Western castling doesn’t exist in historical Indian Chess (Chaturanga). Instead, in some versions, there is a cool move whereby each king can once in the game make a knight's move.

I read that it is can’t be used after check. But is that correct? Is it indeed a right which is lost for good after being checked, or is it something which cannot be exercised while the king is in check, but the right will return after the check no longer applies?

I guess I am hoping that it can be exercised even when in check, because it’s so cool to have a chance to escape otherwise certain loss - like James Bond.

How do the two kings interact directly? Must they avoid being a knight move apart. And if one has used its ability can the the other check it allowing bare K to checkmate e.g. K+N?

And how universal in Indian chess is this rule, please?

EDIT: The reason for this is that I think that Chaturanga can be interesting for chess problems, particularly retrograde analysis, but we need to agree a standard "canonical" version of the rules. The get-out-of-jail-free card is a unique mechanism like nothing else in chess, but I can't find a clear definition of how anyone uses it.

EDIT 2: Some relevant links:

The last link given is the most relevant, see John Gollon's account. However, I submit that the text is not clear, and I'm looking for another primary source.

  • 2
    +1 for "like James Bond." Commented Aug 23, 2020 at 1:51
  • 2
    As a Indian I would tell that 'Indian chess Rules' varies highly between cities and almost no one play theses today. They could move likely understood as home-rules.
    – AKP2002
    Commented Aug 23, 2020 at 9:29
  • 2
    AKP2002 yes there is variation in over the board variants. However my main driver is chess problems: there are some interesting mechanisms, and it would be good to have a canonical version for problem purposes. I have written an article about it and sent to two Indian friends, but hope I could get an answer to the king move thing here
    – Laska
    Commented Aug 23, 2020 at 12:13
  • 2
    Could you give a link or name the resource where you read about this?
    – user24344
    Commented Aug 31, 2020 at 1:54
  • 2
    I think there is a confusion here between Chaturanga (the game played in the 5th century that evolved into Shatranj, modern chess, shogi, Indian chess and many other variants) and Indian Chess, the variant of the game that was played in the 19th century in India. "Indian Castle" is relevant for the latter, not the former. Relevant : chess.stackexchange.com/a/18927/14079
    – Evargalo
    Commented Sep 23, 2020 at 7:00

2 Answers 2


I was reading through "American Chess Magazine Volume 1" on Google Books, which is from 1897. There is an old article on the origins of chess therein, with the basic summary of how the game to be: it's reminiscent of today's Wikipedia as it was at the time. To my surprise, on page 264, there was direct mention of what you speak of! It occurs when speaking of chess's predecessor, Shatranj, which itself is a form of the much older, four-player game Chaturanga, aka "Indian Chess."

Here is the relevant text.

The King was limited Shatranj, as in our game, to a move of one square in any direction, except that once during a game he was allowed the move of a Knight ; i.e, from his own square to King's Knight's second square, to King's Bishop, third square, to Queen's third square, or to Queen's Bishop's second square. This Knight move was omitted and Castling substituted.

At the end of page 265, it says:

In all variations of Chaturanga, the common origin is shown by the retention of one object, viz.: the capture of the opponent's King. When this is accomplished it is a checkmate, "Shiek-Mate," the death of the King--the game is finished.

So there is a primary source, I suppose. It seems to me that the special knight move can only be used by an unmoved king. Also , reading into it, that an opponent can walk their king into a king's knight move range. Shatranj is a variation of Chaturanga, I think, and if the goal is to capture the king. So "stalemate" as we know it doesn't exist, and therefore so doesn't check. I admit that this interpretation of the check is my opinion, so it is still unclear, but what is clear is where on the board "Indian Castling" can take place.

  • 2
    Thanks for this Rewan. I am not entirely objective in this because it seems to me that compositionally and for endgames there will be a lot more fun if the get out of jail free card can be played at any point. If I see some kind of old source which definitely stays that this move is only available to an unmoved king then I will be convinced. But until then I will not extrapolate beyond what I’ve read
    – Laska
    Commented Oct 15, 2020 at 2:58

Chaturanga is a local variant played by Indian traders. This variant was created only 1400 yrs ago from ShadYantra or ShatRanjan (not Shatranj)..

In older version Board Castling happened in real but not like Rook King exchange.

KING possesses scepter or RaajDand. King and minister resides in Palaces marked with X signs. There were 8 places in total in King's state.

In indian Castling, king had to consult any Royal team member and could go to castle again.. but Castle was not present in Chaturanga (variant played by local traders)...

King can do castling unlimited number of times and no such rules to disallow king to save his life were present..

With due course of time, the dashpaad board (10×10) went in secret exile along with Kautilya Arthshastra. Ashtpada board(8×8) had no Castle and fictional castling was introduced by Arabs and Europeans.

  • 2
    Please mention your sources. The "King can do castling unlimited number of times" seems strange (but, plausible, of course). Commented Jul 2, 2023 at 4:13
  • Are you interested to learn the ancient game or just digging resources. The oldest game of chess was known as Dhoot and Dhyut and then Shadyantra.. Unfortunately, this website has some users heavily biased and if I mention my youtube channel then they will ban me.. hence, I dont mention my channel url
    – ShadYantra
    Commented Jul 2, 2023 at 21:03
  • I am interested in facts about chess and its predecessors (and variants for that matter). So, sources that focus on facts rather than tales would be the way to go (there are solid studies on tales, but tales themselves are not reliable sources). Regarding biases, everyone has biases. It's better to work on our own biases before we ask others to rectify theirs. It would be better to base discussion on facts and logic. Commented Jul 3, 2023 at 1:04
  • 1
    At least in the near future, I am not interested in learning the game dynamics of predecessors of chess in detail (or most chess variants for that matter). I am certainly interested in collecting data about old games not necessarily related to chess (many are lost, and many more will follow if don't do this). Commented Jul 3, 2023 at 1:05
  • 1
    @ShadYantra : if you have a relevant Youtube channel but don't want to (or cannot) share it, so be it. But could you please provide some of the sources you used when you work on your Youtube video about castling in Indian chess ?
    – Evargalo
    Commented Jul 4, 2023 at 7:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.