The Kibitzer
by Tim Harding

The Openings at New York 1924

THE THEME OF MOST of the Chess Caf‚ columns this month is
the great New York 1924 tournament which was played 75 years
ago. My brief is to look at the openings, and particularly the
discussion of them by Alekhine in the tournament book in the light
of modern developments. 

This is quite a large topic and since a lot of preliminary material is
required, the investigation of the openings other than 1 e4 will be
held over to next month's edition of The Kibitzer. Moreover,
Richard Forster (The Late Knight) is focusing on Reti and his 1
Nf3 opening, so I shall pass lightly over that.

Here is the list of the players in the order that they finished the
event, with their ages (at the start of play in March 1924) and
countries of residence

Emanuel Lasker (Germany, 55, former world champion) Jose
Raoul Capablanca (Cuba, 35, world champion) Alexander
Alekhine (ex-Russia, 31, future world champion, became French
citizen in 1925) Frank Marshall (USA, 56) Richard Reti
(Czechoslovakia, 34) Geza Maroczy (Hungary, 54) Efim
Bogoljubow (34, became German citizen in 1927) Savielly
Tartakower (37, later a French citizen) Frederick Yates (Great
Britain, 40) Edward Lasker (USA, 38) David Janowski (France,
55)

This tournament was important because it featured eight of the
world's top players in (or close to) their prime. At the bottom of
the table you find, just above one grandmaster who was long past
his best, two IM-strength competitors who found the going rather
tough at this level. 

To me, the most striking thing is the absence of young men nobody
under 30, let alone teenagers! There was one young comet on the
rise at this time   Carlos Torre   but he was not yet 20 and his
brief blaze of glory did not commence until the following year.

Of course in the early 20th century, chessmasters matured much
slower but it does seem to be true that very few people born in the
decade 1895-1904 became grandmasters. Possibly World War I
and the Russian Revolution had something to do with this. It
meant, however, that of the field of established, even middle-aged,
players in New York there was no young tiger hungry to make his
mark (Alekhine would be nearest to that description and he was
31! There were many with reputations to protect, very few with a
name to make. A policy of safety first was likely, especially with
very few easy games in prospect and each opponent to be faced
with both colours.

Stylistically, of these 11, only Marshall and Tartakower were
players with a penchant for the old open tactical sort of game that
had been prevalent up to about 1910. Maroczy, Janowski and Em.
Lasker were veterans of that period and could play in the classical
open style but were adjusting with varying degrees of success to
the new methods. 

Capablanca was one of the two players (the other being the absent
Rubinstein) whose whole way of playing was rooted in taking the
ideas of Steinitz and Tarrasch to a new level, playing for
accumulation of small advantages rather than with any preference
for kingside attack. At this stage in his career (half-way through his
title reign) Capablanca was formidable, not yet reliant solely on
technique as he became in the 1930s.

Of the newer generation, Alekhine and Bogoljubow were players
of great tactical inventiveness who favoured the closed openings as
leading to deeper, more slowly-evolving strategic attacks that were
harder to neutralise than the old 1 e4 openings. At this point in his
career, Bogoljubow (of Ukrainian extraction) had not entirely
secured his connections with the Soviet Union; he was to win the
1925 Moscow international.

Reti was the principal representative at this time of the
hypermodern school that eschewed the occupation of the centre in
the opening.

Another factor to be borne in mind when studying the openings in
New York 1924 is the limited palette with which 10 of these 11
artists, as we may call them, were employing to compose their
masterpieces. It is as if the ten greatest painters of an era were
asked to have an art competition and agreed amongst themselves
that there would be a special prize for the best use of ultramarine
(= Reti Opening) but they would not employ orange, pink and
green.

Normally in a tournament such as this, the players know in
advance who their opponents will be and (after the draw) on which
day they will meet them and with which colour. This gives good
scope for specific personal preparation especially in a double-
round event.

However, at New York being a double-round event, the contestants
could not be sure in advance who they would play or with each
colour - but at leastthey knew they would get a White and a Black
against everyone. This arrangement made only general advance
preparation possible and may partly account for the relatively small
number of specific opponent-directed innovations that were seen.

