For the previous article, please see: A legendary clash. Readers’ reactions confirm that even today’s top engines cannot provide a complete answer on their own, despite being indispensable. In his latest analysis from 2020, Garry Kasparov improved significantly on previous works, especially his own from 2003. However, some additions and corrections are still required. On the other hand, alongside some excellent strategic comments, also a few of the lines given by William Steinitz already in 1896 are still relevant today.
Sources and abbreviations used (in chronological order):
Hof = La Stratégie (15. August 1895), p. 239, notes by L. Hoffer
DWZ = Deutsches Wochenschach (18. August 1895), p. 279
DSZ = Deutsche Schachzeitung (August 1895), p. 237
BCM = British Chess Magazine (September 1895), p. 397, notes by W. Wayte
Stei = Horace F. Cheshire: “The Hastings Chess Tournament” (London 1896), p. 30, notes by W. Steinitz
Sch = Emil Schalopp: “Das internationale Schachturnier zu Hastings” (Leipzig 1896), p. 71
Kob = Alexander Koblenz: “Lehrbuch der Schachstrategie 1” (Berlin 1980), p. 224
Sergei Soloviov: “Emanuel Lasker 1, Games 1889-1903” (Sofia 1998), p. 181
Garry Kasparov: “My Great Predecessors, Part 1” (London 2003), p. 102
KM = Linder, I. & Linder, V.: “Emanuel Lasker” (Milford 2010), p. 71, notes by K. Müller
GK = Garry Kasparov: “Moi velikie predshestvenniki, Tom 1” (2nd ed., Moskva 2020), p. 118
The following answers were created with the help of readers’ responses. In detail they present the latest results on the five original questions, as well as to some extra positions. A few historical perspectives are also included. Special thanks go to ‘albitex’, a longtime contributor to this column, see for example Riddle solved: Fischer could have won!
Answer 1
With 55.Rxd3, Lasker ran into a mating net.
Kasparov is right in his statement. Following 55.Bc7! Ra2+ 56.Kf1 Rgg2 57.Rxd3 Rxh2 58.Kg1, Black wins in the long run. In the mainline, after 58…Rhe2, the total control of the 2nd rank enables him to exchange rooks at will. In the end, Black’s two outside passed pawns on the a- and h-file are the decisive factor, as confirmed by readers’ analyses. However, Steinitz is not far off. In the sense, that his defence 55.Bc7 is best and offers at least some hope in practice.
Perhaps a historical interesting point is that Steinitz’s comment from 1896 was completely forgotten for more than 100 years. In 2003, GK only considered 55.Bg5 Rxg5 56.Rxg5 d2 -+. The first one to (re-)discover 55.Bc7 was Karsten in 2010.
Thus, after 52…Ra1 53.Bd8, the evaluation of Chigorin’s 53…Nd3+ has changed significantly over the years. In 1998, Soloviov claimed it to be excellent (!!), while in 2003, GK still rated it as a good move (!). Since 2010, it has rightly been considered dubious (?!), and should be replaced by 53…Rh6 (KM and GK), or by 53…a6, as noted by ‘albitex’.
To 52…Ra1, Steinitz said: “Hardly as reliable as 52…Rh6“. No one picked up on his words. Today’s top engines rate his old recommendation even a bit higher than Chigorin’s move.
Answer 2
By 48.Rb5 Lasker tried to get away with the loss of a pawn. Following 48…Rxa4 49.dxe5 Nfxe5, instead of 50.Bh4, the Bulgarian team led by Soloviov recommended 50.Rdd5 immediately. As readers’ analyses confirmed, this does indeed offer more resistance, but after 50…Ra1 not salvation.
The commentary of DSZ reveals the negative side of analyses at that time, namely occasionally rather crude tactical oversights. After 48.Rxc4 Nxd4+, White wins by 49.Rdxd4, not Black. Curiously, Steinitz fell for the same trick. The correct reply 48.Rxc4 Nd6 was already noted by BCM and Chigorin. Also, on 48.Bc2 exd4, the old 49.Bxd4 is simply bad.
Both of GK’s lines, involving investing an exchange for a pawn and “rely on the strength of the bishops”, are state of the art.
The second one, 48.Bc2!? exd4 49.Rxc4 Nd6 50.Rxc6 Rxc6 51.cxd4, is more risky, as White needs to free his passive rook on d2.
