Dr Strangelove’s Game
Published by the Penguin Group
Penguin Books Ltd, 80 Strand, London WC2R ORL, England
Penguin Putnam Inc., 375 Hudson Street, New York, New York 10014, USA
Penguin Books Australia Ltd, 250 Camberwell Road,
Camberwe]], Victoria 3 124, Australia
Penguin Books Canada Ltd, 10 Alcorn Avenue, Toronto, Ontario, Canada M4V 382
Penguin Books India (P) Ltd, i [ Community Centre,
Panchsheel Park, New Delhi -110017, India
Penguin Books (NZ) Ltd, Cnr Rosedale and Airborne Roads,
Albany, Auckland, New Zealand
Penguin Books (South Africa) (Pty) Ltd, 24 Sturdee Avenue,
Rosebank 2196, South Africa
Penguin Books Ltd, Registered Offices: So Strand, London WC2H OKI., England
First published by Hamish Hamilton 2001 Published in Penguin Books 2002
Copyright © Paul Strathern, 2001 All rights reserved
The moral right of the author has been asserted
Set in Caslon 540 (Times New Roman 12)
Printed in England by Clays Ltd, St Ives plc
Except in the United States of America, this book is sold subject to the condition that it shall not, by way of trade or otherwise, be lent, re-sold, hired out, or otherwise circulated without the publisher’s prior consent in any form of binding or cover other than that in which it is published and without a similar condition including this condition being imposed on the subsequent purchaser
Scanned: Mr Blue Sky
Proofed: It’s Nor Raining
Date 28/02/2002 V.2.0
Something Out Of Nothing Comes
The Richest Man in the World
The Founding Father
French Optimists and British Pessimists
Brave New Worlds
The Pleasure Principle
Workers of the World Unite
Measure for Measure
Into the Modern Age
Cometh the Hour, Cometh the Man
The Game to End All Games
Epilogue: The Game Goes On
‘The ideas of economists and political philosophers, both when they are right and when they are wrong, are more powerful than is commonly understood. Indeed, the world is ruled by little else. Practical men, who believe themselves to be quite exempt from any intellectual influences, are usually the slaves of some defunct economist. Madmen in authority, who hear voices in the air, are distilling their frenzy from some academic scribbler of a few years back.’
John Maynard Keynes
‘In economics there are some, even if not many, immutable laws - laws of an order of certainty of Calvin Coolidge’s possibly apocryphal dictum that when many people are out of work, unemployment results.’
J. K. Galbraith
Dr Strangelove had a black-gloved synthetic arm; his crippled body was confined to a wheelchair. Shaded glasses obscured his intense twisted features, and his high-pitched, strangulated voice had a menacing mid-European accent. This was the evil genius of the War Room, advising the US president on strategy as the world faced nuclear disaster, in Stanley Kubrick’s film Dr Strangelove, or How I Learned to Love the Bomb.
A sole American bomber has eluded the Soviet defence system, and is now beyond recall. With horror, the Soviet ambassador reveals to the president and his advisor the likely consequences: if the bomber succeeds in reaching its target, it is liable to trigger the Soviet Union’s ultimate weapon, the Doomsday Machine. This will release a vast cloud of radioactive material which will enshroud the entire Earth, destroying all human and animal life for 100 years. The president is aghast. Dr Strangelove is exasperated, and exclaims to the Soviet ambassador, ‘The whole point of hafing a Doomsday Machine is lost if you keep it a secret. Vy didn’t you tell the vurld?’
This is the logic of game theory - the first explicit reference to this new method of strategic thought in a popular movie. The entire notion of nuclear deterrence was based upon game theory.
Dr Strangelove, somewhat frenetically overacted by Peter Sellers, shows signs of growing insanity as the film approaches its nuclear climax. He rapidly conceives of a brilliant plan for human survival, his mechanical arm starts going out of control and attempts to strangle him, he makes agitated references to a ‘kompew-tah’. The evil genius degenerates into a parody of the mad scientist.
