Chess Archaeology HomeChess is a scientific game and its literature ought to be placed on the basis of the strictest truthfulness, which is the foundation of all scientific research.W. Steinitz

Missing Games — Sixth Match, 1834 La Bourdonnais-McDonnell Matches
Researched by Nick Pope

(Return to Main Index)

If game 80 was incorrectly identified previously, and corrected, we arrive with the following color sequence for the games in the sixth match:

Chess Studies: 777879808182838485Total
 Bourdonnais0110010003
 McDonnell1001101116
 Drawn         0

If this sequence is to be believed, it would mean that three decisive games are missing. Two games where McDonnell has White, that would appear to belong between games 77/78 and 78/79, and one additional game where McDonnell had black and would belong between games 80/81. However, based upon the counts for the last match (8 wins for McDonnell, 3 for Bourdonnais, and 1 draw), we know that this cannot be the case.

While there are several ways to resolve this conundrum, the most expedient solution is that games 80 and 79 were simply transposed in the sequence given by Greenwood Walker and followed in Chess Studies. There is a historic precedent already established of Greenwood Walker publishing games out of sequence (see third match, games 44 and 45 in Greenwood Walker and Chess Studies). I do not think it too outlandish that another pair of games could have been given out of sequence.

By swapping games 79 and 80 in the sequence given in Chess Studies, we arrive with the following series:

Chess Studies: 777880798182838485Total
 Bourdonnais0101010003
 McDonnell1010101116
 Drawn         0

Now we end up with Bourdonnais having only two consecutive games with the first move. This appears to indicate that the two missing wins for McDonnell are either decisive games where McDonnell had white in both games which would fit before game 77 and between games 77/78 or McDonnell could have had white in one game and black in the other with one fitting between games 77/78 and the other after game 85.

The draw would be trickier, if not impossible, to accurately place. A quick check of the distribution of previous draws shows that 53% of draws occurred within the first three games in the prior matches, and the greatest concentration of draws occurred as the second game in prior matches. Based upon this distribution I would place the draw as either the first or third game of this match if the pairing was McDonnell-Bourdonnais, and as the second game of this match if the pairing was Bourdonnais-McDonnell.

With a number of various sources, the next step was to comb through the earliest records to see which games were published and when, to compare the games and to check for clues, omissions, and errors.

Lewis gives 50 games. George Walker claimed to have given "38, with the exception of four or five which were not preserved" and later "the whole of the remainder" in Bell's Life in London. William Greenwood Walker gives 83 games (noting that game 14 wasn't worth preserving even though it had been previously given in Bell's and omitting game 85 which had previously been given by Lewis). The Chess Player's Chronicle gives the 85 games in a serialized format. And finally George Walker's Chess Studies gives the 85 games in a single volume.

The thing which really caught my attention was Walker's statements about publishing "38, with the exception of four or five which were not preserved" and "the whole of the remainder" in Bell's Life in London:

On the occasion of La Bourdonnais' recent visit to London, he played 88 games, in the Westminster Chess Club, with Mr. M—, the first English player. Of these, 14 were drawn; and of the remaining number, La Bourdonnais won 44, and lost 30. A selection of 50 of these games was made, and printed by Mr. Lewis; but it is matter of great regret to the Chess World that he did not publish the whole of the 88.

Of the 88 games played in the great match between M. de la Bourdonnais and Mr. M'Donnell, 50 were printed by Mr. Lewis, and the remaining 38, with the exception of four or five which were not preserved, have all appeared in our selection. The thanks of the Chess world are certainly due to us for thus having secured what, without our aid, would never have come before them. Many of the finest games were not included in the printed 50; for instance, that splendid Muzio Gambit (No. 2), and the two specimens of Captain Evans' game (Nos. 6 and 18 of our collection)—all three won by Mr. M'Donnell. Mr. W. Greenwood Walker will reprint the whole of these games in his forthcoming selection.

The total number of games played by La Bourdonnais and M'Donnell was 88. Of these Mr. Lewis printed fifty, and we believe we gave the whole of the remainder before they were re-edited in an improved form, and printed with the numerous other games of Mr. M'Donnell, by Mr. William Walker.

