example, I think special signs could mark the case of (different types of) shared royalness in variants such as Spartan Chess (which has half-royal kings, and the duple check rule). We could use barred kings, kings crossed by a zigzag line or with a one-half sign (½). As a more general alternative, the zigzag line or the one-half could cross the fleur-de-lis sign.
Whenever non-royal kings are present, I find it highly advisable not to use the normal king symbol for non-royal kings (nrK's, also known as commoners or, rather absurdly, manns). I have personally tried custom variants on Chess Dot Com, with a royal king and a non-royal king together, and I find that platform's convention very confusing and annoying. The king with a fleur-de-lis sign is ok, but the other one is not, since it looks exactly like a normal king, that is a royal king! One possible solution, among a lot of others, is to use the Synochess advisor symbol -- that's a non-royal king in Synochess, despite being something else in other related games and sets of pieces. The nrK symbol used for Knightmate and other variants on both pychess-alternates and greenchess.net (links here and here) is alright as well.
A short note on the king's finial. Standard king symbols and figurines/miniatures (I mean standard-style pieces) traditionally have a cross as the top element. That originates as a Christian sign. For this cultural reason, the cross is sometimes questioned. I think this little element can be freely modified or replaced, both in standard chess and in variants. For instance, New Army variants could display a wide range of different finials: I am against the gratuitous non-mnemonic changes typical of those set of pieces ("this change is purely aesthetic and thematic"...), but modifying only little details such as the king's finial would be nice and harmless.
§4 Non-trivial fairy pieces
Some variants have special pieces that are not common nor trivial, such as modified pawns, pawn-something hybrids, or hybrids involving pieces taken from games other than western chess (see Paradigm Chess and its unique Paradigm dragon, which is a bishop-xiangqihorse hybrid). In such cases, I think that individual inventions are welcome, as far as symbols are concerned. That can even include well-known fairy pieces as the ferz, the wazir etc.: one should consider the Chess Dot Com conventions as the most widespread, but there's not a real consensus.
(In September 2021, Chess Dot Com introduced Paradigm Chess30 as a test variant. On that occasion, two fairy pieces were added to the Chess Dot Com stock of fairy pieces: the Paradigm dragon and the xiangqi horse itself. The latter doesn't appear in Paradigm Chess30 nor in any other known variant of western chess, but is an interesting piece and can be used for custom variants. The symbol of the Paradigm dragon is a bishop with dragon elements, probably because dragons can represent or refer to Chinese civilization. That had already been decided by the inventors of Paradigm Chess30. But what about a special symbol for the xiangqi horse? In usual westernizations of xiangqi, the normal knight/horse symbol or piece is used, because of the obvious correspondance between the two. Avoiding confusion with the ordinary knight symbol is important, and the Chess Dot Com staff found a good solution. For the xiangqi horse, we have a knight/horse symbol with the addition, again, of evident dragon elements: in this case, no wings but dragon scales. -- I think the xiangqi horse might also interestingly combine with the rook, the queen, and the king. See my proposals on the web pages My own heterodox chess and Pictures.)
§5 Divergent pieces and movement-only or capture-only hybrids
"Divergent pieces are pieces that move in one way but attack differently. The chess pawn, which moves forward but attacks diagonally forward, is a classic example. Divergent pieces are especially prominent in Orda Chess/Mirror, Empire Chess, and Hoppel Poppel" (quotation from www.pychess.org/variants).
I don't know of any pre-existing general convention for the symbols of divergent fairy pieces. That's a pity: it would be very useful. The only sort of convention I've stepped into is in one of the alternate sets of pieces for Hoppel Poppel on PyChess ("other piece sets are available, which may help avoid confusion" -- great!, avoiding confusion is important in my opinion, so that I'm writing several sections on this very subject). In that set of pieces, a cartoonish image reminding of piece X is the symbol for the piece moving like X and attacking like Y; and a cartoonish Y is vice versa. That's nice and good, but works only for such a simple system: only two pieces that exchange their rules with each other. Anyway, starting from this interesting idea, one could add little sub-icons to the cartoonish symbols in order to provide complete and unambiguous information.