Most people considered Dr Strangelove to be a far-fetched creation, exaggerated for satirical purposes. The facts suggest otherwise. During the mid 1950s a mysterious figure with a Hungarian accent, his crippled body confined to a wheelchair, would be whisked by limousine from his bed at Walter Reed Hospital in Washington to the White House. Here, President Eisenhower, who had previously commanded the entire Allied forces in Europe during the Second World War, would listen intently to the suggestions of his secret strategic advisor, a man who had never even been in the army. The meeting over, the wheelchair-bound figure would be sped back to his hospital room. Here, two armed guards were posted at his door night and day, and he was attended only by naval nurses with top-security clearance. The patient was becoming increasingly deranged and would frequently wake shrieking and babbling in the night. His military minders were there to ensure that any secrets he blurted out would not find their way to a foreign power.
This was John von Neumann, a brilliant Hungarian who was responsible for major breakthroughs in several intellectual fields, from pure mathematics to practical economics. Indeed, in 1944 he was convinced that he had found a method which ‘solved’ economics. From then on ‘wise’ economic choice would simply be a matter of mathematical calculation. The entire process of economic decision-making could be left to computers (which he also helped to invent), and economists of the human variety would thus become redundant. This same man - arguably the finest mathematical mind of the twentieth century - also proposed even more far-reaching uses for his radical new method, which he called game theory. Here was a theory which showed not only how to banish for ever economic uncertainty, but also how to rule the world by nuclear force.
John von Neumann was born the son of a rich banker in Budapest in 1903, one of the brilliant Hungarian generation which was to produce figures ranging from George Soli to Zsa Zsa Gabor. Von Neumann was quickly recognised as an infant prodigy. By the age of six he could read a page from the Budapest telephone directory once and immediately repeat it from memory. By the time he was eight he could divide two eight-digit numbers in his head. (Try dividing 97,572,915 by 18,835,769 on paper.) Before the age of thirty, von Neumann wrote what came to be regarded as the definitive textbook on quantum mechanics. But this work contained von Neumann’s first crucial mistake: an erroneous proof. Yet by now his reputation was so great that the few who spotted this error felt sure they must have overlooked something. The force of von Neumann’s logic was incontestable, but it had been based upon an unjustified assumption. As we shall see, this was to become a characteristic flaw. (Von Neumann’s ‘proof’ would hold back a particular aspect of quantum mechanics for over half a century.)
In 1928 von Neumann came up with a theory which was to transform the long history of mathematical probability. This would become known as game theory. Its intention was to reduce any two-person contest to a precise mathematical game. A player’s alternatives in a poker game could be assessed according to the mathematical probability of their different outcomes. However, game theory was about more than games; it could also be applied to reality. In von Neumann’s words, ‘Real life consists of bluffing, of little tactics of deception, of asking yourself what is the other man going to think I mean to do. And that is what games are about in my theory.’ Game theory was about conflict between two highly intelligent and deceitful partners, conjoined within certain rules. One player could never be sure whether the other was double-crossing.
Von Neumann’s continuing obsession with game theory may well have been linked to his compulsive sex drive. According to his biographer, Steve J. Heims, ‘some of his colleagues found it disconcerting that upon entering an office where a pretty secretary was working, von Neumann habitually would bend way over, more or less trying to look up her dress’. Von Neumann continued womanising throughout his two marriages. Both of his wives were powerful and intelligent women, who proved quite capable of double-guessing what even a genius might get up to in his spare time. In a letter to his second wife after a particular misdemeanour had been discovered, von Neumann wrote guardedly, ‘I hope you have forgiven my modest venture in double-crossing.’
In the light of such behaviour, it is not surprising to find that the inventor of game theory took a somewhat cautious view of two-person games and conflicts. There was only one rational strategy: ‘defeat is inevitable if you aim to win rather than avoid losing’. Damage limitation was the object of the exercise. At every stage, you should work out each possible move that you can make, and then calculate the maximum possible loss that you could sustain if you made that move. You should then select the move which had the minimum maximum possible loss. This became known as minimax theorem — and though it resulted in one divorce, it seems to have minimised the possibility of a second maximum loss for its creator.
