Hans Berliner: A life in chess,computer chess and much more

Abstract

Hans Jack Berliner was born in Berlin on Jan 27, 1929, a birthday he shared with Wolfgang Amadeus Mozart who was born in 1756. His family fled Nazi Germany in 1937 and went to the US, taking up residence in Washington D.C. Hans was only eight years old and spoke very little English with a thick German accent. He attended Henry Cooke Public School on 13^th St. NW where life was not easy for immigrants. Astonishingly, he appears in the essay How I Started to Write by Carlos Fuentes, who also attended that elementary school. The essay is included in Phillip Lopate’s anthology The Art of the Personal Essay (Fuentes, 1994; Lopate, 1994). Fuentes, a Mexican immigrant, describes his school experience and the impact that Hans Berliner made on him at that early age:

Instantly, surprisingly, I became a pariah in my school. Cold shoulders, aggressive stares, epithets, and sometimes blows. Children know how to be cruel, and the cruelty of their elders is the surest residue of the malaise the young feel toward things strange, things other, things that reveal our own ignorance or insufficiency. This was not reserved for me or for Mexico: at about the same time, an extremely brilliant boy of eleven arrived from Germany. He was a Jew and his family had fled from the Nazis. I shall always remember his face, dark and trembling, his aquiline nose and deep-set, bright eyes with their great sadness; the sensitivity of his hands and the strangeness of it all to his American companions. This young man, Hans Berliner, had a brilliant mathematical mind and he walked and saluted like a Central European; he wore short pants and high woven stockings, Tyrolean jackets and an air of displaced courtesy that infuriated the popular, regular, feisty, knickered, provincial, depression-era little sons of bitches at Henry Cooke Public School on Thirteenth Street N.W.

In 2005, Hans Berliner was interviewed for the Computer History Museum at his home in Riviera Beach, Florida by Gardner Hendrie. A video recording of that interview and a full transcription are available online (Berliner, 2005; Hendrie, 2005). About his early years in Washington D.C. and his interest in science, and how that directed him to chess, he said:

When I was 11 and 12, I read every book in the adult section of the library about astronomy. And I had a feeling that I knew a very great deal about astronomy. However, what’s a 12-year-old kid going to do with a phenomenal knowledge of astronomy? There was no school that could handle you. There was no special place to go. And so here I was, and I even was at the point where there were some theories about the formation of the solar system that were patently wrong. And I knew that, because they violated certain physical things that I knew about. School was a piece of cake. School was always very easy. And then into this vacuum comes chess, which was a place where a kid could all of a sudden compete in the adult world. So from that point on I was hooked on chess.

Hans Berliner became a chess master at the age of 20. In 1968, he won World Correspondence Chess Championship with the unprecedented score of 14/16: 12 wins and 4 draws. No WCC champion before or since has ever shown a better performance. Winning the fifth Correspondence Chess World Championship made him an ICCF Grandmaster. In 1990, Hans Berliner was inducted in the U.S. Chess Hall of Fame. The accompanying dedication mentions that ‘he defeated Russian Correspondence Chess Grandmaster Yakov Estrin using the Two Knights Defense in what remains one of the most famous and influential correspondence chess games in history.’ (World Chess Hall of Fame, 1990)

