Chess: why can only pawns capture en passant?

First: what I think I know (please correct me as necessary).

The en passant rule came about as a corollary to the two-move rule for pawns. The two-move rule was intended strictly to speed up the game, not to provide a new way to avoid capture. So a new rule was created under which an enemy pawn may capture your pawn if you move it two squares and the enemy pawn is guarding the square you would have had to pass through first without the two-square rule. Right?

But what I want to know is, why can the capturing piece only be a pawn? If, for example, White moves f2-f4, and Black has a bishop controlling the a8-h1 diagonal, why shouldn’t it be able to capture en passant?

I think it has to do with the pawn’s limitation in capturing. The double move prevents an opposing pawn from moving towards the center as it otherwise would completely eliminate a pawn’s ability to capture another. Without en passant, the double move limits it’s powers in a much more significant way than any other piece.

Other pieces already have to deal with enemy pieces moving through their lines of attack. Allowing pawns to “sneak by” opposing pawns would have been disruptive to the way the game had worked, but letting them sneak past other pieces fits in with how the rest of the pieces alread behave.

The en passant rule (and castling etc) came in hundreds of years ago, so we can’t be sure of the exact motivation.

As Zenbeam points out, the pawn double move didn’t affect other pieces as much as the pawn - which lost its ‘right’ to capture the passing pawn.

I suppose you could add that pawns generally capture other pawns (since they are the same value - you have to have a good reason to sacrifice a piece for a pawn), so en passant does make a difference.

Incidentally, as a chess teacher, this is the trickiest move to explain - generally my pupils who have previously heard of en passant think of it as ‘if you move a pawn anywhere near one of mine, I can take it!’ :eek:

Sorry, I don’t understand. What has the centre got to do with en passant?

If you’re unable to capture pawns towards the interior of the board because they’ve passed you, you’re weakening the passed pawn signifcantly.

I’m not getting this concept! Have I been missing out on a chess move here? Are you saying that a pawn can take another pawn, just by moving past it, and without having to land on it diagonally? When can this happen then?

Yes.

If a pawn makes an initial double-move (d2-d4, for instance) and in doing so, bypasses the space guarded by another pawn (d3 being guarded by a black pawn on e4), in the very next move, the bypassed pawn, e4, can capture the pawn on d4 as if the d4 pawn was actually on d3.

I’m not sure what you mean by this. What is the difference between moving through a pawn’s line of attack (well, not line…) and moving through another piece’s line of attack (well, not necessarily line…)?

Boozahol Squid, P.I. has explained it (note that the en passant capture must be made immediately).

There are 3 special moves in chess: castling. promotion and en passant.

Sorry I still am confused. :o
Why does it matter whether you capture towards the centre and how is a passed pawn involved?

Well pawns do most of the capturing of other pawns, so if a pawn can ‘pass by’ a pawn capture by moving two squares (i.e. on its first move), then an ‘opportunity’ is lost.

True, but occasionally other pieces capture pawns, and is there not just as much of an opportunity lost if a pawn “bypasses” a capture by another piece?

(Not that the rules of the game actually need any defense or anything, but I’m just curious if there’s something I’m missing here in these rationalizations, other than “Meh. Other pieces will have their day; pawns need all the legs up they can get” :))

No, if (say) my e-pawn gets past your d-pawn then the latter has permanently lost all chance of catching the former; if your double move gets your pawn past one of my pieces then the piece can still backtrack and intercept the pawn anyway.

This would be teaching glee to suck eggs, but for the benefit of some others, there’s an exception in the case of Kings. In a King and pawn endgame, the pawn’s double move can make all the difference between a King being able to catch the pawn or not. (Trivial example: Black King on d2, White pawn on e2; White to move plays e2-e4 and the pawn is home and dry, whereas the King could have caught it after e2-e3.) But the owner of the King generally has an option as to how his King will set off after the pawn in the first place, whereas a pawn only has one line of approach to the enemy; so the pawn needs the equaliser more than a King does.

Ah, gotcha.

It’s considered advantageous to control the center of the board, so you have a reason to want your pawns to move towards the center.

However, pawns normally only move straight ahead, except when capturing. So in order to move your pawns towards the center, they have to capture something.

By allowing pawns to move forward two spaces, they can dodge over the space the opposing pawn is attacking. Thus, they deprive the opposing pawn of the chance to capture them and move towards the center.

En passant lets the pawn capture the one who just moved two spaces by moving to the space he would have landed on if he had only moved one space. In other words they get to make the diagonal move they would have made if he’d only moved one space. Thus, they aren’t denied their chance to move towards the center.

Make sense?

Thanks, tim. I wasn’t sure how to clairfy what I was saying. That’s roughly what I was thinking, and well laid-out.

Thanks for the responses everybody!

Many strategic explanations have been proposed, but I’m still not sure they get to the heart of my question (or equally likely, I just don’t understand them).

Let me lay out my case more methodically.

  1. At one time, pawns could only ever move one space at a time.

[Obviously, in every turn in which a player moved a pawn, that pawn was subject to attack by pawn or piece on the square in which it landed.]

  1. The beginning of the game came to be considered slow as each player maneuvered their pawns into favorable positions one square at a time.

  2. So a new rule was adopted allowing each player to initially move their pawns two spaces instead of one, at their discretion.

[This was only intended to move the game forward, and not to confer any new powers of escape to the pawns. In other words, the result of the game should be just the same as if the two-move rule never existed, only faster.]

  1. Thus the en passant rule was adopted, allowing a properly positioned enemy pawn to capture as if the two-move rule did not exist.

[But this still allows pawns to escape capture from pieces.]

So it still seems to me that allowing a pawn to escape capture from anything except another pawn still violates the ostensible purpose of making the game faster while not altering it in any other way.

The parenthetical–and especially italicized–parts are my own conjecture, obviously. Is there no record of debate–correspondence, articles, letters to editors–that might shed some light on how chess players of the time may have viewed these new rules?

I think the issue here is you are assuming the two-move rule was intended only to speed the game “while not altering it in any other way.” What I suspect happened was that the two move rule was introduced and it generally improved the game, but people sometime thereafter realized that it spawned up a pretty darn big loophole/strategic problem.

When, say, white had moved its pawns far enough up the board that black’s pawns could scoot by, it significantly reduced the white pawns power because they could not control their opposing pawns. Thus white would avoid advancing its pawns this far (and vice versa) and the middle of the board would get jammed up with these hesitant pawns.

To solve this problem, they patched things up not by revoking the two move rule, but by adding the concept of en passant capture. It didn’t totally change everything back to how it was before, but rather just fixed a particular problem that the two move rule created.

Sorry :o , but no, this doesn’t make any sense at all. :eek:

  1. It’s difficult to arrange to control the centre by capturing. Normally you just advance your d+e pawns (plus sometimes the c and f pawns).
    OK, the Saemisch variation of the Nimzo-Indian (1. d4 Nf6 2. c4 e6 3. Nc3 Bb4 4. a3 Bxc3+ 5. bxc3) does use a capture towards the centre. However Black gets counterplay against the doubled pawns. But this sort of variation is rare - and doesn’t involve en passant anyway.

  2. The centre is e4/d4/e5/d5. There is no en passant capture that finishes by controlling any of these squares. Indeed the capturing pawn would probably be moving away from the centre!
    In a variation of the Petroff (1. e4 e5 2. Nf3 Nf6 3. d4 exd4 4. e5 Ne4 5. Qxd4 d5 6. exd6 en passant), the White pawn is already on e5. It then captures away from the centre (and is recaptured anyway).