[Retros] Uniquely reconstructible openings (not Re: Variant PGs)

Noam Elkies elkies at math.harvard.edu
Wed Jan 27 17:38:08 EST 2010

> I suggested that it might be of interest to compose the longest

> proof game featuring an "official" game debut [=opening]

> [...]

I've wondered about this too. One natural start is the symmetrical
Bishop's opening line starting 1 e4 2 Bc4 3 d3 4 Be3, and now several
possibilities beginning with 5 Bxc5 seem feasible though I don't know
that any is "official". Likewise on the Kingside, with the additional
possibility 1 d4 2 Bf4 3 e3 4 Bd3 5 Ne2 6 0-0 7 Re1 etc. Another direction
is the Trompowsky line 1 d4 Nf6 2 Bg5 Ne4 3 Bh4 g5 4 f3 gxh4 5 fxe4,
though this doesn't reach your target of 6-7 moves; nor does the "Fred"
variation 1 e4 f5?!? 2 exf5 Kf7?!? 3 Qh5+ g6 4 fxg6+ Kg7 5 gxh7 Rxh7
-- yes, that's a recognized (though very deprecated) line. Somebody
might have started a game 1 e4 Na6?! 2 Bxa6 bxa6 3 Nf3 Bb7 4 0-0!? Bxe4
5 Re1 Bxf3 6 Qxf3 Nc6, which looks reasonably plausible, though probably
not enough to be a canonical opening line.

What about complete games? There are plenty of games that were agreed
drawn after 0.5 or 1 moves and are thus trivial examples. There's the
possibly apocryphal Fingerfehler game 1 e4 e5 2 Qh5 Ke7??? 3 Qxe5#.
There might be a complete game 1 f4 e5 2 fxe5 d6 3 exd6 Bxd6 4 Nc3??
Qh4+ 0-1, though alas that's not quite uniquely reconstructible.


More information about the Retros mailing list