1. From Chess to Computer Chess

After reaching the ultimate summit, Hans Berliner retired from correspondence chess with a record of 94 wins, 1 loss, and 10 draws, and turned his attention to computer chess. In 1969, he went to Carnegie Mellon University to study computer science, and in March 1974 he completed his PhD Thesis entitled Chess as Problem Solving: The Development of a Tactics Analyzer (Berliner, 1974). His thesis advisors were Allen Newell and Herbert Simon, pioneers in the field of artificial intelligence who jointly received the Turing Award (‘The Nobel Prize for Computer Science’) in 1975. In 1978 Herbert Simon was awarded the Nobel Prize in Economics ‘for his pioneering research into the decision-making process within economic organizations.’ Not every PhD student has the luxury of having Turing Award and Nobel Prize laureates as their advisors. When Alan Newell died in 1992, Hans wrote a touching obituary in the ICCA Journal: ‘He was my friend and my mentor, the most significant force in my life. His passing has created a void in me and at Carnegie-Mellon University that will be impossible to fill.’ (Berliner, 1992) Nine years later, when Herbert Simon passed away, Hans wrote an equally kind and moving obituary (Berliner, 2001). It deeply saddens me that, when Hans Berliner died in 2017, nobody wrote an obituary or an In Memoriam for him in the ICGA Journal. I hope that this contribution makes up for that grave injustice.

The opening statement of the abstract of his PhD thesis (Berliner, 1974) reads:

This thesis concerns itself with progress that has been made in the development of a better model of computer chess. We consider the fact that chess programs have made almost no gain in strength, as measured on the human scale, in the period 1968–1973, as indicative that the popular model of computer chess is near the limits of its exploitability.

A warning that the future of computer chess would depend on a different approach. At that time, the Elo-rating of the best computers was around 1600, so there was still a very long way to go to reach master level (2300+), let alone grandmaster level (2500+) performance. In 1981, Ken Thompson and Joe Condon won the Fredkin Prize when their chess machine Belle achieved master status. Eight years later, the team behind Deep Thought won the Fredkin Prize for reaching Grandmaster level play (Berliner, 1989).

The shift from software-implemented chess programs to hardware-based chess machines was a breakthrough in computer chess strength. The first successful chess machine was Belle, developed by Ken Thompson and Joe Condon at Bell Labs. In 1976, the first incarnation of Belle was built as a software-based chess program that ran on a PDP-11 computer which was augmented with a hardware move generator. In 1978, a hardware evaluation function and a hardware-implemented transposition table were added. This second generation of Belle was capable of evaluating 5,000 positions per second. In 1980, Belle was further improved and additional chess functions, such as pawn structure evaluation, were implemented in hardware. This final version of Belle evaluated 100,000–200,000 positions per second, at that time a phenomenal achievement. Belle won the ACM North American Computer Chess Championship five times (1978, 1980, 1981, 1982 and 1986) and became Computer Chess World Champion in 1980.

Around that time, Hans Berliner and his PhD student Carl Ebeling, with the help of several other PhD students, built the Hitech chess machine. Hans was convinced that a fast hardware implementation of his profound chess knowledge and experience could provide Hitech with an important advantage over chess machines built by people without great chess skills. Carl Ebeling was responsible for building the hardware which featured 64 VLSI circuits (one for each square on the chessboard) which operated in parallel (Ebeling, 1986; 1987). One of the key elements in Hitech was the tradeoff between speed and knowledge. As Ebeling described it in the introduction of his PhD thesis:

[ $\dots$ ] over the past decade, computer chess has become the province of the fast brute-force searching programs, primarily because we understand better how to use computers for fast, simple computation than for complex knowledge-based understanding. Yet it is the opinion of many chess experts that searching speed alone is not sufficient to produce top quality chess play – for computers to penetrate the top levels of human play, they must both search fast and know a lot.

Hitech analyzed about 200,000 positions per second, about as much as Belle did, but with much more chess knowledge implemented. The first time Hitech played in the annual ACM North American Computer Chess Championship (Denver, 1985), it won with perfect score of 4/4. Hitech did not play in the next two editions of the ACM tournament, and when it participated in the 1988 championship in Orlando, FL, it finished in fifth place. It looked like the baton of computer chess domination was passed on from Hitech to Deep Thought, the chess machine developed by PhD students from Carnegie Mellon University.

