Another ongoing discussion, this time concerning central pawn expansion in the QGD

Recently Paul Leggett and I have been discussing a plan created by Botvinnik in the QGD, in which as White he forgoes the minority attack of yore (even back then!) and instead plays for central expansion. The key line runs 1 d4 d5 2 c4 e6 3 Nc3 Nf6 4 cxd5 exd5 5 Bg5 Be7 6 e3 0-0 7 Bd3 Nbd7 8 Qc2 Re8 and now 9 Nge2 with the plan of castling, tucking away the king, and then an eventual f2-f3 and e3-e4, gaining much space in the center and kingside. This plan was first seen, I believe, in Botvinnik-Keres, USSR Ch (Moscow) 1952. (The move order given above is a transposition.)

Prior to that, 9 Nge2 had occurred in this position nine times. According to Kasparov's MGP series, Botvinnik had the following to say after the additional two-ply of 9...Nf8 10 0-0:

This continuation was unusual for that time. After Ng1-f3 White normally used to castle on the kingside, whereas when he played Ng1-e2 he would castle queenside. But in the present game a 'hybrid' variation has been employed, with the aim of making it difficult for the opponent to choose a plan. After White's kindside castling it is harder for Black to obtain active play, in which Keres always felt confident. - M. Botvinnik

(That idea of Nge2, castling queenside and then attacking the kingside with pawns still had some poison in it at the top levels into the early 1980s, by the by.)

Botvinnik-Keres, USSR Ch (Moscow) 1952

after 9 Nge2

Paul brought this up within the context of a recent game of his, which he posted on Facebook*. Later in the discussion, Paul made the following comment:

I am almost finished reading "Rubinstein Move by Move" by GM Zenon Franco (an excellent book), and what do I find towards the end of the book? Botvinnik-Keres 1952 and Kasparov-Andersson 1988, annotated in the context of Rubinstein's contribution to the QGD. These games will live forever.

To which I replied that I had just noticed the game Rubinstein-Bogoljubov, London 1922, which seemed to be heading in the direction of the Botvinnik game above, but after playing Nge2, Rubinstein then played e3-e4 followed by f2-f4. It seemed that Rubinstein was going for a direct kingside attack. (I have no idea if this game is mentioned Franco's book. I'm sure Paul can give an answer to that one.)

Alas, my abilities to visualize the action were sadly lacking. On vacation a few days ago I managed to finally get to the game, which goes as follows: 1 d4 d5 2 c4 c6 3 e3 Nf6 4 Nc3 Bf5 5 cxd5 Nxd5 (a key difference in pawn structures!) 6 Bc4 e6 7 Nge2 (our star move) Nd7 8 0-0 Qh4, and now Rubinstein committed a tactical oversight and played 9 e4, which simply lost a pawn. 

But in the tournament book, Alekhine made the following comment about this position:

Rubinstein-Bogoljubov, London 1922

after 8 ... Qh4

An inoffensive reply to which White should simply play 9 f3, and if 9...Bd6 10 g3 and Black is forced to withdraw his queen to e7, for if 10...Qh3 11 e4 Nxc3 12 bxc3 Bg6 13 e5 followed by 14 Nf4 and wins. - A. Alekhine

Clearly Alekhine was looking to central expansion here, and I imagine that he believed that was Rubinstein's original intent. But that knight on d5 induced a tactical error!

It's hard to believe that Botvinnik wasn't aware of both the game and Alekhine's commentary on the game, and perhaps it inspired him. Probably not in 1952, but rather in 1938, against another one of the giants of the game. For the famous Botvinnik-Capablanca game at the renowned AVRO tournament featured Botvinnik playing 9 Nge2 with an eventual 0-0, f2-f3 and e3-e4 against Black's pawns on the a-d files and an absent Black e-pawn.

I rather imagine, sans any evidence, that this was all a chain of inspiration over time, one thing leading naturally to another. It makes for a nice story, but I have no idea how much if any validity it merits. I do know that in his book One Hundred Selected Games Botvinnik mentions as early as Black's sixth move in the Capablanca game that White's basic plan is to play f2-f3 and e3-e4, breaking through in the center. He goes on to mention after 9 Nge2 that the game recalls Lilienthal-Ragozin, Moscow 1935, but with a couple of minor differences. Clearly Botvinnik had given this strategical idea some thought! 

Those with better paper archives might find more on the genesis of the Nge2, 0-0, f2-f3, e3-e4 idea in these structures, and of Botvinnik's inspiration. (I only have three books by Botvinnik, the one mentioned which only covers games up until 1946, Championship Chess on the Absolute Soviet Championship of 1941, and a book that he sorta kinda maybe co-authored on the Grunfeld Defense in the 1970s, so I have nothing by him on his games against Keres or Larsen save what Kasparov quotes.) And those with better db skills may well turn up some stones plowing fields of data.

So for now, I leave it at that.

...

But I do have two additional comments of a tangential nature. First, the Rubinstein-Bogoljubov game, and Alekhine's notes to it, have more interesting features, which I hope to return to soon. Second, can you imagine one of the top players now writing a tournament book? Imagine Carlsen or Giri writing notes to the recent BilBao tournament, for example, which was also a six player double round robin. Come on, guys, it's only 30 games, and I'm sure your seconds will kick in a good deal of the work. Let's make tournament books great again! 

* I hope that's the correct link. Facebook is a real pain in the posterior to deal with on such matters.