Besides, so-called divergent pieces can be seen as a subset of the special compound/hybrid pieces for which it's necessary to specify peaceful-movement rules and/or capturing-movement rules separately:
normal hybrids, X plus Y, nothing more to specify;
special hybrids (a), basically X but also Y for peaceful movement;
special hybrids (b), basically X but also Y for capturing movement;
special hybrids (c) aka divergent pieces, only X for peaceful movement and only Y for capturing movement.
For that reason, I invented a comprehensive general convention, covering all three kinds of special compound/hybrid pieces by means of two diacritical signs. See my proposal on my web page Pictures, section §2.
(As for what I called special hybrids (a), "basically X but also Y for peaceful movement", see the Spartan Chess piece called lieutenant or tagmatarchos, and the last paragraphs of section §2, Modified versions of Paradigm Chess30, in My own heterodox chess.)
§6 Spartan Chess (and Spartan Mirror Chess)
(Here and in the following sections, my observations on PyChess sets of pieces are mostly based on some notes I took in April 2024. Things might change as time passes.)
Spartan Chess was invented by Steven Streetman around 2010 and is a beautiful variant. The choice of symbols is not bad. It might be better, and I'd love the game even more.
King. I don't like the helmet as a king symbol (see rook-king hybrid below). A normal chess king symbol, perhaps with a modified finial, would be the perfect choice. (The finial could be a little helmet.) Since the Spartan army has two kings, each provided with half-royalness, the king symbols had better be crossed by a zigzag line or a one-half sign, as long as both kings are present. See the Royalness section above.
Hoplite. Ok, I like it.
Rook-king hybrid (a general or strategos in Spartan Chess terms). A normal self-explanatory and unambiguous rook-king hybrid symbol would be the best solution. Anyway, the official Spartan Chess choice is not so bad: it's a sideways helmet, recalling the standard rook symbol in its top part. On the other hand, why to display another helmet? (See king above.) Is that a riddle?! Do they aim at misleading blitz and bullet players?
Knight-bishop hybrid (a warlord or polemarchos in Spartan Chess terms). The optimal solution is obvious. The official Spartan Chess choice is far from worst, but the little shield had better be blank or, anyway, different from the hoplite symbol.
Captain or syntagmatarchos. Lieutenant or tagmatarchos. Ok, I like these pieces and their official symbols (but tyntagmatarchos is a typo!).
§7 Synochess
Synochess is an interesting variant designed in 2020 by Couch Tomato (Daniel Lee).
Royal king (general in ChinKor terms, but king in Synochess terms). I think a chess king symbol would be much better, perhaps with a modified finial. (The finial could be a dragon, as a symbol of Chinese civilization.) In my opinion, it's important to better distinguish the royal king from the non-royal king. See the Royalness section above.
Non-royal king (advisor in Synochess terms). This piece corresponds to the two advisors of ChinKor chess, but is only one and follows the rules of a non-royal king. Ok, I like the piece and its symbol. It's very important to distinguish it from the (royal) king symbol. See above.
Rook (chariot in both ChinKor and Synochess terms). A standard rook symbol is what we need, according to my principles, since it is a rook. No wheels, thank you. I can't stand "purely cosmetic changes", if the platform or application doesn't give any alternative. Some (western-style) sets of pieces for ChinKor chess display cannons with wheels: on Xiangqi Dot Com, for example. That's ok for cannons, and that's one of the reasons why I don't like wheels for a rook/chariot symbol. (The standard conventional rook symbol, that is the tower symbol, is alright. I think a tower is not offensive for anybody. Well, towers have no relation with the Asian names of the rook... but not even with its current English name! The English rook is exactly a "chariot" as well, from the etymological point of view.)
By the way, I've happened to see a Youtube video with a perfect user-friendly set of pieces for Synochess: no cosmetic change, no ambiguity at all. Rooks are rooks and kings are kings. www.youtube.com/watch?v=1mIOMMV5clc&pp=ygUJc3lub2NoZXNz (Synochess testing on Cutechess). On PyChess, such user-friendly sets of pieces are available for xiangqi but, strangely enough, not for janggi and Synochess.
§8 Some scattered and general notes on some other New Army variants
Orda/Khan's Chess, Empire Chess, Chak, Shinobi+ etc., all designed by Couch Tomato and available on PyChess, are probably very good chess-style games. Anyway, no user-friendly sets of pieces are available for any of these variants, and that's really a pity.
