How can you deliver checkmate with a king?

If that is true, then you wouldn't say the player delivers the checkmate either since he does not personally attack the opposing king. And the FIDE laws are pretty clear about that.

Unless you want to semantically separate "delivering a checkmate" from "checkmating" which was never my intention and which is pretty irrelevant for the debate on "king killer". I only used the expression "delivering the checkmate" to separate the action from the state it causes. Doing that facilitates the comparison with similar operations like "delivering stalemate" which is the same as "stalemating" or "delivering check" which is the same as "checking".

I have given all arguments I have and I don't think we will come closer than this. I hope you can find like 4 billion friends on your viewpoint, otherwise I bet you will find the majority against you

That being said, it's still impossible to deliver checkmate with a king. Which is the question that was asked. A king may not check (or checkmate) the opposing king. As Martin said, the achievement is where the king moves and the result is checkmate. That could be a discovered check, or castling.

In a castling checkmate I'm sure you would agree it's the rook that delivers checkmate, not the king. Even though the king moves, so does the rook, and of course the rook is the last piece to move. But it's still considered a "killer king" award because the king was involved.

That is precisely what I do not agree with because castling is considered a king's move and the motion of the rook is a collateral effect, not a move . That decision is separate from the delivery of checkmate - and it could have been made differently. Had the rook been assigned the ownership role of castling then indeed the rook would also be the deliverer of the checkmate. Two different issues which occasionally deliver the combined result of a castling checkmate.

Btw, it is not the rook to move last in castling, that is only one implementation. The rules say that you can move the king and rook simultaneously. Choosing the king first is only a practical way to avoid debate about the touch-move rule. But it is perfectly legal to first touch the rook, then touch the king as well and finally make the castling move with both. Mathematically it is all nonsense since a move is indivisable, it does not have 2 parts. That is why the ownership is an arbitrary choice, it does not come from the sequence in the castling move.

Btw, btw, There is a fairy chess variant (MDR) which permits splitting moves in parts based on the execution of illegal or incomplete moves in a chess game and FIDE correction instructions. You can imagine that it dictates the rook-part of castling after the king-part was already played which creates separate events with separate ownership and causation. The splitting is also used in "joke puzzles".

Yes, move your king and if it can checkmate like become two rook mate, then you got it.

like someone's example in front

That seems pretty confusing. If moving the rook during castling isn't a move, what is it? How does the rook get to the other side of the king without moving it? I know it's not magic, I know it's not osmosis, I know it can't be done using the Force. If the rook isn't part of the castling move, how do you explain how it gets to where it goes?

If the rules say you can "move" both king and rook simultaneously, why does rule 4.1 say that each move must be made with one hand only. How does someone use one hand to simultaneously move the king and rook? Besides, doesn't FIDE say the king must be moved first and if not, then it's a separate rook only move.

All of this doesn't change the fact that in castling, it's the rook that delivers checkmate, not the king. It is against the rules for a king to check, or checkmate, another king.

That's a good point. Today, it appears the rules say you cannot use both hands to castle. You may touch the rook first but you have to move the king first in castling which means the rook comes second.

This is all implementation. For instance in digital environments you don't move the rook at all. And perhaps tomorrow there is a program where you click on 0-0 and don't even touch (click) any of the involved pieces. A "chess move" has a formal definition within the abstract chess system and is indivisible which is something very different from the natural language verb "move" which refers to the description of a immense number of tiny changes in physical piece positions which aggregate to the execution of a complete chess move. E.g., in the physical sense of the word "move" the chess move Bc1-h6 consists of a large number of muscle contractions and displacements of the bishop through the air before landing on h6. In the abstract universe of chess, a move is only the "total change" on a chessboard brought about in a single turn of one player (including moving captured pieces off the board).

Chess is not an analog system and has never ever been an analog system. It was dreamt up by a human mind as a game in discrete steps (moves). There is move 1 and move 2 but not move 1.3 or 1.8. And there is no slow change from move 1 to move 2. Except that we had to play it in an analog physical universe for thousands of years. Only in this century chess is returning home to the discrete mathematical game it always was by human design - with the help of the computer which is its natural, discrete, environment. All the FIDE paragraphs like 1.4 about how moves and corrections are executed are non-essential and disappear in the modern chess GUI's- without affecting basic concept definitions of moves, castlings, checks and checkmates.