Von Neumann emigrated to the US in the 1930s. Not yet thirty, he was appointed alongside Einstein to the newly founded Institute for Advance Study at Princeton. The Institute, which was devoted exclusively to theoretical research, quickly became a Mecca for the finest scientific brains in Europe and America. While at the Institute, von Neumann would later play a leading role in the development of the first computers. Despite the ban on all practical work at the Institute, he managed to assemble in the boiler room a large prototype computer, an early version of the Mathematical Analyser, Numerator, Integrator and Calculator (known as MANIAC).
Von Neumann also made vital contributions to the development of the first nuclear bombs. As a result, he became a leading member of the US Atomic Energy Commission, advising the president on the use of the hydrogen bomb. The Cold War against Soviet Russia was now entering its most frigid period. Von Neumann saw this global conflict as an ideal opportunity for putting game theory into practice. Here was the two-person game to beat all two-person games. The application of game theory to this situation led von Neumann to but one conclusion, the logic of which was incontestable. The only possible course of action was to strike first. As von Neumann earnestly explained to the president, game theory dictated that he drop an H bomb on the Russians at once. The only way to maintain a winning position in the game was to destroy the Russians before they could develop their own H bomb. Any other course of action would be completely contrary to the logic of game theory. (Never before, even at home, had he come across such an opportune game situation.)
Once again, we can see that von Neumann’s logic is irrefutable. However, his assumptions appear to have been subjected to somewhat less rigid analysis. Regardless of von Neumann’s urgent insistence (‘bevor it is too late’), President Eisenhower continued to hesitate. Even his secretary of state, John Foster Dulles, became convinced by von Neumann’s logic. But still Eisenhower dithered. He found himself unable to counter the compelling argument put forward by these two wizards of global strategy. Despite this, the man who had won the Second World War in Europe couldn’t help feeling that something was amiss. Then the Russians announced that they also now had the H bomb. It was too late. Despite the illogic of its position, sanity had prevailed.
But game theory could be applied to situations that were more than just catastrophic (nuclear strategy, divorce) or trivial (poker, marriage). By its very nature, it referred to any human activity involving conflict. This meant that game theory could be applied to the most complex and vital human activity of them all, namely, economics. As early as 1939 von Neumann had been approached by another brilliant Austro-Hungarian member of the Zsa Zsa Gabor generation with just this view in mind.
Oskar Morgenstern had studied economics and philosophy in Vienna. This had led him to conclude, ‘I was an idiot [to have studied] this silly philosophy.’ His opinion of economics, or more precisely economists, was little better. He dismissed the work of the leading Austrian economist Friedrich von Hayek as ‘higher nonsense’. Despite this, he was willing to succeed Hayek as director of the celebrated Vienna Institute for Business Cycle Research. Morgenstern’s ability to antagonise all and sundry came to a head with the Nazi takeover of Austria in 1938. Though anti-Semitic, Morgenstern was viewed as ‘politically unbearable’ by the Nazis. He accepted a post on the economics faculty at Princeton, and remained in exile for the rest of his life. His American colleagues proved an equal disappointment: ‘Economists simply don’t know what science means. I am quite disgusted with all this rubbish.’
Not content with his low opinion of others, Morgenstern compounded this with delusions of his own grandeur. He claimed to be an illegitimate descendant of the German Kaiser Friedrich III, and kept a portrait of his ‘grandfather’ on prominent display in his apartment (even after America went to war with Germany). Morgenstern soon became a recognisable figure about Princeton. Wearing one of his tailor-made three-piece outfits, he took to riding on horseback through the streets.
In the words of the American economic historian Robert J. Leonard, Morgenstern was a man of ‘enormous intellectual ambition and limited theoretical ability’. He wished to mix only with those he considered to be his intellectual peers, and quickly gravitated to the Institute for Advanced Study. Here, the likes of Einstein, Godel and the resident Nobel Prize-winners soon learnt to avoid him. Not so von Neumann, who established a curious rapport with Morgenstern. It seems Morgenstern was willing to overlook von Neumann’s Jewishness - possibly on account of his titled ‘von’, as much as his intellectual pre-eminence. Morgenstern’s view of economics was shared by von Neumann, who declared, ‘economics is simply a million miles away from the state in which an advanced science is, such as physics.’
