Continuing with 1921 Capablanca - Lasker, Game 11, I had a hard time figuring out what is going on in the diagrammed position. The notes of Capablanca and Kasparov were not particularly helpful on their own, but by considering them together and trying my own ideas, I finally grasped the thread of the game.
Of 32.Nd2, Capablanca wrote, 'This was my sealed move (*) and unquestionably the only move to keep the initiative.' It's clear that White is better. The Black Pawns are split into three islands, with the b- & d-Pawns vulnerable. The Black King is exposed to attack and White has one each of the four pieces, meaning there is lots of opportunity to rearrange the pieces into various attacking configurations. Black's own trumps are the Knight on d5 and the ownership of the c-file.
At this point there is a difference of opinion about how to continue. Black played 32...Nf8, and Capablanca noted, '32...Rc3 would have been met by Qa1', implying that Lasker's move was satisfactory. Kasparov assigned 32...Nf8 a '?!', quoted Lasker ('Better is 32...Rc3! 33.Qa1 Nf8 34.Ne4 Rc7 gaining an important tempo'), and added 'I agree: after 35.g3 White has a marked advantage, but even so not as great as in the game.' Q: What is this 'important tempo'?
Havana 1921 (g.11)
Lasker, Em.
Capablanca, J.R.
After 32.Nf3-d2
After 32...Nf8, the game continued 33.Ne4, and Capablanca commented,
The White Knight stands now in a very commanding position. Black's game is far more difficult than appears at first glance and I believe that the only good system of defense would have to be based on ...f5, after ...h6, driving back the White Knight. (Q: Why is it so important to 'drive back the White Knight'?)
After a further 33...Qd8, White played 34.h4. Kasparov gave this a '!, preventing ...f5'. Now on 34...Rc7, comes a flurry of remarks by the two World Champions.
Capablanca: 'This might be said to be the losing move. Black had to play ...h6 in order to be ready to continue with ...f5, forcing the White Knight to withdraw.'
Kasparov: 'Playing the Rook onto the 7th rank leads to conceding the c-file.'
Capablanca: 'There is much more than meets the eye in this position. This is a crucial point in the game. Apparently there is not much on either side, yet if Black can save the game it must be done at this point, and the chances are that the only move that may save the situation is 34...h6, threatening to drive the Knight away with ...f5. The situation is most interesting and will repay study.' (Primer of Chess, p.222)
(Q: Why was 34...Rc7 'the losing move'? What's this about 'conceding the c-file'? Why the obsession about 'driving the White Knight away'; what is it threatening?)
Then White played 35.Qb3 and Capablanca noted, 'White's plan consists in getting rid of Black's powerfully posted Knight at d5, which is the key to Black's defense.' Now it all clicked: White will play Bc4 and Nc3, forcing the Knight off d5; that's why control of the c-file was so important. It was also important to kick the Knight off e4 in order to prevent it from going to c3. Once the Black Knight is gone from d5, the e-Pawn becomes vulnerable.
Only three moves have been played on each side since the diagram, but what a lot of chess knowledge was packed into those six moves. Incredible! Now if I could just figure out what was meant by that 'important tempo'...
---
(*) I'm almost certain that adjournment analysis was frowned on in 1921, but I need to verify this. When did adjournment became fair game for the sort of extensive analysis that disappeared when computers became too strong?