ChipTest, the predecessor of Deep Thought, was developed in 1985 by Feng-Hsiung Hsu and Thomas Anantharaman. It was basically the Belle chess machine miniaturized onto a single VLSI chip. The 1987 version of ChipTest evaluated about 500,000 positions per second. It won the 1987 ACM tournament while an early version of its successor Deep Thought won the 1988 edition of the championship. In 1986, former Hitech member Murray Campbell had joined the team to create Deep Thought. ‘The king is dead, long live the king!’ However, Hitech appeared to have one last trick up its sleeve. In the 1989 ACM North American Computer Chess Championship in Reno, NV, Hitech finished tied for first place together with Deep Blue. Because Deep Blue had defeated Hitech in the third round, it seemed that Deep Blue was the superior chess machine. However, Deep Blue had lost its game in the final round against Mephisto X, a commercial chess program running on a single 68030 microprocessor. In the Communications of the ACM, computer chess veteran and professor at McGill University in Montreal, Canada, Monty Newborn summarizes this battle of the kingdom as follows: ‘Despite entering ranked almost a class above the field, a last-round loss forced Deep Thought to settle for a first-place tie with Hitech at the 20th Annual ACM North American Computer Chess Championship.’ (Newborn, 1990). However, Hans Berliner saw it differently. In a message titled ‘Hitech wins International Chess Tournament; beats former World Candidate’ which he posted on the internet forum rec.games.chess, he describes how Hitech beat the legendary David Bronstein in the 1989 Aegon Man-Computer Tournament in The Hague, the Netherlands. Bronstein, who is ‘widely considered to be one of the greatest players not to have won the World Championship’ (Wikipedia, 2023) was 65 years old at the time. Nevertheless, Hans proudly celebrated the victory as ‘the strongest all-time player ever to be defeated by a computer.’ An innocent-looking remark in the same post sparked a heated debate full of controversy and venom: ‘in November [Hitech] took first place in the North American Computer Chess Championship Tournament on tie-break over Deep Thought’ (Berliner, 1990a). Feng-Hsiung Hsu responded: ‘Hitech SHARED first place. Deep Thought won the individual encounter against Hitech decisively for the third time, but lost to Mephisto in a difficult positional game, which I am sure they would keep on reminding us in their various ads. There was no stated tie-breaking system for the event, and Deep Thought should have retained the sole possession of the title.’ (Hsu, 1990). Hans Berliner replied: ‘There WAS a stated tie-breaking system in this tournament and it involved breaking the tie according to the Sonneborn-Berger system. You may choose to conveniently FORGET such niceties, but that is how Hitech got the first-place trophy which is on display in the Hitech lab. The fact that Deep Thought was lost in almost every game it played in this event before finally being nailed by Mephisto is, I suppose, also something that can be conveniently forgotten.’ (Berliner, 1990b). The discussion goes back and forth a few more times, but it is clear that there is no love lost between the father of Hitech and the man behind Deep Thought.

In 1988, Hitech finished tied for first place with Cray Blitz, Bebe and Sun Phoenix in the triannual World Computer Chess Championship in Cologne, Germany. The ‘sum of opponent scored’ was used as a tiebreaker and resulted in Cray Blitz winning the title with Hitech relegated to third place. It was the closest that Hans Berliner came to capturing a second world title, this time for his Hitech chess machine. In subsequent world championships, Hitech was past its prime and finished fourth (Edmonton, 1989), seventh (Madrid, 1992) and eighth (Shatin, 1995).

The aforementioned Aegon Man-Computer Tournament was organized by the Duch Computer Chess Association (CSVN) and hosted by the Aegon insurance company in The Hague, the Netherlands. Humans and computers competed against each other in a six round Swiss-system tournament (with the exception of 1986 and 1989 when there were seven and five rounds, respectively). Hitech participated in four consecutive editions (1990–1993). On two occasions (1991 and 1993), Hans Berliner came to The Netherlands to operate Hitech, while I babysat the chess machine that ‘eats International Masters for breakfast’ (Berliner, 1988) in the absence of its father in 1990 and 1992. The 1990 Aegon Tournament saw 14 computers and 14 humans compete. While the humans defeated the computers 47–37, Hitech won the tournament. It played the two grandmasters who participated that year, drawing against Jeroen Piket and defeating David Bronstein. It was the only time in its twelve-year history that a computer won the Aegon Tournament. In 1991, Hitech finished 24^th in a field of 20 computers and 20 humans (five grandmasters), 16^th in 1992 (24+24, four GMs), and 25^th in 1993 (32+32, three GMs).

