Last month I mentioned briefly in the game notes a doctrine of a chessfriend of mine: chess by smell. I find this doctrine attractive, and I want to consider it more fully now. The main idea of chess by smell is that, when confronted with a truly difficult choice, the player should focus on the fundamental features of the position and give strongest consideration to the move that most accords with them. In general, to employ a metaphor that will be meaningful to most chess players, he should play the move that smells best.

I don't want to try to confer upon this mere practical rule the status of a deep chess truth. Nevertheless, the doctrine of chess by smell has been particularly useful to me in my correspondence play. There is in the correspondence game, I think, a tendency to elevate specific calculations, and also specific theoretical pathways, to too high a place in one's reckonings. When specific calculations reveal the definite superiority of some continuation, well and good. But when that isn't the case and the decision is a truly difficult one, it is time to close the books, shut down the database, fold up the analytical notes, and take stock of some very simple things. At those times -- and they occur not only when the position is calm but often also when it is tactically charged -- I ask myself such things as where all the pieces are, how active they are and how their activity can be improved, what the weaknesses are, what the various plans are, what could possibly happen and what could not possibly happen, and what sorts of endings might be coming up. What, in short, are the demands of the position?

You may easily answer, well, that is all too easy; I do all that already. But I know that when I analyze my own defeats I can generally identify some point in the game at which I became hungry for something a little too specific and which turned out later to be rancid. At some point I did not stop to smell the position and sniff out the most satisfactory approach to it. And when I win, I think I can often see the same thing in the play of my opponents.

To see what I mean, take a look at this position, which occurred, surprisingly enough, in two of my games in US13P05, in each of which I was White. White has just played 13. Bd2.

White has the basis of a strategic victory in his two bishops and his imposing queenside pawn majority, and he stands to win a pawn on c3. But he lags in development and his king's field is somewhat weakened. There is also the thought that Black's knight on c3, if not eliminated, will function as destructively as a battalion of Greeks within the walls of Troy. No doubt is was the glittering prospect of maintaining the c3 knight while also exploiting the e1-h4 diagonal that led my opponents into their respective choices, 13...Nfd5?! and 13...Nfe4?!. Each of these moves has ingenious points, and I have no doubt that a great deal of calculation went into each of them. But as will be seen in the games given below, each is fatally flawed.

I couldn't denigrate the play of my opponents even if I wanted to, since I myself, when I was preparing this line, didn't credit Black's best move very much, and I spent lots of time analyzing the moves my opponents eventually played. But if you look at the position, I think that chess by smell suggests Black's best move rather clearly. And for what it is worth, my chessfriend, the doctrine's originator, was confident of the best move almost instantly. It's just 13...O-O! This could be a woodpusher's move, but a little sniffing shows that it is also a master's. After White wins his pawn on c3, he will have only his queen developed, Black's development will be almost complete, and it will be Black to move. Indeed, after 14. Bxc3 dxc3 15. Qxc3 Black has 15...Nd5! with further gain of time. Moreover, White's elimination of the c3 knight obliges him to open the d-file, and of course, open lines are invariably happy for the player with greater piece activity. Finally, with so little support from his pieces, White's queenside pawns will be exposed to attack by means of ...a7-a5. So 13...O-O! smells quite good, and as you see, there are no deep calculations here. You can find the details in the notes to the games.

Well, if you saw 13...O-O! and were sure of it as quickly as my friend was, go to the head of the class. But something is wrong when three strong correspondence players (each obtained a high score in the section) were all hypnotized by the prospects of the deep complications that are offered by funny moves in the diagrammed position. I think it goes to prove my point that there is too much emphasis on specific calculations in correspondence chess and not enough chess by smell.

Game 1. Morss-Dolgitser, US13P05.

Game 2. Morss-Dyson, US13P05.

Next month: the main line of the Two Knights.