21 September 2008

Pillsbury’s Mate

Errors abound. An error made and published will reproduce itself. While assembling my "Checklist of Checkmates," a still unpublished booklet, I presented this position from Laschet-Schaffer, 1997 as an example of Pillsbury's Mate.

I asserted, "Harry Nelson Pillsbury won several games this way in the nineteenth century." Now, I cannot find any games that support that assertion. I must dispute the accuracy of my claim (and revise it before publication).

It is possible that I was thinking of the game Pillsbury-Lee, London 1899 as given in Georges Renaud and Victor Kahn, The Art of the Checkmate (New York: Dover, 1962 [1953]).

Pillsbury, H - Lee, F [D53]
International Chess Congress London, 1899
1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.Bg5 Nbd7 5.e3 Be7 6.Nf3 b6 7.cxd5 exd5 8.Bb5 Bb7 9.Ne5 0-0 10.Bc6 Rb8 11.Bxb7 Rxb7 12.Nc6 Qe8 13.Nxe7+ Qxe7 14.Nxd5 Qe4 15.Nxf6+ gxf6 16.Bh6 Qxg2 17.Qf3 [17.Qf3 Qxf3 18.Rg1+ Kh8 19.Bg7+ Kg8 20.Bxf6+ Qg4 21.Rxg4#] 1-0

Renaud and Kahn state, "Black resigned on his seventeenth move, since a refusal of the sacrifice would have cost him a rook" (131).

Position after White's 17.Qf3

The diagram position and conclusion certainly accord with my assertion that Pillsbury won at least one game with Pillsbury's Mate. Unfortunately, the game score differs in The Book of the London International Chess Congress 1899 (London: Longmans Green and Co., 1900), the official tournament book. There are some differences in move order, an absence of the move 10...Rb8, and 15...Rfd8 instead of 16...Qxg2. After that point the games are completely different.

Pillsbury, H - Lee, F [D51]
International Chess Congress London, 1899
1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.Bg5 Nbd7 5.Nf3 Be7 6.e3 b6 7.cxd5 exd5 8.Ne5 Bb7 9.Bb5 0-0 10.Bc6 Bxc6 11.Nxc6 Qe8 12.Nxe7+ Qxe7 13.Nxd5 Qe4 14.Nxf6+ gxf6 15.Bh6 Rfd8 16.0-0 Kh8 17.Qh5 Rg8 18.f3 Qe6 19.d5 Qe7 20.Bf4 Rac8 21.Rac1 Nc5 22.Rfd1 Qd7 23.b4 Nb7 24.Rc6 Rg6 25.Rxf6 Rcg8 26.Be5 1-0

The game score in Renaud and Kahn must be inaccurate, unless the tournament book put forth egregious errors.

A Similar Game

The position below with black to move appeared in Barassi-Ferrari, Trieste 1923. Aside from the position of Black's queen rook, it is identical to that presented by Renaud and Kahn as the final position in Pillsbury-Lee. It bears an even stronger resemblance to Pillsbury-Lee as presented in the tournament book until move 15.

That game did not conclude with Pillsbury's Mate, although it easily could have. Here's the game score as I found it in the ChessBase database.

Barassi, A - Ferrari, R [D53]
Trieste-chB+ Trieste, 1923
1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.Bg5 Be7 5.e3 Nbd7 6.Nf3 b6 7.cxd5 exd5 8.Ne5 Bb7 9.Bb5 0-0 10.Bc6 Bxc6 11.Nxc6 Qe8 12.Nxe7+ Qxe7 13.Nxd5 Qe4 14.Nxf6+ gxf6 15.Bh6 Qxg2 16.Qf3 {diagram position} 16...Qg6 17.Bxf8 Rxf8 18.0-0-0 Kh8 19.Rhg1 Qh6 20.Rg3 Rd8 21.Rdg1 f5 22.Qxf5 Qc6+ 23.Kb1 Nf6 24.Rg7 Rf8 25.f3 Qe6 26.Qg5 Qd6 27.Qh6 c5 28.Rxh7+ 1-0

Could it be that Renaud and Kahn mixed up game scores from two games to create one that never existed?