To my knowledge, no one had ever identified which 38 games were published in Bell's Life in London. In fact, as far as I know, no one has ever cited Bell's Life in London in any prior research about the Bourdonnais-McDonnell matches. This appears to be an untapped resource.

Taking inventory of the even strength games published between Jan 4, 1835 and May 1, 1836 (the date Walker first states that he published 38 games), I was able to identify 36 games being from the six matches. This leaves the tantalizing possibility that two additional games could be found amongst the remaining even strength games published in Bell's Life in London during this period. The candidates are Bell's games XXVI, XXXI, XLIV, XLVII, LV, LVIII, LXXV, LXXVI, LXXX, LXXXI, and LXXXIII.

I eliminated the following three games as being previously identified:

XXVI is a King's Gambit Accepted. Jay Whitehead identifies this game Walker-Nixon although I have not found his source for the attribution.

XLIV is another King's Gambit Accepted. This game is found in Chess Studies, game 453, as Brand-Fraser.

XLVII, yet another King's Gambit Accepted. It is found in Chess Studies, game 871, as Walker-NN.

I then eliminated game LVIII, which is a King's Knight's Gambit where White plays 5.0-0 in response to 4...g4, due to a note to the game as given in Bell's: "This game was played by two young players [...]".

This left me with seven candidates, three wins for Black: LXXV, LXXXI, and LXXXIII; and four wins for White: XXXI, LV, LXXVI, LXXX.

LXXV is an Italian Game, where White plays 4.0-0 instead of 4.c3. This game is found in Chess Studies, game 994, without identifying either player. Bourdonnais invariably played 4.c3 in the four previous Italian Games, so it seems odd if he had adopted it now. While not impossible it is an unlikely candidate.

LXXXI is a KP1 (Sicilian) Opening. In the previous twenty KP1 openings McDonnell had the first move. This game is found in Chess Studies, game 924, without identifying either player. McDonnell never adopted a single KP1 defense (Sicilian or French) in any of the six matches and it would appear to be an extreme time to adopt a new defense. This appears to be a highly unlikely candidate.

LXXXIII is an Italian Game, Rousseau variation, where Black plays 3...f5. This game is found in Chess Studies, game 620, without identifying either player. McDonnell invariably used the standard defense of 3...Bc5. As in game LXXV, while not impossible, it seems to be an unlikely candidate.

None of these wins for the second player appear to be likely candidates. This leaves four wins for the first player to be checked:

XXXI is an Evans' Gambit. A popular opening played during the six matches. This particular Evans' Gambit reached the following position: 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.b4 Bxb4 5.c3 Be7, which had also occurred in the fourth match, game 51.
This was the only time Bourdonnais successfully won an Evans' with the black pieces and it would appear to fit the narrative of the sixth match with Bourdonnais resorting to a line he was previously successful with, especially if he was falling behind in the match. Another point in it's favor is that this game was published with another game from the match, Bell's game XXXII, which was verified as game 68 from the fifth match. This game is published in Chess Studies as game 705 without identifying either player. This is a likely candidate.

LV is a King's Gambit. More specifically it is a Bishop's Gambit wherein White plays 5.Nf3. McDonnell usually played 5.Nc3. McDonnell had also virtually given up playing the Bishop's Gambit after going 0 for 4 in the first match (see games 11, 18, 20 and 22). He did play it one more time, in the third match (game 46), again losing. He was more successful with the King's Knight's Gambit so I find it unlikely that he would revert back to a variation in which he had lost with all five attempts. This game is found in Chess Studies, game 818 with no names given. This appears to be an unlikely candidate.

LXXVI is an Italian Game. This game is found in Chess Studies, game 563 with no names given. McDonnell preferred an Evans' Gambit over a regular Italian Game during these matches and would have had no reason to start playing it at this point in the contest (i.e. the sixth match) as he was doing just fine with the Evans' Gambit. This appears to be an unlikely candidate.

LXXX is a Bishop's Opening. This game is found in Chess Studies, game 788 with no names given. McDonnell played the Bishop's Opening three times in games 24, 30, and 47 getting one win, one draw and one loss.
He had previously played this variation, 3.c3, in game 47. This is a likely candidate.