Most of them (Tartakower being the exception) had rejected the
romantic style openings such as the King's and Evans Gambits,
Italian and Vienna Games, as well as the Four Knights Game
which had enjoyed a brief vogue. Rudolf Spielmann had written
his famous article "From the sickbed of the gambits" and,
Tartakower apart, the contestants in New York seemed to agree
with him. So if you opened 1 e4 you expected Black to reply 1...e5
and there would be a Spanish Game (Ruy Lopez). The principal
alternative that you might meet was the Sicilian, but in this era
there was no Dragon, Najdorf, Sveshnikov etc. as Black played the
Sicilian slowly with positional objectives and postwar all-out-
attack methods against it had not yet been developed.

Alternatives to 1 c4 and 1 d4 were limited, also, although 1 c4 was
occasionally essayed and this tournament (as we shall see later) did
make some contribution to recognising the importance of 1 Nf3.
The biggest opening surprise in New York was when Tartakower's
visit to the ape-house at the New York zoo inspired him to open 1
b4 against Maroczy in round 4.

He obtained a draw but in the tournament book Alekhine sniffed at
it: "An old move, the chief drawback of which is the fact that
White discloses his intentions before knowing those of his
opponent. He need not necessarily be at a disadvantage thereby,
but is not that altogether too small a satisfaction for the first
player?" 

In a way Alekhine was right but, 75 years on, his reaction to
another of the few enterprising opening choices in New York
seems very strange

"Capablanca took the liberty once of playing this Joke Opening. He
was favoured by the fates as his opponent, Ed. Lasker, voluntarily
renounced taking possession of the centre through d4... Naturally,
this experiment has no claim to any theoretical significance." 

It was really only in the 1950s, when the Austrian grandmaster
Karl Robatsch began to play it regularly, that this "joke opening"
(along with the allied Pirc Defence, in which Black prepares the
fianchetto by 1...d6 and 2...g6) began to be respectable. Since the
1970s the Modern Defence, as it is now known, has been
considered a major and more-or-less sound defence. As for 1...a6,
with which Tony Miles once defeated Karpov, what would
Alekhine have said about that!

Against 1 d4, as we shall see in more detail next month, Black
employed only a tiny fraction of the modern repertoire. Even
several defences that were known at that time were not used, such
as the Gruenfeld and Nimzo-Indian. The English Opening was also
virtually absent - Tartakower used it once and Alekhine had a
game that transposed to the King's Indian. So the Reti was really
the only relatively novel opening employed in this event (five of
the players opened 1 Nf3 at least once).

The more that you concentrate on the games of the leading
contenders for the top prizes, the more my remarks apply.
However, because this was a tournament, they had to meet the
other opponents too, which did introduce some variety. 

Seven of the New York field had played in the London 1922
tournament and four had competed at Mahrisch-Ostrau in July
1923; the games of both these events would be familiar to all in
New York. However, there were relatively few top games played
and so much less to study and ideas developed more slowly.
Moreover, if a master had a new idea he could try it out in local
events or a simultaneous display with little risk that rivals would
find out about it - there was no TWIC in those days!  Whether even
the leading masters of this period practised openings preparation as
understood by grandmasters of today is doubtful. Of course there
were no computers then either, but the leading GMs of the 1950s
and 1960s certainly kept very detailed notebooks and/or card
indexes. The players would no doubt have traveled with some
periodicals and the books and bulletins of recent top tournaments
but of course such events were few and far between.

Unfortunately the biographers of players like Alekhine,
Capablanca and Lasker tended to concentrate on the principal
incidents in their lives and chess careers rather than reveal their
methods of study. It is known, from a fairly recent biography in
German, that Ernst Gruenfeld was (round about this time) probably
the first professional player to develop a kind of portable openings
index that he could bring to tournaments in a suitcase.

Of the 110 games in New York, 28 were played in openings with
the Informator/ECO "A" classification, including the Reti. Twelve
were "B", 38 were "C", 27 were "D" and only five were "E"
openings. White in fact opened 1 e4 on 47 occasions out of 110, as
two of the "B" group openings featured a form of Modern Defence
transposing from 1 d4 Nf6 2 Nf3 g6 3 Nc3 or 3 Nbd2 followed by
e4. The opening move 1 d4 was almost as popular as 1 e4.

After the age of the players, the next most striking thing compared
with the tournament practice of today is the very small number of
Sicilian Defences played. Of the 12 games classified as "B" just
half were Sicilians and the other six various semi-open games
other than the French. Since "B" has been the most popular
openings group since the 1970s (just look at the relative fatness of
the ECO volumes if you have them on your shelf) this testifies to a
major shift in the past 75 years, in fact a shift that really began in
the 1950s.

