Chess, May 2008

Harold Lommer

Numbers play an important part in everything, but in chess composition they don’t just serve to prefix the moves. It has become quite common to include, in the titles of printed anthologies of studies or problems, the number of compositions the anthology contains. There are three such study anthologies in English. The first, A Thousand End-Games, was published in two volumes in 1910 and 1911 and was produced by Creassey Edward C. Tattersall (1877-1957). The second was 1234 Modern End-Games Studies, compiled by M. A. Sutherland and Harold M. Lommer, while the third was 1357 End-Game Studies, compiled by Harold M. Lommer and published in 1975. To enthusiasts, these books are known by their numerical titles: 1000, 1234 and 1357. They are mandatory items for anybody looking to form their own endgame study library.

Tattersall’s work was groundbreaking, collecting together the best known work by the giants of the time, and classifying it by material used. In this book can be found classic studies composed by Kling, Horwitz, Troitzky and Rinck, amongst others. Tattersall himself was an occasional composer and some of his own work can be found in his collection. 1234 followed on in the same vein and recorded the work of a later era. I know nothing about Mr Sutherland, not even any studies by him, though the name of co-author Harold Lommer, owner of a night-club in Soho after World War II, will be familiar to many chess enthusiasts. In later years Lommer produced 1357, covering the years 1935-1973.

Lommer was greatly interested in studies showing multiple promotions and/or underpromotions and produced many examples during his lifetime, but for this column I have chosen two simpler pieces as examples of his output.

Harold Lommer

British Chess Magazine, 1946 (version)

1K1B4/3r4/7k/2p5/b7/1SP5/5P2/8

White to play and draw

Both White pieces are under attack and there is no way of saving either of them. Playing to gain the maximum from the inevitable piece-loss by 1.Nxc5? doesn't work: 1...Rxd8+ 2.Kc7 Ra8 3.Kb7 Ra5 4.Kb6 Rb5+ 5.Kc6 Rb4+ 6.Kd5 Rg4! 0-1 (6...Bb3+ 7.Kd6 Rb8 8.Kc7 Rb5 9.Kc6 Bc4! 0-1) In both lines Black is able to maintain his material advantage. Instead White has to actively sacrifice his bishop on a square of his own choosing. This is necessary for a couple of reasons that will become clear as we play through the solution. 1.Bg5+ Kxg5 2.Nxc5 Rd8+ 3.Kc7 Ra8 4.Kb7 Ra5 5.Kb6 Rb5+ 6.Kc6 Rb4+ 7.Kd5 Bb3+ If Black tries 7...Rg4 (or 7...Rh4) White has 8.f4+ (such check not being possible if the black king were still on h6) 8...Rxf4 (8...Kxf4 9.Nxa4 =) 9.Ne6+ =, and we have discovered the first reason for White's initial check. 8.Kd6 Rb8 Black has other possibilities, but they lead to repetitions or the same finale delayed. 8...Rb6+ 9.Kc7 Rb5 10.Kc6 Rb8 11.Kc7 =; 8...Rb5 9.Kc6 Bc4 10.Ne4+ Kg4 11.Nd6 = 9.Kc7 Rb5 10.Kc6 Bc4 11.Ne4+ Now we discover the other reason why it was important for the black king to be decoyed to g5 on the first move: 10 moves later the white knight needs to gain a tempo (by checking the black king) on its trip from c5 to d6 for the fork. 11...Kf4 12.Nd6 ½–½

Our study for solving is taken from the New Statesman, a journal that did much in the middle years of the last century to encourage endgame studies in Great Britain. Its then chess columnist, Heinrich Fraenkel ('Assiac'), was the one who did the encouraging.

Harold Lommer

New Statesman, 1967

3s4/8/4p3/2Q5/5k1q/6p1/4KS2/8

White to play and win


Developed and maintained by Brian Stephenson.
Implemented with HTML5, MySQL, Perl (with, inter alia, CGI::Simple, HTML::Template & XML::LibXML) & CSS/Javascript (jQuery, Bootstrap & DataTables).