The first one, 48.Rxc4!? Nd6 49.Rxc6 Rxc6 50.dxe5 Rxe5 51.Ba2, is a bit more comfortable. However, here White must focus all his efforts on the activity of his pieces.
Ultimately, both ideas prove to be just good enough to hold White’s position. Taken together, with 48.Rb5 White misses a possible save of a precarious situation. It has to be classified as a decisive mistake (?), costing half a point.
Answer 3
As just seen, after 47.Rd2 Nc6 Lasker found himself clearly on the defensive.
In this very complex position, White has several reasonable options, and the game could develop in a variety of different ways.
By 47.dxe5 Nc6 48.e6 (Vasyukov and Nikitin), the play becomes tactical, but Black can equalise quickly.
After 47.Ba2 Nc6! 48.Rxc4 Nd6 (Romanovsky), White maintains a slight advantage, but nothing significant.
Chigorin’s 47.d5 is the most difficult to evaluate. White enjoys a big advantage, but the position is closed and progress is slow. In the end, a win for White seems more likely than not.
A refinement of this idea is 47.Bc2 Nc6 48.d5. White probably wins, but not in a particularly convincing way, leaving some room for doubt.
Finally, there is GK’s 47.Bc2 Nc6 48.Rbb1. This proves to be the best and wins in the long run. White keeps the tension in the centre and waits for a better opportunity to play d4-d5. Black has various defensive systems, but his tactical attempts to break out are refuted. Otherwise, the mainline resulting from a joint conversation with ‘albitex’ demonstrates that Black is gradually being pushed back, involving a king’s march from e2 to a3.
Overall, 47.Rd2 turns a narrowly won position into a worse one. Thus, GK’s classification as “a rude blunder” is justified. It is a decisive mistake (?) as well, costing half a point.
Classical commentators viewed the position from a different perspective. Besides the statements of DSZ and Steinitz (cited in the presentation of the game last time), Chigorin noted (according to Koblenz): “Only with 47.d5 could White save the pawn. This could be followed by Nf7-d6 or Nd8-b7-d6 to execute b6-b5.” The BCM statement is representative of several others: 47.Rd2 is “the losing move. He had nothing better than to return 47.Bc2; Black’s Nd8-c6 being pretty obvious. He must now lose the a-pawn.”
While it is true that all of the recommendations at the time offered improvements, their aim was merely to maintain the balance. Thereby, they seriously underestimated White’s potential. Only in 1998, Soloviov was the first to suggest that White could maintain a clear advantage (±) by 47.Bc2 Nc6 48.Rbb1.
Answer 4
With 37…Nf6 38.Bxf4 Nh5 39.Be3 f4, Chigorin tried to regain control of f4 by his knight. But that doesn’t work as desired, and Lasker could keep a winning advantage until move 47.
The first candidate for improvement is 37…Ne7. It is a better realisation of Chigorin’s idea, because the knight aims for g6, instead of the somewhat decentralised h5. But ultimately, it doesn’t succeed either.
The interesting attempt 37…Nc7 38.Bxf4 e5 is to find counterplay in the centre. However, White has a marginal win and piece play will dominate for a long time to come.
And there is Karsten’s 37…Ne3. As a result of 38.Bxe3 fxe3 39.Kxe3 Ra5 White’s pair of bishops is eliminated. But after 40.exf5, White has a very marginal win, mainly due to his active king and superior minor piece.
Therefore, despite three more attractive alternatives to Chigorin’s 37…Nf6, Black is lost at this point.
Answer 5
As seen, after 36…gxf4 37.e4, Chigorin ended up in a losing position.
By playing 36…h6, Black stabilises his pawn structure and refuses to open up the position. Nevertheless, his defensive task remains challenging.
Even better is the spectacular 36…g4! 37.fxg4 Nd6! This is a defensive pawn sacrifice that creates a new blockade, particularly the central e4.
In both cases, White might claim a slight advantage for a long time, but no more.
All in all, by 36…gxf4 Black enters the losing zone from an objectively playable position. It has to be called a decisive mistake (?), costing half a point.
Another moment of highly controversial discussion throughout the years is Chigorin’s decision here. From jubilation to condemnation, everything is included. Like: “35…g5! This opens up the g-file for Black’s rooks to attack” (Schallopp). To GK’s: “35…g5? Black does not feel the danger and opens the position (Levenfish).” But it was only after Lasker’s acclaimed 36.Ke2, that Chigorin erred with 36…gxf4.