The Caro-Kann and Scandinavian (Center Counter) were known
but not much employed at this time; they got one and zero outings
respectively in New York. Alekhine had introduced his defence
1...Nf6 in 1924 but employed it only once in New York, as did
Tartakower. The game was of no theoretical interest because
Maroczy   so far from preparing anything approaching like an
attempt at refutation   chose to defend from move two with 2 d3.

It is surprising that the Sicilian Defence took so long to develop.
For example, Howard Staunton had written half a century earlier in
his "Chess-Players' Handbook" (e.g. page 371 of the 1872 edition)
that

"In the opinion of Jaenisch and the authors of the German
'Handbuch' with which I coincide, this is the best possible reply to
the move of 1 e4 as it renders the formation of a centre
impracticable to White and prevents every attack." With these
three major authorities backing it, the Sicilian still languished
through the second half of the 19th century, Louis Paulsen being
its main exponent and it was his variation that still held sway.

In his new book "Secrets Of Modern Chess Strategy", American
IM John Watson writes (page 245) about Alekhine's apparent
obsession with an early Be2 move for White against the Sicilian
Defence (i.e. before playing d4). Janowsky-Lasker (round four)
went 2 Nf3 Nc6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 d6 6 Be2 e6 with a
form of Scheveningen Variation that had been introduced the
previous year. Of course in 1924 the Sozin and Richter-Rauzer
attacks had not yet been invented, and nor of course had the
Boleslavsky variation (6 Be2 e5). In his notes, Alekhine said that
2...Nc6 was preferable to 2...e6 because if 3 Be2 Black can make
White block his c-pawn by 4 Nc3 (in view of 4 e5? Ng4 5 d4 cxd4
6 Nxd4 Qc7). Strange logic to the modern eye.

After 1 e4 c5 2 Nf3 e6 (Lasker-Tartakower, round 5) Alekhine
criticises White for playing 3 d4 instead of his favoured 3 Be2
which keeps the c4 option open. Watson finds this both amusing
and revealing about Alekhine's approach to the defence"At first, if
one only follows his written notes, he seems to be almost absurdly
dogmatic about 1...c5 (not to mention wrong)." On the other hand,
I do not necessarily agree that 3 Be2 is a weak move.

Alekhine did not play the Sicilian himself in New York. It was
adopted by Tartakower (three times), Emanuel Lasker (twice) and
Bogoljubow (once); White won three, Black two with a draw. Reti-
Tartakower began 1 Nf3 g6 2 e4 c5 and now, Alekhine wrote, 3 c4!
should have been played. Of course nowadays that is only one of
several moves White might consider.

Moving on to the "C" classification, there were nine French
Defences (C00-C19) of which one transposed from 1 e4 Nf6. Five
of the others were in the MacCutcheon Variation (1 e4 e6 2 d4 d5
3 Nc3 Nf6 4 Bg5 Bb4) see diagram. Alekhine wrote that the New
York games "have contributed considerably to a further clearing up
of the problems as the result of two of the lines adopted".
Bogoljubow had employed 5 Nge2 to defeat Tarrasch (Pistyan
1922) and now he scored a comfortable win with it against Reti in
round four. In a later round against Dr. Lasker, Reti improved at
move 71 e4 e6 2 d4 d5 3 Nc3 Nf6 4 Bg5 Bb4 5 Nge2 dxe4 6 a3
Be7 7 Bxf6 gxf6 (instead of 7...Bxf6) but after 8 Nxe4 he played
8...f5 and lost again. A superior treatment was seen in Bogoljubow-
Alekhine (8...b6, with a draw in 85 moves) which led to "a
bearable game" for Black but was no refutation of 5 Nge2 as
Alekhine said. His line is probably as good as any against 5 Nge2
which remains a playable move for those who wish to avoid the
complications of 5 e5.

Capablanca preferred to meet the MacCutcheon by 5 exd5, against
both Alekhine and Bogoljubow. If you believe "ECO", Black
should transpose to an Exchange Variation by 5....exd5 but as a
rule MacCutcheon players do not like to do this. In retrospect, the
future world champion preferred Bogoljubow's way of meeting
this line (5...Qxd5 6 Bxf6 gxf6 7 Qd2 Qa5!) to his own (6...Bxc3)
although in the sequel it was Alekhine who scored half a point
while Bogoljubow eventually lost. Nowadays 7 Nge2! is
considered more critical than Capablanca's 7 Qd2 but Black has a
wider choice of replies to 5 exd5 than he has to 5 Nge2 and his line
can be avoided if Black does not like it.

