 |
 |
Editor: Dr Tim Harding |
 |
 |
 |
 |
 |
 |
|
|
Editor: Dr Tim Harding |
|
|
|
|
|
|
© Dr Tim Harding
Last modified:
6 October 2026
As we explained in the opening article of this series, the international tournament at Hastings in the summer of 1895 was a great event in which almost all the leading masters of the world at that time were participants. The energetic local chess club in Hastings, led by tournament secretary Herbert Dobell, ran the event very well in many respects.
One of the features of the event was a beautiful tournament book which was planned from the start and produced very quickly. It was edited by a Hastings club member, Horace Cheshire, and published by Chatto and Windus of London early in 1896, just a few months after the congress concluded. Cheshire’s book included signed black-and-white studio portraits of all the players, commissioned from local photographer Bradshaw. (Original prints of these photographs can be seen at the National Archives in London.)
In that book, the game scores and annotations, of course, were presented in the descriptive notation used throughout the English-speaking world at that time and up to the 1960s at least. At that time the French also used descriptive notation, as can be seen in Samuel Rosenthal’s book of the Paris 1900 tournament but most of the rest of the world preferred the algebraic notation which is universally used today. We suspect that Mr Cheshire, an amateur with limited editorial experience, was not very fluent in algebraic notation.
Unsurprisingly, since most Europeans were not familiar with descriptive notation, there was a need for a tournament book in algebraic, which was commissioned by Veit & Co., the Leipzig publishers of the Deutsche Schachzeitung, and edited by the Berlin chess master Emil Schallopp, who had edited the (incomplete) book of the great Paris 1878 tournament for the same publisher. Schallopp was a scholarly man who worked as the chief stenographer for the Reichstag, the German federal parliament.
In his introduction, Schallopp lists many of the sources that he used to compile his book, including newspapers and magazines from Britain, Germany and other countries. He writes that before the official book appeared he already had found the scores of about one third of the games.
He then explains that Mr Cheshire, the editor of the English tournament book, “only in a few places makes his own comments on the text”.
Instead, as Schallopp observed, the annotations in the official book were written by 12 of the competitors, whom he lists: Pillsbury, Lasker, Tarrasch, Steinitz, Schiffers, von Bardeleben, Blackburne, Gunsberg, Tinsley, Mason and Albin. (The other ten participants did not contribute notes, it seems.)
Schallopp evidently found (as we do) Cheshire's methodology to be curious, if not unsound, remarking that the games were distributed for editing, “but in such a way that none of them got to annotate a game they had played themselves.”
While some of annotators were possibly more diligent than others, the editor's strange procedure was always liable to have the critical moments of games go unnoticed by someone who did not play it. Even more seriously, we do not know what checks (if any) Cheshire made to reconcile with the manuscripts he received back from the annotators against the original scoresheets which were presumably provided to him at the end of each round.
While I was researching my book Steinitz in London, Dr Thomas Niessen sent me a file generated by his software (which compares PGN files) that had identified 36 games from Hastings 1895 revealing discrepancies between the game scores in the Cheshire and Schallopp books. I since raised queries on five other games from that tournament. You can download Thomas’s free software for yourself from his website.
Some discrepancies between the tournament books are minor (concerning move order or an additional move at the end) but several are complex. What is more, the books frequently differ from contemporary newspaper reports on the games and chess columns that covered the tournament soon afterwards.
Sometimes the German score looks more likely (or even certainly) to be correct, and sometimes the Cheshire version. Occasionally, neither is completely to be trusted.
In a major tournament today, both players have to submit their scoresheets, providing a cross-check, and usually they have a carbon copy to retain. In the Hastings regulations, only the winner (or White in the case of a draw) was required to submit a legible score. The loser (or Black in a drawn game) probably retained his own scoresheet or threw it away. It is unknown whether journalists present were able to copy the official record or had to make their own copies, or perhaps ask the players for a copy.
Historians always prefer to go back to the primary sources (texts created at the time of the event) to be sure of having their facts right, but Cheshire’s dubious methodology means that one can by no means trust his version unquestionably, and in some cases at least his edition was clearly wrong. In particular, when one of the participants in a game published a game score differing from Cheshire’s, we will always tend to prefer theirs.
We cannot be certain how those publications obtained their game scores but Leopold Hoffer (who reported for The Standard, The Field and The Chess-Monthly) was normally very reliable and was writing very soon after a game had been played. We can be fairly certain that several others had reliable sources of information. Tinsley was in a good position to ensure the accuracy of the reports in The Times though he possibly delegate the reports to his son Edward who was also a strong player.
The youngest player in the tournament, Carl Schlechter, published 56 of the games in his weekly column in the Vienna sports paper Allgemeine Sportzeitung. You might expect that he would publish accurate scores but for reasons of space in the publication, he may have truncated some games, and that indeed seems to be the case with his first round game against Pollock, as we shall see in our discussion of Round One.
Some time ago we wrote to the secretary of the Hastings Chess Club to enquire whether perhaps any of the original game scores still survive in the club’s archive but we were informed that they do not have them. Probably they were lost or destroyed after the first book was published. There must be a suspicion that the official scoresheets were given to the annotators and not received back with their annotations for checking.
There is at least one case where the finish of a game in the Cheshire edition is blatantly wrong. The game Lasker-Steinitz is one such blatant case about which I wrote in my Steinitz book. Cheshire’s version is nonsense whereas Schallopp agrees with The Times. There are also cases where the opening sequence or a move in the middle of a game differs between the two books.
To confuse matters further, two new tournament books were published in 1995, the centenary of the tournament, both using algebraic notation. One was produced in Dallas, Texas, by Sid Pickard while the late Colin Crouch and Kean Haines did some additional research into the tournament and published a book in England. This was a big missed opportunity to set the record straight.
Sometimes these modern editors follow Cheshire, sometimes they follow Schallopp and occasionally they give something different. Sadly neither of these books provide any discussion of the issues we are discussing here, or indeed show awareness of the problem. Probably the editors just worked from the databases available to them. In the case of the Lasker-Steinitz game, Crouch and his co-author clearly saw there was something wrong in the Cheshire book but did not check Schallopp and came up with a third “solution” for which there is no primary source.
There were several other major tournaments of the pre-First World War period where the record is incomplete because contemporary tournament books were not produced or were missing some of the games. Paris 1878, Wiesbaden 1880, Cologne 1898, Monte Carlo 1902 and Ostend 1906 are the best known. Hastings 1895, though, is the only example we are aware of where supposedly all the games were preserved but where in fact the record is so unreliable.
In rounds 4, 5 and 16 no problems, happily, were found in any of the games. All the other 18 rounds have a problem with at least one game.
Each discrepant game must be analysed on a case-by-case basis and that is what we shall be doing, round by round, in our subsequent articles.
Each article will discuss only the games where we see problems, but when the series is complete we shall provide a PGN file for the whole tournament which is as accurate as we can make it at this time.
This may not be our final word because there are still some further sources that we would like to check against our provisional results.
Readers are also welcome to send in their own discoveries and comments.
The recent problems at the British Library have delayed our work on this topic but we do hope eventually to see some newspapers that are currently unavailable to us, and some other magaiznes of the period at the Koninklijke Bibliotheek.
We are considering eventually writing a "variorum edition" tournament book that will spell out all the differences with our conclusions.
Back to the introduction page.