Copying the Error

In How to Beat Your Dad at Chess (1998), Murray Chandler cites Pillsbury's "16-move win over Lee in London 1899" as "the most amazing" example of Pillsbury's Mate (50). On the next page, the diagrams given and alleged to be from Pillsbury's win, show a black rook on a8 as in the Barassi-Ferrari game. Stating that the game lasted sixteen moves, rather than seventeen, also suggests closer correspondence with the later game.

Eric Schiller's presentation of the game in Encyclopedia of Chess Wisdom (2003 [1998]) almost matches that in Renaud and Kahn. But Schiller moves the game backward a decade to 1889.

On his Chess History Center website, Edward Winter includes Schiller's error in a catalogue of faults appended to an article that originally appeared in Inside Chess.

Nonsensical game-score: Pages 321-322 have a game 'Pillsbury v Lee, London 1889'. The two did not even meet that year. Ten years later they played a game which opened similarly, but the continuation given by Schiller was in fact what occurred, up to a point, in a different game, Pillsbury v Newman, Philadelphia, 1900. In short, yet another shambles.
Winter, "A Sorry Case,"
Schiller's limited annotations to the game offer the same alternative for Black at move 10 that appears in Renaud and Kahn. Perhaps, it is not entirely a case of plagiarism. Schiller offers an insight at move 16 that is not in The Art of the Checkmate. "It is unlikely that a modern player would be foolish enough to capture at g2, opening up the dangerous g-file, even if the combination itself was not anticipated" (Schiller, Encyclopedia of Chess Wisdom, 3rd edition, 323).

The Pillsbury-Newman game cited by Winter deviates on move 17, where Pillsbury played 17.Kd2 instead of Qf3. At least that's how the game score appears in Rhoda A. Bowles, "Mr. H. N. Pillsbury's Chess Career" British Chess Magazine (August 1902), 343.


  1. Interesting website. Have you thought about submitting some of your writings to NORTHWEST CHESS for possible publication

    Russell Miller


  2. Russell,

    The thought runs through my mind from time to time, but it usually escapes before I can tie it down. I ry harder to catch it the next time.

    I have submitted annotated games to Northwest Chess, such as my games against FM David Sprenkle in the Spokane City Championship that appeared in the August 2008 issue.

  3. I've never heard of "Pillsbury mate" rather, "The Pillsbury Attack" was a valid approach Pillsbury prospered with the white pieces and prompted the popularity of QGD in the late 1890's and turn of last century. This attack was always marked with a Ne5, attacking the defending knight on f6 ( usually with teh bishop or if teh NE5 is exchanged a pawn on e5 drives it away) adn followed up with a queen to f3- h3 and Bishops focusing on h7.

    The Game with Lee has the characteristic Ne5 ( on move 8). But, you might want to check with Batgirl on Chess.com as she's a better historian than I as she lists these mates quite elequently here:

    Morphy's mate I heard of and it appears that the "Pillsbury mate" is a Bishop variant of a similar pattern. Hmmm, you've got me curious. I'll have to do some research with chess base

  4. Thanks BlunderProne,

    I checked your link to chess.com, where the user has posted several of the best known checkmate patterns, including Pillsbury's Mate. With a few exceptions, however, she offers no history of any of the patterns.

    Dozens of books and many more websites offer lists and diagrams showing basic checkmate patterns, but histories of the names are more elusive. Although several patterns are present in Pedro Damiano's Questo libro e da imparare giocare a scachi et de li partiti (1512), most of the names used today appear to be a twentieth century contribution. The smother mate pattern given in your link is from Damiano, but some of the pieces have been removed from his example.

    The Art of the Checkmate (1953) might not be the original source of the term Pillsbury's Mate, but it is the oldest source that I've located. I long for correction on this point.