There was one potentially theoretical clash in other lines of the
French. Yates adopted the Albin-Chatard-Alekhine Attack (1 e4 e6
2 d4 d5 3 Nc3 Nf6 4 Bg5 Be7 5 e5 Nfd7 6 h4) and was probably
not surprised that Maroczy answered 6...f6 since the Hungarian
GM had introduced this move at Mannheim 1914. (See Diagram)

This was evidently a good opportunity for a talented IM to cause
an upset by springing a major tactical innovation against his
eminent opponent to create enormous complications which he
could spend many happy hours analysing in his cabin crossing the
Atlantic. However, instead of 7 Bd3 (analysed by Alekhine in the
tournament book) or 7 Qh5+!? (discovered only in the 1930s)
Yates played the tame 7 exf6?! so perhaps he had not done any
serious opening preparation after all!

Moving on to the 1 e4 e5 Open Games (C20-C99), there were 29
of there; in four cases Tartakower bucked the trend and played the
King's Gambit. The other 25 comprised one Philidor Defence, one
Scotch Game, three with forms of the Three Knights Game (two
via Petroff and one via 2...Nc6) and one slow Italian/Two Knights
hybrid. 

The Ruy Lopez was played 19 times; three miscellaneous
defences, one Exchange Variation, one Modern Steinitz (3...a6 4
Ba4 d6). Nine of the games went down the main line known as the
Closed or Chigorin Defence. The most important discussions in
New York for the 1 e4 openings concerned the main lines arising
from 1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 Nf6 5 0-0 Be7 6 Re1 b5 7
Bb3 d6 8 c3 0-0 (See Diagram).

This position occurred several times, with White playing 9 d3
(twice, including a case which went 8 d3 and 9 c3), 9 d4 (played
three times by Yates) and 9 h3! twice each by Emanuel Lasker and
Maroczy. 

This move 9 h3 only scored 50%; it had not yet become the GM-
killer which it later evolved into in the hands of such mighty
exponents as Keres and Fischer. After 9...Na5 10 Bc2 c5 Maroczy
timidly played 11 d3 against Capablanca and was duly punished by
defeat. Following the correct 11 d4 Qc7 12 Nbd2 a position arose
which was tested hundreds of times at a high level in later years,
especially the 1950s and 1960s. In New York, Black chose lines
involving the opening of the c-file whereas in later years Black
frequently left to White the decision about whether to open the d-
file, close the centre or maintain the tension. For example,
Emanuel Lasker v Ed. Lasker continued 12...cxd4 (12...Nd7?! was
played in Lasker-Reti but after 13 d5 Reti chose 13...f5?, rightly
criticised by Alekhine, and White easily won) 13 cxd4 Bd7 (The
same position arose in Maroczy-Reti via 12...Nd7 13 Nf1 cxd4 14
cxd4 etc.) 14 Nf1 Rfc8 and now the former world champion
answered the threat to his bishop by 15 Re2 whereas Maroczy
preferred the simpler 15 Bd3.

In either case, Alekhine wrote in his openings survey that "White
obtained a lasting positional advantage... as a result of the almost
irremovable obstacle of lack of freedom of movement of the
opposite pieces". In fact both these games were drawn, with the
Lasker-Lasker encounter featuring the well-known "miraculous"
position draw with White having knight against rook and pawn, but
that doesn't mean Alekhine was wrong. However, perhaps neither
of the moves seen in 1924 is best. The teenage Bobby Fischer
played 15 Ne3 to beat Bisguier in the 1958 USA Championship
and Shirov-Piket, Madrid 1997, is a more recent example, also
won by White.

Black has a multiplicity of plans after 9 h3 but only a few are seen
nowadays in master play. In postwar years Smyslov tried the move
12...Bd7 with...Rfe8 while 12...cxd4 was played in conjunction
with 13...Bb7 and other moves but eventually the fashionable
11...Qc7 was virtually abandoned. Also the Keres move 11...Nd7
did not quite stand the test of time. In the mid-1960s Black looked
for other ways of meeting 9 h3 in the diagram position. Breyer's
old 9...Nb8 was revived and Smyslov's 9...h6 (to enable...Re8
and...Bf8) was also popular. I witnessed a game at a London open
tournament (circa 1968) in which IM Michael Basman (still
playing orthodox openings in those days) was so astonished by
9...h6 and convinced that it must be a weakening "patzer move"
that he played 10 Bxf7+??! against an opponent who knew what he
was doing, and Michael soon had to resign!