2. Other interests and The System

Throughout his life, Hans Berliner was fascinated by science – ‘One of my favorite subjects has always been chemical synthesis’ (Berliner, 2005) – and genuinely interested in problem solving and the challenges that it offered. Chess and computer chess were the most notable fields to which he applied his problem-solving skills and excelled at the highest level. While working on the problem of how a computer should evaluate and compare the immense number of board positions in the chess search tree, he decided to tackle a game in which good and bad positions are much easier to distinguish (Berliner, 1980):

What I wanted was a domain where it is possible to compare two situations and make a judgment about which one is the better without having to worry about the exhaustive analysis that chess positions require. When I became acquainted with backgammon, I realized it had the desired properties, and I set out to develop a backgammon program.

While evaluating board positions in Backgammon is much easier than in Chess, the game itself is far from trivial (Berliner, 1977):

The thing that makes backgammon an interesting object of study of AI is that in any given position (of which are 10²⁰), there are 21 possible combinations that the throw of two dice can produce. Each of these can be played legally in the average board position about 40 different ways. Thus if one were to investigate a backgammon position by tree searching, it would be necessary to deal with a branching factor of more than 800 (!!) at every node.

In 1979, his Backgammon program BKG 9.8 defeated world champion Luigi Villa by 7–1 in match played in Monte Carlo. ‘It was the first time a computer program had beaten a world champion at any board or card game.’ (Berliner, 1980)

Another challenge that fascinated Hans Berliner, was a solitaire card puzzle which he called Superpuzz (Berliner and Goetsch, 1985), the origin of which goes back to his early childhood (Berliner, 2005):

Way back as a little kid in Germany there was a game that I think I knew as ‘Patience’ or something. It’s a card game played with a four by thirteen raster of cards that were laid out four deep and thirteen across. And there were certain rules for moving – you took up all the Aces which left four blanks and then you moved the cards around and then presumably you eventually got to the point where there were no legal moves and you were allowed to deal the cards again and see if you could finish in three. This was an infantile creation, but there was a real game there. I knew I couldn’t play four by thirteen because that was too much, but I started playing four by six, and four by seven, four by eight with computers. And computers could solve the four by six game almost all the time it would win, but no human being could. It was a combinatorial thing, seeing sequences that no human being could.

Between 1925 and 1927, Aron Nimzowitsch published his views on chess strategy and positional play in five brochures that were collectively published as his famous chess book Mein System, which was translated in English in 1929 as My System (Nimzowitsch, 1925; 1987). He identified eight ‘Elements’ of chess strategy: (1) The center; (2) Play on open files; (3) Play on the seventh and eighth ranks; (4) Passed pawns; (5) The pin; (6) Discovered checks; (7) Exchanging; (8) The pawn-chain. In 1999, Hans Berliner published his views on the perfect opening for White in a book with a very similar title: The System (Berliner, 1999). While Nimzowitsch left room for discussion by calling it his system, Hans was convinced that it was not just his system, but the system. His adversaries (there were many) accused him of arrogance, but I think it was merely the scientific attitude that Hans applied throughout his career. Just like Einstein published the theory of general relativity, not his theory of general relativity. That scientific approach is also evident from the Introduction section of The System:

I liken The System to Mendeleev’s Periodic Table of atomic elements. At the time this was formulated in the mid-nineteenth century, many elements were as yet unknown and the valences and atomic weights of others were in doubt. Yet, he was able to show a certain pattern for elements of valence +1, 0, and −1 [ $\dots$ ] Later, with the discovery of much missing data, the theory was confirmed and now is established fact.

