Gamesmanship, Legitimate Play, or a Tall Tale? Or, why it is Harder to Write History than one Might Think.

I am writing a relatively long post to show the way chess historians must be careful not to buy anything they are told -- but also not to dismiss everything they are told, either. While many laymen believe everything people recall about the past due to exaggerated respect to eyewitness testimony, amateur historians tend to fall into the opposite extreme: to dismiss all accounts that have minor variations in them as "inventions", and rely only on "published documents", forgetting that psychological factors make recall fuzzy as time goes on, which hardly means the eyewitness is lying, and that documents can be just as wrong as eyewitness statements.

The worst cases of this are seen in holocaust denial, where amateur historians (David Irving most famously) ignore the context and the times, relying only on published documents, and thus take at face value the Nazi's talk of "evacuations" and "fighting partisans", and claim minor discrepancies in eyewitness testimony as "proof" they are all lying. The reality, as historian Michael Shermer notes, this is getting it backwards: the Nazis deliberately used obfuscations in their official documents, while it would be extremely surprising if all eyewitnesses, each with their own point of view, would have told exactly the same story.

Something similar, if on a much more benign level, happened to me, when trying to establish the truth about an historical incident. I was told yesterday by a veteran Israeli player, IM A, that another Israeli player, IM B, had used the following trick:

In a lost position, with both players in desperate time pressure, IM B had deliberately hanged his queen! He slammed down his undefended queen on d4, forking the opponent's queen on a7 and the king on g7. His opponent, famous American GM C, was shocked, and started thinking furiously. He might indeed have lost on time, as IM B intended, if it weren't for IM B's unfortunate mistake: he slammed the clock so strongly his own flag fell! After the game, GM C and his colleagues told me, angrily: "this is just not done!"

The problem? I checked the database and no game between IM B and IM C which fits the description took place. What's more, IM B apparently never reached such a position in any formal game. Incidentally, the position itself occurred quite rarely, even with the queen defended: only about 200 times in nearly 3 million games in my database.

It is easy to now dismiss the story as an invention, a tall tale. But this would be hasty. There was indeed a game between IM B and GM C in the said time and place which IM A had noted. What's more, the game ends, two moves before the time control, in the following position:

In this position, after Black's move (... h5) the database says Black won (0-1). Indeed he is about to mate White. But, if here White played Qd4!? in an effort to make Black think a few seconds and lose on time, it would fit well with IM A's story. It is not surprising that if this is so, this last "tactical shot" was not noted in the database.

But this is not the end of the story. Does this mean IM B actually tried to "pull a fast one"? Possibly: objectively, Qh1, Qe2 or even Qxg4 are better (although they still lead to a dead-lost position) but would not have such a psychological effect on the opponent. So an attempt at a "swindle" seems possible. On the other hand, he might have been so short on time he might have blundered in earnest: after all, even great players sometimes commit oversights.

If I knew how hard it would be to get to the truth in chess history, I might never have taken up this hobby...