Eventually Igor Zaitsev and Anatoly Karpov recognised that 9...h6
was an unnecessary and weakening move because after 9...Bb7 10
d4 Re8 White can only get a repetition of moves by playing Ng5;
therefore the Smyslov variation has disappeared from master
practice along with the old Chigorin method. The Zaitsev and
Breyer lines (along with the Marshall/ Anti-Marshall complex)
remain the critical lines of the Spanish at the end of the century.

In Alekhine's opinion, one of the main contributions of the New
York debates was to show that (from the above diagram) 9 d4 is
inferior to the preparatory 9 h3, because of a discovery evidently
made in pre-tournament preparation by Bogoljubow. Both versus
Capablanca (London 1922) and Lasker (Mahrisch-Ostrau 1923)
Bogoljubow had replied 9...exd4 but now against Yates he played
9...Bg4! with a pin which puts the white centre under pressure.
(See Diagram)

Although 9 h3 has proved superior to 9 d4 when Black adopts the
7...d6 8 c3 0-0 move order, this position remains important as it is
one of White's two main ways to avoid the Marshall Attack. When
Black employs 7...0-0 (instead of 7...d6) White can play 8 d4 and
after 8...d6 9 c3 (there is nothing better) 9...Bg4 the diagram
position frequently arises.

In later years, 10 d5 became popular but in New York, Yates
preferred 10 Be3 exd4 11 cxd4 Na5 12 Bc2 Nc4 (nowadays 12...c5
is considered to give Black a good game) 13 Bc1 c5 14 b3 which is
still reckoned to be a main line (ECO C91, row 9 on page 461 of
the 3rd edition). Now Bogoljubow tried 14...Na5 (heading for c6)
and Yates allowed this, eventually losing, whereas 15 d5! is given
in today's theory books as somewhat favourable to White,
following various games such as Bronstein-Smyslov, Petropolis
interzonal 1973. The tournament book of that event (edited by
Wade and Blackstock) suggested 15...Nxd5!? 16 exd5 but after the
fireworks Black still seems to be worse17 Nc3 b4 (17...Bxc3?? 18
Qd3) 18 Bb2 bxc3 19 Bxc3! Bxc3 20 Qd3 g6 21 Qc3 and it is hard
to see why they claim equality in view of Black's offside knight
and weakened kingside.

KING'S GAMBIT

Tartakower bravely played the Little Bishop's Gambit (1 e4 e5 2 f4
exf4 3 Be2) four times. First mentioned by the Russian analysts
Petroff and Jaenisch in the mid-19th century, it had never been as
popular as 3 Nf3 or even 3 Bc4. After all the bishop appears to
attack nothing but on the other hand it is not exposed to attack
either. The absence of direct threats can make it hard for Black to
decide what to do. If Black checks the white king he can end up
losing time. (See Diagram)

Tartakower won his first two games with it against Bogoljubow,
after a tough struggle, and Yates, who replied with the ineffectual
3...Nc6. However, he only scored a draw (v Alekhine) and a loss (v
Capablanca) later in the event. They were stronger players and also
had time to prepare something. 

Alekhine himself chose to experiment with 3...Ne7 (preparing
4...d5) and in his openings survey he looked at this in more detail;
he also suggested that 3...f5 4 exf5 Qh4+ would be a more logical
line for Black, not mentioning the superior 4 e5. 

He was probably right to say that none of the four games was in
itself really significant theoretically; nevertheless anybody
contemplating 3 Be2 should study the material on it to be found in
the tournament book. I had to meet the move in a correspondence
game many years ago and eventually lost; I never thought of
looking in the New York book for guidance.

Tartakower-Alekhine went 3 Be2 Ne7 4 d4 d5 5 exd5 Nxd5 6 Nf3
Bb4+ 7 3 Be7 8 0-0 0-0 9c4 Ne3 10 Bxe3 fxe3 11 Qd3 Bf6 12 Nc3
Nc6 13 Nd5 and now Alekhine said that after 13...Re8 (he actually
played 13...Bg4) "White will have difficulty in winning back the
gambit pawn and still have a free game". 

White can find slightly different ways to play to avoid Alekhine's
recommendations, for example 4 Nc3 is probably better than 4 d4.
So 3...d5, as Capablanca played, may be as least as good. In the
1960s the British player Norman Littlewood (brother of John and
father of Paul) employed the 3 Be2 gambit with some success and I
don't consider that it is dead and buried yet.