That you continue to believe that the rook delivers the checkmate shows you do not understand the mathematical abstraction and do not understand that chess is a discrete game and that the assignment of delivery is arbitrary. That FIDE and WFCC gave it to the king in castling was a free choice but nevertheless a fact now. It is an illusion to believe that the check-giving piece has some priority over the other units participating in the incarceration of the opposing king. Since none of them can claim it, it has been given to the last unit making a change to the board position by executing a chess move which is the king in castling:

Wikipedia: Under the FIDE rules and enforced in most tournaments, castling is considered a king move ......

Yes, castling is a king move. But it's also a rook move. There is no possible way to castle without moving the rook. According to the rules the rook must also be moved, otherwise it's simply not castling. You are not allowed to move the king two squares in any other circumstance. It's probably considered a king move because the king moves first. That would make sense.

A discovered checkmate where the king moves out of the way is also a king move. But in both cases the king is not the checkmating piece. In both cases it reveals the checkmating piece. The rules do not allow the king to check, or checkmate the opposing king because the king itself would also be in check.

Whether chess is abstract or not has nothing to do with the rules that prevent the king from checkmating. The rules are the rules. Until they are changed, the king is not allowed to checkmate the opposing king. Yes, the rook is the checkmating piece, not the king. The reasons for check and checkmating even existing require the rook to be the checkmating piece.

I understand that there are different rules for different kinds of chess games, like fairy chess, fantasy chess, compositions, and certain other variations. In those variations there are probably all kinds of different rules that allow all sorts of crazy things that we don't see in real chess. But in real chess, where FIDE rules and USCF rules apply the king may not check or checkmate the opposing king. Only other pieces, such as a rook, may checkmate.

So the rules (so far) do not allow a king to deliver checkmate. In the case of castling, the king move of castling reveals the checkmating piece. Which is the rook.

Not true. Castling is only a king chess move by law. The concept chess move is fundamentally different from the word move in natural language. The ownership of the chess move has been assigned to the king, not the rook though that was an arbitrary decision by FIDE/WFCC. But obviously there is a practical advantage for choosing the king since a king skipping a square is easier recognizable as an "intention to castle" than a rook doing the same which could have "dual intention".

As I wrote before (and this is the last time) the rear piece in a discovered check, at best attacks the king and never checkmates since the checkmate is the result of an evaluation of the whole board position, not one piece. In fact, it is impossible to set up a position in chess where only 1 piece completely covers all squares of and around the king. There are always at least 2 pieces involved (which might include blocking pieces of the opponent). In order to select one unit, FIDE has decided to assign the checkmate delivery to the owner of the last full chess move whether or not it contributes to the checkmate.

Different rules of fairy/fantasy chess have nothing to do with the issue. But they can be useful to understand how concepts in chess may be different in variants. Chess moves have no parts in standard chess - they are indivisible - but that may be different in a fairy. The things you invent do not exist in chess but they do in some fairies.

It is not true that a king cannot check another king. It's ludicrous. It is true that a king cannot be in check by another king (a state) but it can play a checking move (an action). There are a massive amount of chess players and experts (about 99%) who do not know the chess rules well. You will find numerous discussions on chess.com proving that. On top of it the rules themselves are not everywhere clear which is why the FIDE laws are updated about every 10 years. As I explained it does not matter a lot how some rules are understood when playing games. Whether or not you know that a king delivered the checkmate does not change the outcome of your game. It only matters for how we communicate about chess and how we give killer king awards or instructions for compositions.

So "delivering checkmate" is not a matter of the game rules and they don't allow or disallow anything on that issue. It is about how we talk and what we mean by it!

The "state" of checkmate is when one piece (or two in the case of double checkmate) checks the king and the king cannot escape and the checking piece cannot be removed or blocked. So a king cannot checkmate, it's against the rules. Usually a single piece checkmates, such as a rook. The other pieces on the board, that help create the "state" of checkmate are not checkmating.