The premise of The System is that White has the benefit of the first move and should consolidate that advantage:

The System is a theory of how to integrate board control and development into a unified whole. At the start of a game of chess, White is ahead in development by half a move by dint of his being ahead 1 move after his turn, and 0 moves after Black has moved. The broad question is whether White can turn this small advantage to account as the game goes on. If White plays passively, Black will be able to mimic his moves, and after 7 to 10 moves, being to move will no longer constitute as much of an advantage as it does at the start of the game. Therefore, White must do something to wring concessions from Black even as they are both trying to complete their development and compete for control of the board. This is what The System is about, and this book is about how to do this.

Like the eight Elements of Nimzowitsch, Hans Berliner defines nine Principles: (1) Tactics is King; (2) Piece Location; (3) Development; (4) Board Control; (5) Options; (6) Response Pairs; (7) Transpositions; (8) Resolution; (9) When there are no Threats. Each Principle is explained in half a page or less, and illustrated with key moments from various chess games, including the Rook ending in his famous game against Estrin in the Final of the WCCC in 1968.

3. The Kasparov – Deep Blue matches

While the playing strength of computers steadily increased, the holy grail of computer chess – beating the human world champion in a match under tournament conditions – appeared on the horizon. Who would make that happen? Monty Newborn (2003) described the rivalry between the strongest chess computers:

Although Deep Thought had established a clear lead over competing chess programs, these, too, were moving forward at impressive rates. If the Deep Thought team’s ultimate goal was to defeat Kasparov, to remain ahead of the silicon competition in the meanwhile was a second and almost equally important goal. The Deep Thought team was not the only one with its sights on the world champion. Robert Hyatt’s Cray Blitz and Hans Berliner’s Hitech were a notch weaker than Deep Thought, and both Hyatt and Berliner were talented individuals. Neither’s program could be ruled out in the race to challenge the world champion. Berliner would have given his left arm to be the first; he had invested the better part of his adult life aiming for this opportunity.

In 1989, Feng-Hsiung Hsu and Murray Campbell joined the IBM Thomas J. Watson Research Center in Yorktown Heights, NJ, to continue their work on Deep Thought, which was renamed Deep Blue in honor of its new home. A match between Deep Blue and Gary Kasparov took place in Philadelphia in February 1996. Despite losing the first game, Kasparov won the match 4–2 (three wins, two draws, one loss). Hans Berliner attended the event and even commented on the games, but deep in his heart he wished that a super-duper upgraded version of Hitech was playing Kasparov. Instead, it was the brainchild of two of his former students that claimed its place in the spotlight. But at least he still had both of his arms.

The following year, a rematch took place in New York. After five games, the score was 2½–2½ (one win, one loss and three draws each) with one more game to go. That game lasted just 19 moves because Kasparov made a very unusual 7^th move to which Deep Blue responded almost immediately with a Knight sacrifice. The match was over and Deep Blue beat Kasparov 3½–2½. A historic moment of first-man-on-the-Moon proportions.

Monty Newborn published a book describing the long journey from the first ideas about programming a computer to play Chess (by Alan Turing and Claude Shannon around 1950) to the Deep Blue – Kasparov match in Philadelphia in 1996. Hans Berliner was asked by the American Mathematical Society to review the book (Berliner, 1998). His opening statement is vintage Hans:

This book was published after the 1996 match between Kasparov and Deep Blue, but before the 1997 match in which Kasparov was defeated. Given the hype and speculation following that event, I find it difficult to believe that anything I say in this review will persuade anyone to either buy this book or pass on it. However, here goes.