Based upon the fact that both game XXXI (game 705) and LXXX (game 788) were in McDonnell's opening repertoire; 5...Be7 being a move Bourdonnais had used in a previous Evans's Gambit; and the fact that Walker indicates that he published the remaining games twice, via "the remaining 38" and "the whole of the remainder", I put forth that these two games are possibily the two missing McDonnell wins and belong before game 77 and between games 77 and 78.

If this Evans' Gambit, game 705, is one of the missing games, it also solves a minor mystery concerning the counts relating to the number of Evans' Gambits played. The summary of which was presented by Ernest Adams as follows:

The number of won games are stated by both writers to be 20. I find 12 won by La Bourdonnais, 7 by M'Donnell, and 3 drawn. Mr. Greenwood Walker records 10 won by La Bourdonnais, and 10 by M'Donnell. His errors must, I think, have arisen from some accidental transposition of the players' names in some of the games.

The following tables illustrate the counts of the games as published by Greenwood Walker with the addition of game 705. There were three drawn Evans' Gambits which are not included in the following tables (games 52, 59 and 65).

The first table is counts of the games as published in Greenwood Walker's book. The second table factors into account that Greenwood Walker gives game 73 incorrectly (previously given by Lewis with the correct pairing).

As published
 Game  LCM  AM 
2601
5110
5310
5610
6010
6310
6601
6710
6910
7101
7201
7301
7501
7701
7910
8010
8201
8301
8401
Total910
70501
Total911
   
Lewis correction
 Game  LCM  AM 
2601
5110
5310
5610
6010
6310
6601
6710
6910
7101
7201
7310
7501
7701
7910
8010
8201
8301
8401
Total109
70501
Total1010

After correcting for game 73 we arrive at the mysterious 10 wins each as reportedly stated by Greenwood Walker. Admittedly, I have never found a printed summary from Greenwood Walker stating that each player had won 10 Evans' Gambits out of twenty decisive games, but I can see how such a conclusion could have been reached.

Now the next three tables are the totals based upon the printed games given in the Chess Player's Chronicle, Chess Studies, and my own correction to game 80. The Chronicle corrects game 82, for a total of 11 wins for Bourdonnais and 9 for McDonnell. Chess Studies goes further and corrects games 71 and 82, for a total of 12 wins for Bourdonnais and 8 for McDonnell.

Chess Player's Chronicle
 Game  LCM  AM 
2601
5110
5310
5610
6010
6310
6601
6710
6910
7101
7201
7301
7501
7701
7910
8010
8210
8301
8401
Total118
70501
Total119
   
Chess Studies
 Game  LCM  AM 
2601
5110
5310
5610
6010
6310
6601
6710
6910
7110
7201
7310
7501
7701
7910
8010
8210
8301
8401
Total127
70501
Total128
   
Pope
 Game  LCM  AM 
2601
5110
5310
5610
6010
6310
6601
6710
6910
7110
7201
7310
7501
7701
7910
8001
8210
8301
8401
Total118
70501
Total119

If these two decisive games were from the match, why didn't George Walker include them as such in Chess Studies? Perhaps an explanation can be found in the following:

But during a succession of months, who shall say this gentleman was never absent one or two days from illness? And may not two or three games have been lost in this manner? I think writers are seldom expected to make oath to the truth of their statements. I looked over the majority of the games as they were played myself, but was out of town for a fortnight towards the end of the contest.

During the last two or three weeks of the great match (for the whole may be termed one great match broken into chapters) I was out of town, and thus only got hold of the last games in gross, which may have probably added to the difficulty.

I have no early file of Bell to which I can refer, but think that many notes accompany the games, valuable as being gathered by me at the time from the players themselves.

I suspect George Walker lost track of what he had published in Bell's Life in London over the next eight years but he managed to keep the loose-leaf MS copies given to him from Greenwood Walker of the games (most likely with just color indicators, and no names, at the top of each page). So when George Walker got around to compiling games for Chess Studies he included them without recognizing them as being games from the match or games he had previously published in Bell's Life in London (i.e. "I have no early file of Bell to which I can refer").