Morgenstern saw his chance. Here was just the man to make up for those minor theoretical deficiencies which so often seemed to hamper the full flowering of his exceptional intellect. If only he and von Neumann could collaborate on a project. Morgenstern began looking through von Neumann’s previous work and came across his paper on game theory. Yes, this was it! Here at last was the inspiration they had both been waiting for, he assured von Neumann. With his economic genius, and von Neumann’s mathematical acumen, they would together rescue economics from its neolithic state. They would transform it into an exact science, whose incontestable logic would leave no possible room for error. This would be done by the application of game theory.
In 1941 Oskar Morgenstern and his new pal ‘Johnny’ von Neumann began collaborating on a paper which demonstrated how game theory could be applied to economics. This paper soon grew into two papers, and then blossomed into a 100-page pamphlet. Despite war shortages, Princeton University Press was persuaded to publish — as soon as the authors completed their final draft. But by now Morgenstern and von Neumann had become obsessed with their topic, and the pamphlet soon began to expand into a book. At this stage von Neumann was doing stints at Los Alamos, where he was working on the Manhattan Project to build the first atomic bomb. He was also in demand as a consultant to the military and the government in Washington. But in between the distractions of advising how to win the war and devising how to split the atom, von Neumann would return to Princeton to continue with his collaboration. He and Morgenstern worked night and day. Von Neumann would rise while his wife was still asleep and meet Morgenstern at the Nassau Club. Here, they would review their previous work over breakfast and then sketch out further possibilities for the expansion of game theory into economic activity. Discussion would continue at Morgenstern’s apartment across Nassau Street above the Princeton Bank. Under the watchful eye of the Kaiser, Morgenstern would scribble intently in his pad as von Neumann gazed into the middle distance, spouting a succession of abstruse mathematical formulae. (The child who was capable of dividing eight-figure numbers in his head could now conjure up formulae which reduced all economic activity to a game.) In the afternoon they would retire to von Neumann’s house at 26 Westcott Road, so that he could spend some time with his wife Klari during his brief spell on leave from Los Alamos and Washington. Klari would serve coffee as the two collaborators continued with their conversation. The talk would continue over more coffee, over drinks, over dinner, and yet more coffee. Having been ignored for eight hours at a stretch, Klari would finally turn them out of the house, where they would walk the night-deserted streets of wartime Princeton until the early hours. Eventually Morgenstern noticed, ‘Klari was often rather distressed by our perpetual collaboration and incessant conversations.’ So at the weekends they would spare her this upsetting experience. Instead, ‘we drove occasionally to the seashore and walked up and down the boardwalk at Sea Girt, in particular, discussing matters’. In the end they all travelled over a thousand miles to the Texas coast on the Gulf of Mexico: ‘I was vacationing in Biloxi with Johnny and his wife Klari. Again, day after day was spent in discussing the theory... Incidentally, we always spoke German.’ One can only wonder what the locals made of all this. At the time the entire eastern seaboard was rife with rumours of Germans recording the movement of Atlantic convoys to Europe, and stories about enemy spies being put ashore from submarines. By April 1943 their work was at last complete. As Morgenstern explained, ‘The people at the Press were quite overwhelmed seeing a manuscript of about 1,200 typed pages full of graphs and uninhibited mathematical notations.’ Without apparent irony, its authors decided to call this magnum opus General Theory of Rational Behaviour. But eventually they felt this ‘was not descriptive enough of our work’, and it was re-titled Theory of Games and Economic Behavior. The work is prefaced by an assurance that ‘no specific knowledge of any particular body of advanced mathematics is required. However...’ A glance at the ensuing hundreds of pages of tightly packed formulae containing many ‘uninhibited mathematical notations’ should suffice to indicate whether your mathematics rises to this humble category. The work itself opens with a modest comparison between the application of game theory to economics and the effect of Newton’s discovery of gravity on physics. It passes on to analyse such diverse subjects as the ‘Robinson Crusoe’ economy, ‘The Adventures of Sherlock Holmes’ and ‘Poker and Bluffing’, coming to a close with ‘Economic Interpretation of the Results’ for markets consisting of two people, and extending even as far as markets ‘greater than three people’. At the end there is an index which includes such items as ‘psychological phenomena, mathematical treatment’; ‘winning’ has sufficient entries to include two sub-sections: ‘certainly’ and ‘fully’. There are just two entries under ‘losing’.