His main criticism was that Newborn was in a hurry to get his book out and did not wait until the after rematch:

The real story of Computer Chess Comes of Age should be written about what the Deep Blue team did between the matches of 1996 and 1997. I hope this book will be written some day. In the meantime, I will give Newborn’s book a 7 on a scale of 1 to 10 and hope that someone who knows more about chess will revise it.

The book that Hans wished for was written four years later by Feng-Hsiung Hsu (2002) who knows considerably less about chess than Newborn.

Fig. 1.

Hans Berliner at the invitation of Yasser Seirawan (left) commenting on the Kasparov-Deep Blue match in Philadelphia in 1996. (photo: Dap Hartmann).

Fig. 2.

A gathering of computer chess minds at the Kasparov-Deep Blue match in Philadelphia in 1996. FLTR: Jonathan Schaeffer, David Levy (background), Feng-Hsiung Hsu, Sanjoy Mahajan and Hans Berliner. (photo: Dap Hartmann).

4. Retirement

In 1997, Hans Berliner received the Allen Newell Medal for Research Excellence for his work in computer chess. The following year, he retired from Carnegie Mellon University and moved to Florida. We still exchanged occasional e-mails in which we discussed chess, computer chess, problem solving, science, life, the universe and everything. In November 2003 Hans wrote:

I have been reading some articles about the current big problems in astrophysics, and I have thought about them as hard as I could, and believe I have a very important answer. Now, I realize that for someone like me, who is way out of the mainstream of this subject to say this, is asking for a lot of laughter. But I have applied all I know to the problem, and my solution is not only holding up, but it gathers strength all the time. So, what I need to do is to bounce it off someone who knows a lot more about the subject than I do, which is where you come in. I have a writeup which is almost finished, but I keep finding ways of saying things simpler and better. Also, I am a little worried about the fact that ideas cannot be patented, and the only way to get credit is to actually publish this paper which would be difficult for a non-member of a society. What I think I will do is call you at home and talk about this. Then maybe I will find out that this has all been rejected before or something else is wrong, but – we’ll see.

A few days later, we talked on the phone but, unfortunately, I do not remember to which big problem in astrophysics he had an important answer, and he never sent me his writeup. But I do recall that he was mistaken, and he kindly thanked me for setting him straight. Of course, I wish that he had come up with an important new insight into astrophysics, as a last spark from his brilliant mind.

Hans Berliner died on Jan 13, 2017 in River Beach, FL, aged 87.

He was a very intelligent, highly demanding, but most of all a truly kind Mensch.

References

Berliner, H.J. (1974). Chess as Problem Solving: The Development of a Tactics Analyzer. PhD Thesis, Carnegie-Mellon University. Available at https://apps.dtic.mil/sti/pdfs/AD0784881.pdf (visited: 2023.10.16).

Berliner, H.J. (1977). BKG, a Program That Plays Backgammon. https://www.bkgm.com/articles/Berliner/BKG-AProgramThatPlaysBackgammon/index.html (visited: 2023.10.16).

Berliner, H.J. (1980). Computer Backgammon. Scientific American, 242(6), 64–72. doi:10.1038/scientificamerican0680-64.

Berliner, H.J. (1988). Re: More AEGON tourny info. Personal communication. e-mail to the author dated Dec 1, 1988.

Berliner, H.J. (1989). Deep Thought Wins Fredkin Intermediate Prize. AI Magazine, 10(2), 89.

Berliner, H.J. (1990a). Hitech wins International Chess Tournament; beats former World Candidate. Message on rec.games.chess, dated May 25, 1990, 11:08:48 PM https://groups.google.com/g/rec.games.chess/c/RXQGrtdw9r4/m/A3ndE8WpP5kJ (visited 2023.10.16).

Berliner, H.J. (1990b). Message on rec.games.chess, dated: May 26, 1990, 2:03:39 PM https://groups.google.com/g/rec.games.chess/c/RXQGrtdw9r4/m/yaoUSpPrrAMJ (visited 2023.10.16).