There is precedent for George Walker losing track of what he had previously published. Greenwood Walker claimed that game 14 had not been preserved, however, it had already been published by George Walker in Bell's Life in London, May 31, 1835, long before Lewis gave it to Staunton for publication on August 2, 1841.

As these two decisive games had found their way into George Walker's Chess Studies (with no names attributed), I began checking Chess Studies for any unattributed draws published by George Walker. The hope being that if he had published the two decisive games then perhaps he may have published the missing draw.

In searching through Chess Studies I found four draws that were not identified by George Walker (games 593, 859, 881, and 897).

I was able to quickly eliminate game 881 as it is attributed to Lewis vs Sarratt, 1816, in the Oxford Encyclopedia of Chess Games.

UPDATE 20 April 2020: Game 593 has been identified from Bell's Life in London, 1839.05.19, stating that it was a "Game just played by two of the leading players in the Westminster Chess Club." Thus removing it from the pool of candidate games.

UPDATE 21 April 2020: Game 897 has been identified as being played in 1839 by two English players, according to Walker in Bell's Life in London, 1839.09.01, this was: "A quiet and neat game just played between two of England's leading players: a sort of Claude Lorraine affair—round, if not very spicey. One of the combatants is provincial, the other London, both being members of the Westminster Chess Club."

This left me with three remaining candidates: Leaving only one possible hope that the missing draw could be game 859:

593 is an Italian Game. Jay Whitehead attributed it to George Walker vs NN in his historic database, but I have not found the original source for Jay's attribution. We know McDonnell eschewed the normal Italian Game for the Evans' Gambit, but Bourdonnais did play it occasionally during the matches, so it would probably be Bourdonnais-McDonnell if this were the missing draw. This is a possible, but highly unlikely, candidate in my opinion.

859 is a King's Knight's Gambit. McDonnell was the main exponent of the King's Gambit during these matches, however, in this particular variation we see White playing 5.0-0 after Black's 4...Bg7. McDonnell had previously played 5.d4 in the third match, games 42 and 44 with split results. Bourdonnais also played 5.d4 in the fourth match, game 55, with a draw. This line seems out of character for both players, however, we cannot rule it out completely. This too is a possible, but unlikely, candidate in my opinion. This is the only remaining candidate for the missing draw (which may also fall in short order).

897 is a Queen's Gambit Accepted. Bourdonnais was the main proponent of the Queen's Gambit and the one time McDonnell ventured forth with it Bourdonnais declined the gambit pawn. This game reaches the following position: 1.d4 d5 2.c4 dxc4 3.e3 e5 4.Bxc4 exd4 5.exd4 Bd6 6.Nf3 Nf6.
This position had been reached previously in the first match, game 12, where Bourdonnais played 7.0-0. In this game White ventures forth with 7.Qe2+, which seems to be a slight improvement. Of the three drawn games, this game seems to be the most likely candidate.

While the evidence around this draw is not as strong as the evidence for the two decisive games, it does fit the framework of the 1834 matches. It was an opening frequently used by Bourdonnais with good success (11 wins, 3 losses, and 1 draw) and it appears to be a line that Bourdonnais had played previously (in game 12). I tentatively put forth that this game could be the missing draw
.

Now what about the statement appearing in Bell's Life in London, May 1, 1836, regarding "with the exception of four or five which were not preserved"? Was doubt starting to creep into George Walker's totals due to the forthcoming publication of Greenwood Walker's book?

Of the 88 games played in the great match between M. de la Bourdonnais and Mr. M'Donnell, 50 were printed by Mr. Lewis, and the remaining 38, with the exception of four or five which were not preserved, have all appeared in our selection. The thanks of the Chess world are certainly due to us for thus having secured what, without our aid, would never have come before them. Many of the finest games were not included in the printed 50; for instance, that splendid Muzio Gambit (No. 2), and the two specimens of Captain Evans' game (Nos. 6 and 18 of our collection)—all three won by Mr. M'Donnell. Mr. W. Greenwood Walker will reprint the whole of these games in his forthcoming selection.