So maybe you have been misinformed, I don't know. But a king cannot checkmate another king. The rules of chess explain why, but it's also universally known that would be an illegal move.

And no, castling is not a king only move "by law". No such rules say any such thing. In fact FIDE 3.8 a. says the exact opposite of what you are saying. Here is the quote,

"There are two different ways of moving the king:
by moving to any adjoining square not attacked by one or more of the opponent’s 
pieces
or by ‘castling’. This is a move of the king and either rook of the same colour along 
the player’s first rank, counting as a single move of the king and executed as 
follows: the king is transferred from its original square two squares towards the 
rook on its original square, then that rook is transferred to the square the king has 
just crossed"

Notice the rules of chess specify that castling involves MOVING both the king and rook. Which is obvious since it's impossible to castle without moving the rook.

Again, the rules are the rules. No amount of wishing or hoping can change that at the moment. A king cannot deliver checkmate, because that creates an illegal position. Castling is a king move, where the rook also moves. And castling (or discovered check) by a king reveals the checkmating piece.

Maybe some day the rules will be changed to accommodate your wishes, but for now we have to go by the rules of chess to determine things like checkmate, castling, etc.

That is an achievement where the king moves and as a result, checkmate. It's not the king delivering checkmate though. The king moves to reveal the piece that delivers checkmate. The rook is delivering checkmate because only a checking piece can deliver checkmate. The king cannot check a king. In this example, would you say the pawn, or the bishop,

is delivering checkmate?

The pawn, because the pawn moving results in the checkmate.

OK. So the pawn is delivering checkmate, right? How is the pawn checking the king?

And if the pawn is not checking the king, how can a piece that is not checking the king deliver checkmate?

The dictionary defines deliver as "to send (something aimed or guided) to an intended target or destination". So how is the pawn aimed or guided at the target, which is the king? Which piece is actually aimed, guided towards the intended target? Which piece is "sending" checkmate?

In the "state of checkmate" nothing is known about which unit delivered the checkmate (though some things can be deduced) since that precedes the state. In the "state of checkmate" the king cannot participate, but it can in the move causing the checkmate which gives it ownership and the title "checkmating piece" - or killer king for that matter. In the precise same way that "a goal" in football means that the ball is in the net (a state) but it was delivered by the player shooting it (the action). He is the "goalmaker". Though the king's role is of lesser significance (that is not the point of the argument).

Articles 3 and 4 in the laws define the chess moves and how they are executed in a physical environment. To do that they use our natural language word "move". The result is the definition of the formal chess move. The game of chess is played with formal atomic chess moves - indivisible and without parts. This is clear on the game record sheet - which only contains complete chess moves no "moving parts" or execution sequences. And equally relevant is that none of these parts and execution sequences exist in the chess GUI's which indicates that they are irrelevant. A chess move is a "what" and not a "how". The "what" of all changes on the chessboard in a single turn. The remainder of the rules is exclusively based on full chess moves and not about "moving parts".

No law says castling is a king-only move because it is an implication not a law. Perhaps I overstated that. But the laws contain this definition in the glossary:

castling: 3.8.2 A move of the king towards a rook. See the article. In notation 0-0 kingside castling, 0-0-0 queenside castling.

and in 3.8.2: ‘castling’. This is a move of the king and either rook of the same colour along the player’s first rank, counting as a single move of the king and executed as follows:.......

Btw, your claim that a king cannot check(mate) another king comes completely from your imagination. It is nowhere in the FIDE laws. It is only a reasonable implication when you assume that moves are always delivered by players, never by pieces.

The problem is that the FIDE laws have no real use for delivery assignment to chess moves and therefore refrain from focusing on detailed definitions (as given in my next post) . Nothing keeps me from saying that Nb1-c3 is a bishop-move (or giraffe-move for that matter) as long as I execute it correctly. The assignment of a move to a unit only matters in the shells of the game - tournaments, compositions and communication. But Wikipedia and tournaments and problemists and chess.com folks got the spirit and only assign the castling move to the king. The fact that several units move has nothing to do with how we classify ownership of a chess move. These are not just two english words but a fixed term referring to certain defined actions in a chess game. So Bc1-h6 is a chess move, Rz6-y3 is not, while Rh1-a8 is debatable.