Instead GK supports an old recommendation by Levenfish, 35…Nd6, with the idea “to prevent the breakthrough e3-e4”. Today’s top engines rate both 35…Nd6 and 35…g5 as being roughly equal good, with an evaluation of around +0.5.
As is evident from several remarks, contemporary commentators considered the position to be balanced. Therefore, if Black wants to play for a win, he has to take action, and Chigorin’s g6-g5 fits well into this understanding. Later analysts, though, rated White’s potential much higher, noting his pair of bishops and his large number of central pawns. This is why they are urging caution.
I think it makes little sense for Black to play 35…g5 if he can’t follow up with 36…gxf4. So, from a human perspective, I consider 35…g5 to be dubious (?!), and 35…Nd6 to be a more suitable concept.
For more analytical details, please see the annotated game and for more historical quotes and suggestions, please see the second game in the list.
Here’s the game with detailled analysis:
And here is the game with extended historical quotes:
Summary
Our intensive investigation only begins at move 35, but three decisive mistakes occurred during this phase, each costing half a point.
Due to 36…gxf4 Chigorin entered the losing zone from an objectively playable position. By 47.Rd2 Lasker turned a narrowly won position into a worse one, and with 48.Rb5 he missed a possible save of his precarious situation.
From today’s perspective, several of the other decisions made by both players were suboptimal. Of note is 44.Rb4, with which Lasker initiated a multi-move manoeuvre to attack the weakest black pawn on c4. But at this point, this is too slow and stubbornly sticking to it for three more moves turned out to be a disaster. Ironically, Lasker could have successfully implemented this same plan with 41.e5 earlier on.
Last considerations
In the previous phase, there are also a few innovations to report.
Although Steinitz recommended 15.e4, this advance is a strategic mistake. Instead, White should opt for c3-c4, which is a likely winner in the moves 15 to 17. After that, 19.Bxg6! hxg6 20.Rg5 is a random winning chance. But even 20.h4, making use of “the useless assertion of the rooks on the g-file” (Stei), could be enough to keep a decisive advantage. For Black, 29…Rf7, with the idea of Rf7-f6-h6, makes his defensive task easier. The same applies to 30…Rxb5, which allows him to get rid of the excluded rook. Steinitz’s 33…g5 is possible, but it should be followed up by 34.Ba3! Kg7!. The problem is the same as in move 35, Black is well advised to avoid g6-g5 at all.
And probably there is much more to be discovered.
Unfortunately, the days when chess teachers could present this game as either a strategic masterpiece or a demonstration of how to handle a pair of knights seem to be over. The experts are right: “Black is inferior” (Stei), after granting White the pair of bishops and a large number of central pawns early in the opening. In the semi-endgame after move 13, White is “virtually winning” (GK). However, the strategic considerations resulting from the asymmetrical position usually fade into the background. Instead, short-term and dynamic demands often take centre stage. GK’s description of it as “a very complex game rich in tactics” therefore seems quite accurate. Today more than ever.
Postgame
Ultimately, Hastings 1895 was to become one of the best-seeded, most exciting and most important tournaments of all time. Nowadays, one rule seems almost exotic: “The pairings for each individual match day were announced only shortly before the start of play on the same day” (Sch). In addition to the element of surprise, this also resulted in some unusual colour distributions. From round 11 onwards, Lasker received the white pieces four times in a row, followed by the black pieces for the next five rounds.
Chigorin got off to the best start and was ahead in the standings for most of the way until the 12th round. After that, however, a three-way contest developed between him, Lasker, and Pillsbury, with the lead changing hands several times. In the end, it was the relatively unknown 22-year-old from Boston who emerged as the winner. The final standings after 21 rounds: 1. Pillsbury 16.5, 2. Chigorin 16, 3. Lasker 15.5, 4. Tarrasch 14, 5. Steinitz 13.
Rather than clarifying the true hierarchy in the world of chess, Hastings produced a fifth worthy contender. It was only after the Four Masters Tournament in St. Petersburg in 1895/96 and the one in Nuremberg in the summer of 1896, that Lasker, who won both events, was generally recognised as the best player.
Although Chigorin triumphed in this first encounter with Lasker, he was never able to repeat this success in regular tournament games. Over the next few years, including their last meeting in Cambridge Springs in 1904, the personal score became 10-3 (+8, =4, -1), clearly in Lasker’s favour.