Taking into account that games 14 and 85 are missing from Greenwood Walker's book, coupled with the two missing McDonnell wins and a missing drawn game, would this not account for George Walker's statement about there being "four or five which were not preserved"? I suspect George Walker knew that there were going to be just 83 games published by Greenwood Walker and that he made an assumption that four or five games were not preserved as he knew there should be 88 games in total.

One final issue about the sixth match being unfinished.

Chess.—We cannot tell our Correspondent where to get the French poem he alludes to, but believe it may be procured through any foreign bookseller. The author is the celebrated Mery, the title "Le Revanche de Waterloo," and the subject, we are told, for we have not yet received it, is the victory of La Bourdonnais over M'Donnell. We hope that the author is honest enough to sing how COMPLACENTLY the French hero went through the first twenty-one games, of which he won no less than sixteen, through Mr. M.'s nervousness: how SMILINGLY he pursued the match through the next stage in which, though M.'s play was fast improving, the Frenchman kept his advantage; lastly, how FROWNINGLY Monsieur performed the third act of the drama (the said third act being equal in length to the other two, which comprised forty-two games, of which our lamented countryman won twenty-two; and last of the lastly, by way of epilogue, we hope M. Mery laments the fact, in suitable rhymes, that M. La Bourdonnais, having lost eight games out of the last eleven they played, was unfortunately compelled to business to depart for Paris at an hour's notice, leaving the match unfinished. We think it probable that, unless M. Mery has done just to both sides, a friend of our's will translate his poem, cum notis variorum, which the truth will be told in round terms.

Nearly all the facts quoted by Mr Sullivan are wrong, which makes his decision mere chance-work. It is not true that the match between De la Bourdonnais and M'Donnell consisted of four matches of 21 games each. We ought to know something about this, as we helped to capture those matches personally. They began with a match of 21 games and as draws were not, of course, counted, the match actually included 25; of which De la Bourdonnais won 16, lost 5, and drew 4. Their second match consisted of 9 games, and the third match of 11, with one draw. Their fourth match was also to compose 11, but consisted of 18, there being 7 draws. What says Mr Sullivan to this? And why not have written to England for the facts, before assuming the judicial cloak. The fifth match also consisted of 11, with one draw; and the sixth, and last match, was also to have been 11, but was left unfinished, 9 games only being played. Thus, instead of four, they played six matches, of which we have the dates now before us; the total of games being 85, of which De la Bourdonnais won 46, lost 26, and drew 13. The second match was won by our countryman, 5 to 4.

The sixth and last match was never finishednine games only being played, of which De la Bourdonnais won five to four. The match was suspended, owing to the Frenchman's being obliged to leave London for Paris on business, and his opponent's being equally forced by circumstances to make a visit to his native town, Belfast.

During the last two or three weeks of the great match (for the whole may be termed one great match broken into chapters) I was out of town, and thus only got hold of the last games in gross, which may have probably added to the difficulty.

Based upon my research, it would appear that the sixth match was fought to a completion of 11 decisions. George Walker, as stated, was out of town during this final match and both Bourdonnais and McDonnell left London prior to his return. His absence, coupled with the incomplete presentation of the games played in the sixth match in William Greenwood Walker's book, leads me to suspect that George Walker had made a conclusion that the sixth match must have been left "unfinished". However, neither Lewis or Greenwood Walker ever make mention of the sixth match being unfinished. All we have is the word of George Walker, who directly stated that he was not present for the final match.

It is my belief that 12 games were played in the sixth match; one draw and 11 decided games. And if the match was for 11 decisions, as previously stated, then it was brought to a conclusion and was not "unfinished". This would make the 8 to 3 score in favor of McDonnell a conclusive result and aligns the history of the match with the earliest historical records.

I would like to thank David Moody, John Hilbert, Joost van Winsen and Nikolai Brunni, for providing me with a much needed sanity check regarding this research. It is far too easy to fall victim to confirmation bias and I would not have put forth this research if their reviews had raised any serious doubts regarding my conclusions.

[Archive] [Excavations] [Gallery] [Journal] [Library] [Links] [Legend] [Market]
© 1999-2021 Jacques N. Pope. All Rights Reserved.