There are other points in your post I will not respond to. To me it's clear you lack the basic understanding of the mathematical nature and structure of chess. And you can forget about the high ground of "rules". As said before, the subject is about communication not rules though rules always play a part in justification. I'm sure you have had similar debates about these issues with others. Many people understand these things intuitively, some need to get a masters degree in math first and the remainder should stick to playing games.

Btw, the understanding level of the chess laws is much higher in the composition community than in the game community - simply as composers always visit the edges of the system to land their innovations in. For instance the players and arbiters in the game universe have no clue on how to handle the dead position rule or 75-move rule - even though it's obvious to anyone with a brain. And they have a ridiculous circular definition related to "series of legal moves". A famous arbiter described the meetings of FIDE technical bodies as purely political events where all common sense was left at the front door when entering the building. No wonder Elvis already left it some time ago.

You are right! Do not follow lfPatriotGames as he will take you from one confusion into the next. The delivery of the last full move determines the identity of the checkmater Though it does not matter in playing games, it will help you through communicating about puzzles and in other chess subjects. Like the "killer kings".

Btw, there are a number of ways to legitimately assign delivery of a (checkmate/check/stalemate/..) state which are all in the same action chain:

1. The chess laws consistently refer to players as deliverers of check or checkmate
2. Since that gives little information about what precisely happens on the board you can narrow it down to the full move containing the action, like "Pd4-d5 checkmated black".
3. Chess units on the board are often considered representing the player in a move. Stating that a pawn (or specifically Pd4) delivered the checkmate recognizes that the player animated the pawn to deliver the checkmate.
4. What applies to pawns equally applies to kings in delivering checkmates or other states. Only there is a catch in that 2 units are involved in a castling move, something I have been discussing at length with lfPatriotGames. Whatever the verdict on this does not affect anything in the first 3 points.

We also have USCF rules about check and checkmate. The emphasis is mine.

4. Objective and Scoring 
4A. Checkmate.
The objective of each of the two players in a game of chess is to win the game by checkmating the opponent’s king. 
A player’s king is CHECKMATED when the square it occupies is ATTACKED by one or more of the opponent’s pieces and 
the player has no move that escapes such attack. See also Rule 12, Check; 12C, Responding to check; and 13A,

In our example, which piece is attacking the square the opponents king occupies? The pawn or the bishop? Or to put it simpler, which piece is giving check? The pawn or the bishop?

We can't just conveniently change the rules of FIDE, USCF, and the definitions in the dictionary because we happen to not like them. All three make it literally impossible for a king to give check or deliver checkmate. If for some reason people feel that a king can give check, try it. Play in a tournament and use your king to attack the square the opponents king occupies. See how that goes.

Not move out of the way to reveal another piece that attacks the square the opponents king occupies, no, use the king itself to attack that square. See how that works out.

Oh yes, we can ignore everything USCF declares as this is an international platform and no-one here cares about the local folklores. In fact, they corrupt the proper execution of chess laws regarding the termination of endgames as has often been discussed in chess.com forums. It is ridiculous to group FIDE laws and USCF scribbles in a single sentence.

What is worse is that you do not understand what the USCF rules are stating. None of the quoted text refers to the anatomy of (the act of) "checkmating" only to the "state of checkmate" - which is only a diagram snapshot. To return to my football metaphor, it describes the presence of the ball in the net but not the way it was delivered there. The fact that the goalmaker is not in the net does not imply that he didn't score the goal. So, your quotes are about checkmate - all of which I agree with - and not about checkmating which they do not address.

As you know very well by now is that nobody claims that one king can attack another. But it can check(mate) the opposing king in the last move leading up to the state of check(mate) by making that move. I see nothing in your USCF rules which contradicts that - simply because those rules were not quoted or do not exist. None of which would surprise me for USCF, btw.

As long as you can't separate actions from states you will not get this. It's a choice of your own fabrication. But the people who invented the "killer king" are certainly aware of the distinction. Plus a few billion others. For some reason you still believe you are not alone. Sorry, you are.

The point is, the "Killer King" achievement is achieved by moving the king and delivering checkmate.