The publication of Theory of Games and Economic Behaviour in 1944 was greeted with ecstatic acclaim. The American Mathematical Society Bulletin described it as ‘one of the major scientific contributions of the first half of the twentieth century’ - thus placing it alongside relativity, quantum theory and the Keynesian economics which had combated world recession (though Morgenstern was of the opinion that ‘Keynes is a scientific charlatan, and his followers not even that’). More cautious critics like the economist Leonid Hurwicz still remained hugely optimistic: ‘Ten more such books and the future of economics is assured.’ The groundswell of opinion in favour of Theory of Games and Economic Behavior rose to a climax in March 1946, when the book made front-page news in The New York Times - on a par with Churchill’s declaration in the same month that an ‘Iron Curtain’ had descended across Europe. Many were convinced that game theory would become the foundation of all economic theory. Economics would be reduced once and for all to mere calculation. There would be no room for human argument: a decision would be either right or wrong.
Dr Strangelove’s game could have put an end to the world as we know it. Might it one day do something similar to economics? The answer lies in our definition of economics. Initially the subject consisted of little more than a series of insights concerning commercial activity - what exactly it was, how it worked, how to improve upon it. From such beginnings the study of economics gradually evolved. Over the centuries, this has generated an epic narrative, replete with a cast of varied and colourful personalities. But instead of a traditional story, with a central hero whose actions illustrate and develop his character, we have a central idea. This evolves in the hands of a succession of individuals, including some of the greatest thinkers of their age, as well as a motley collection of wayward geniuses, moralists, eccentrics and charlatans. Some have attempted to save the world, and might have destroyed it. Some have seen themselves simply as mechanics, doing their best to keep the engine running. Others have conjured up Utopias or apocalyptic visions. A number have been drawn by compassion to try to remedy the horrific reality they saw unfolding before them. The ever-evolving economic idea has not always resulted in progress - far from it. Yet it has led to a deepening social self-understanding, and a growing insight into how society works. From an initial series of insights, economics has now spread into every facet of our lives. It has made us increasingly aware of who we are and what it is we are doing. What follows is a narrative of how this happened, together with the lives and ideas of those who helped to make it happen.
Something Out Of Nothing Comes
That telling cultural symbol, the zero, first arrived in Europe from the Levant around 1200. Prior to this, calculation had best been carried out on the abacus. With no concept of zero, it had been mathematically impossible to conceive of negative numbers. With the advent of the zero, more complex commercial calculations could be carried out on the page, and you could end up with a minus quantity. Accountancy as we know it had been born.
By the late thirteenth century merchant banks were conducting international trade throughout Europe and beyond. Ledgers in Brugge recorded batches of sealskins from Greenland to pay papal dues. Marco Polo (or the genuine sources he used to confect his memoirs) observed Genoese ships trading on the Caspian Sea. The first great European bankers were the Bardi family of Florence, who flourished from 1250. By the following century they had risen to become the sole papal bankers, holding a monopoly on the collection of papal revenues throughout the Continent. Such was their wealth, they could even afford to finance kings and wars. It was their support for the extravagances of Edward III, who led England into the Hundred Years War with France, which eventually precipitated their downfall. When Edward reneged on his colossal debts, there was nothing the Bardis could do about it. Commerce had money and influence, but as yet no real power. In 1345 the House of Bardi went bankrupt, contributing to an economic downturn throughout Europe.