Berliner, H.J. (1992). Allen Newell 1927–1992, in Memoriam. ICCA Journal, 15(3), 144–145.

Berliner, H.J. (1998). Kasparov Versus Deep Blue: Computer Chess Comes of Age. Notices of the AMS, 45(4), 503.

10.

Berliner, H.J. (1999). The System. A World Champion’s Approach to Chess. Gambit Publications Ltd. ISBN 1-901983-10-2.

11.

Berliner, H.J. (2001). Herbert A. Simon (1916–2001): A Life’s Appraisal. ICGA Journal, 24(1), 60–61. doi:10.3233/ICG-2001-24117.

12.

Berliner, H.J. (2005). Oral History. Interview for the Computer History Museum by Gardner Hendrie. Video available at https://www.computerhistory.org/chess/orl-43343bb768f00/ (visited 2023.10.16).

13.

Berliner, H.J. & Goetsch, G. (1985). A Study of Search Methods: The Effect of Constraint Satisfaction and Adventurousness. In Proceedings of the 9th International Joint Conference on Artificial intelligence (Vol. 2, pp. 1079–1082). Available at https://www.ijcai.org/Proceedings/85-2/Papers/083.pdf (visited: 2023.10.16).

14.

Ebeling, C. (1986). All the Right Moves. A VLSI Architecture for Chess. PhD thesis, Carnegie-Mellon University. 1986.

15.

Ebeling, C. (1987). All the Right Moves. A VLSI Architecture for Chess. MIT Press. ISBN 0-262-05035-8.

16.

Fuentes, C. (1994). How I Started to Write. In The Art of the Personal Essay: An Anthology from the Classical Era to the Present. (edited by Philip Lopate), Doubleday 1994. ISBN 0-385-42298-9.

17.

Hendrie, G. (2005). Oral History of Hans Berliner. Transcription of the video interview (Berliner, 2005). http://archive.computerhistory.org/projects/chess/related_materials/oral-history/hans_berliner.oral_history.2005.102630824/berliner.oral_history_transcript.2005.103630824.pdf (visited 2023.10.16).

18.

Hsu, F.-H. (1990). Message on rec.games.chess, dated May 26, 1990, 1:59:29 AM. https://groups.google.com/g/rec.games.chess/c/RXQGrtdw9r4/m/sQOy9Cp0Y7wJ (visited 2023.10.16).

19.

Hsu, F.-H. (2002). Behind Deep Blue. Building the Computer That Defeated the World Chess Champion. Princeton University Press. ISBN 0-691-09065-3.

20.

Lopate, P. (Ed.) (1994). The Art of the Personal Essay: An Anthology from the Classical Era to the Present. Doubleday. ISBN 0-385-42298-9.

21.

Newborn, M.M. (1990). The 20th annual ACM North American Computer Chess Championship. Communications of the ACM, 33(7), 92–104. doi:10.1145/79204.79211.

22.

Newborn, M.M. (1997). Kasparov Versus Deep Blue. Computer Chess Comes of Age. Springer. ISBN 978-1-4612-7477-3.

23.

Newborn, M.M. (2003). Deep Blue: An Artificial Intelligence Milestone. Springer. ISBN 978-1-4684-9568-3.

24.

Nimzowitsch, A. (1925). Mein System. Ein Lehrbuch des Schachspiels auf ganz neuartiger Grundlage (pp. 1925–1927). Berlin: Bernhard Kagan.

25.

Nimzowitsch, A. (1987). My System (reprinted ed.). B.T Batsford Ltd. ISBN 0-7134-5655-8.

26.

Wikipedia (2023). David Bronstein. https://en.wikipedia.org/wiki/David_Bronstein (visited 2023.10.16).

27.

World Chess Hall of Fame (1990). https://worldchesshof.org/hof-inductee/hans-berliner (visited 2023.10.16).