Each individual New Army variant has its idiosyncratic conventions, with a lot of links to historical and cultural matters, that can be interesting, but generally have no relation with the chess-style rules, and will never help you to win or draw any game. One piece is a chess knight, but has a totally different appearance; another piece, in another variant, is represented by the knight symbol, but its rules are wholly unrelated... as if it were a riddle.
Providing self-explanatory and unambiguous chess symbols for most pieces of such variants would be easy, but the PyChess community seems to pursue other priorities, as is demostrated by the impressive alternative set of pieces based on Mayan glyphs (though "also suggestive for the movement rules"), made available in 2023 for the game called Chak. It's very original and praiseworthy, but ordinary chess-style player-friendly sets are still missing. As things stand, only a few deeply motivated chess-variant aficionados can reasonably play these New Army variants; and, again, that's really a pity.
§9 About Mansindam aka Pantheon Tale
Mansindam is a very good and very difficult variant, invented by a person from Korea (Daphne Snowmoon), and playable on PyChess since Summer 2023. Mansindam is mostly based on shōgi (shogi, Japanese chess), but has the western chess non-pawn pieces, including the knighted and the crowned fairy pieces -- even the threefold hybrids, which are so rare in variants.
I won't say almost anything about the official names or identities of the pieces, and will consider only the western-style symbols (both the East Asian names/characters and the western or English names/symbols are official and original, for this variant).
Most of the symbols are very well chosen: self-explanatory and unambiguous, as I like them -- compound symbols for compound pieces. Based on the promoted rook and the promoted bishop of shōgi, all promoted pieces are crowned ones in Mansindam: any promoted X is exactly a crowned X, that is a (non-royal) X+K hybrid. For example, the N promotes to a N+K. (The only exceptions are apparent exceptions, the P becoming a pure non-royal K, and the Q, the Q+N and the K remaining unpromoted.)
As a crowning diacritical sign, a sort of sun has been chosen. It's simple and iconic. I like it. One minor but relevant problem is that the sun sign is placed on the top of the basic symbol. Putting it at the bottom would be better, for two reasons:
No confusion between the Q+N and the N+K could arise;
The sun sign, at the bottom, could perfectly combine with the already compound symbols too, since the R+N promotes to a R+N+K, and the B+N promotes to a B+N+K (threefold hybrids).
As I have said, most symbols are very well chosen. Not all, unfortunately. Not only the two threefold hybrids, but even the R+K and the B+K -- for absolutely no reason (?) --, have independent unitary symbols, with no trace of the sun. We find, regrettably, a ship, a rhinoceros, a tiger, and an archer. No rational principle, not even any thematic coherence, and the graphic design of these four pieces is not good-looking. (That's quite a big difference with respect to Tomato's symbols: Tomato's symbols are always pleasant to the eye.)
What a pity, again. I hope that an entirely self-explanatory set of pieces for Mansindam will be available soon... ASAP. Sooner or later. It would be so easy and obvious, just extending the use of the sun sign to any crowned piece. (... But, for the time being, February 2025, that's not a problem for me personally: I'm a ridiculous beginner at shōgi, and at Mansindam I'm maybe worse than level zero of any AI engine.)
See my thoughts and ideas about Mansindam subvariations at this link (section §11). As for shōgi or Japanese chess: link here in English (whole page) and collegamento qui in italiano (sezione §6).
A note on my atheism
Couch Tomato's article Designing Internationalized Pieces for Eastern Forms of Chess (2021, then 2024 as a PyChess page, link here) is accurate, detailed and interesting. I like some passages very much: for example, the section Recognizability (language-based imagery and peripheral vision, etc.). But, as far as the main points are concerned, Couch Tomato is in favour of internationalization but against westernization. That is so arbitrary, standing in the middle between strictly traditional local conventions and present-day unbound innovation: it sounds like some kind of religion that I cannot agree with. I cannot be against westernization (as one of the users' options, I mean). I am in favour of free choices and of comfortable choices for any relevant group of people, and for as many people as possible: and western chess, aka international chess, is an important reference game for more than half a billion people all over the world! (Choosing [western-style] king symbols and [western-style] bishop symbols without a cross is reasonable and good, if that is the question, but crossless sets of pieces are available for western chess too -- well, not so widely used, unfortunately. One should prefer to use pieces and symbols with no trace of Christianity or other religions, for any kind of chess and similar games.)
See section §4 of my page on shōgi.