Friday, 30 April 2010

Review: Gary Danelishen's The Final Theory of Chess

Gary M. Danelishen, The Final Theory of Chess. Phillidor Press 2008, Berea OH.

A week ago, I received the first chess book that has been sent to me for a review in this blog. I must confess it was a surprise for me to get a book to be reviewed, as my original idea was just to write about chess books I had already bought. But I certainly will review any book sent to be reviewed by any chess publisher!

Gary M. Danelishen has taken a huge task: he aims at a computer-generated analysis that should at some day be able to solve the game of puzzle. Even if the game of chess is extremely complicated, it is a finite game and thus it is solvable at least in theory. In practice it is of course impossible for humans to solve the game of chess. On the other hand, computers have already been able to solve many simpler games, for example the checkers.

The Final Theory of Chess should of course be seen only as an attempt to lay a foundation for the theoretical solution of chess. Most of the book contains analysis conducted between 2004 and 2008 using a network of six computers running different versions of the famous Fritz software. As both the hardware and software evolve continuously, it is more than probable that the latest 64-bit computers with terabytes of memory and either Rybka or FireBird could correct the analysis here and there. But as I don't own neither a modern multi-CPU computer or the latest chess software, I haven not tried to refute any of the published analysis.

Both the idea of trying to progress towards the solution of the chess and the heavy use of our dear friend Mr. Fritz has lead the book to concentrate on concrete variations. The published variations have been color coded to give the reader an idea about the robustness of the analysis. Unfortunately the book is printed in black and white which makes it at times rather difficult to decipher the color coding. Finally each published variation gives an evaluation of the variation as given by Fritz.

In addition, the variations are practically devoid of any verbal explanations. Thus the reader will not find here any help about strategic nuances of the opening. This makes the book very inpractical for anyone interested in just playing chess on the board during the club evening of in an weekend tournament. The book is, however, quite useful as an guide when analysing the games afterwards. This is especially true for those chess enthusiasts who don't have a computer to help in analysing the game. In addition, this book could be extremely useful for correspondence chess players who are looking for new ideas in their tactical pet openings.

The search for the solution of chess has lead the author to concentrate on forcing tactical openings. This explains the about 110 pages dedicated to Blackmar-Diemer-Gambit, an opening of debatable correctness. In addition to BDG, the book presents computer checked analysis in many other tactical off-beat variations. If you are interested in them, you might well like to order this book from Amazon.

In case you are not sure, whether this book is for you or not, you should take a look at the web site The Final Theory of Chess. You might even decide to contribute yourself to the work in progress!


  1. well nice idea but what if black plays 1.. Nf6?
    then after 2.Nc3 d5 3.e4, trying to get the BDG
    again, black plays Nxe4!, the Huebsch gambit
    which is much better for black.
    so exit 'final theory of chess..
    Or a restart ?
    But then more of the chess community should
    be involved in the wiki, and they need
    a (big) database, eg of chess assistant
    or bookbuilder ,

  2. Actually Mr Danelishen suggests 2. f3 as the way to transpose into BDG. It is certainly obscure enough for a practical player but I seriously doubt if 2. f3 is the starting move of a sequence leading into forced win.