I read the second chapter about open files and outposts today. There were some interesting points about the definitions. It seems that usually today we define an open file as one where the pawns of both sides are absent. On the other hand, when your pawn is absent but the opponents is present we call it semi-open. To Nimzowitsch the file is considered open when your own pawn is absent. Although he is not responding to our definition he has a justification for defining it this way. He argues that attacking a piece (including a pawn) and attacking a square are the same thing. I thought that was an interesting point and definitely gave me something to think about.
Going on from there he discusses means of creating open files by centralizing the pieces and causing a trade off. This was another point I hadn't really considered too much. I guess I always sort of treated open files as something that just came along, and centralizing my pieces as good strategy for the effectiveness rather than forcing them to be traded off. Of course, the advanced player is able to plan the creation of open files and this is something for me to consider in the future.
He also makes the point that the goal of operations on an open file is penetration to the seventh or eighth rank. I think this is something which I basically understood but that he put in a more pointed way that I think will be useful to remember.
Then he covers some ways one can attack a defended pawn on an open file and that an open file can be used as a means to mobilizing a heavy piece to another file. These were nice points but I think they weren't new or surprising to me.
Then he talked about outposts. This was another point where I think our customary definition today is different. I think the idea I had usually seen before like in Stean's "Simple Chess" is that an out post is a square protected by a pawn but where there is very little or no chance of an enemy pawn approaching and dislodging a knight placed on the outpost. The value of the outpost is entirely due to the excellence of the position of the knight and that it is firmly ensconced there.
Nimzowitsch sees things differently. To him an outpost is not of value in itself, it is of value as a means to creating play down the file precisely because pawns will be advanced to dislodge the knight which will weaken the pawn that is actually on the file. So for Nimzowitsch a crucial element of the outpost is that it is on an open file so that when the knight is moved the weakened pawn is exposed to attack.
This was a bit of a revelation to me and I think I will have to think about it. I don't think one is better than another, they are just two different situations. I think I have been more aware of the previous notion and not very aware of Nimzowitsch's concept here. I'm sure with a bit of thought the ideas can be fruitful though.
Another interesting rule of thumb that will bear more thought is the idea that whereas a central outpost should mostly be a knight, a flank outpost should be for a rook. There's not a lot of explanation of this point and I think I should devote a bit of thought to trying to understand the point.