The secret life of I. J. Good. (English) Zbl 07708432

Zabell, Sandy

Stat. Sci. 38, No. 2, 285-302 (2023).

Summary: I. J. (“Jack”) Good was a leading Bayesian statistician for more than half a century after World War II, playing an important role in the post-war Bayesian revival. But his graduate training had been in pure mathematics rather than statistics (one of his doctoral advisors at Cambridge had been the famous G. H. Hardy). What was responsible for this metamorphosis from pure mathematician to applied and theoretical statistician? As Good himself only revealed in 1976, during the war he had initially served as an assistant to Alan Turing at Bletchley Park, working on the cryptanalysis of the German Naval Enigma, and it was from Turing that he acquired his life-long Bayesian philosophy. Declassified and other documents now permit us to understand in some detail how this came about, and indeed how many of the ideas Good explored and papers he wrote in the initial decades after the war, in fact, gave in sanitized form, results that had their origins in his wartime work. Drawing on these sources, this paper discusses the daily and very real use of Bayesian methods Turing and Good employed, and how this was gradually revealed by Good over the course of his life (including his return to classified work in the 1950s).

MSC:

62-XX

Statistics

Keywords:

Alan Turing; banburismus; Bayesian statistics; bletchley park; cryptanalysis; deciban; enigma machine; I. J. Good; weight of evidence

References:

[1]	ADAMS, J. F. (1985). Maxwell Herman Alexander Newman, 7 February 1897-22 February 1984. Biogr. Mem. Fellows R. Soc. 31 437-452.
[2]	ALEXANDER, C. H. O’D. (1945). Cryptographic History of Work on German Naval ENIGMA. HW 25/1, UK National Archives, Kew, London.
[3]	BANKS, D. L. (1996). A conversation with I. J. Good. Statist. Sci. 11 1-19. · Zbl 0955.01547 · doi:10.1214/ss/1032209661
[4]	BARNARD, G. A. (1946). Sequential tests in industrial statistics. Suppl. J. R. Stat. Soc. 8 1-21. · Zbl 0063.00211
[5]	BEESLY, P. (1977). Very Special Intelligence: The Story of the Admiralty’s Operational Intelligence Centre, \(1939-1945\). Hamish Hamilton, London. Reprinted in 2000 by Greenhill Books, London, with a new Introduction by W. J. R. Gardner, and a new Afterword by Ralph Erskine, and an updated Bibliography by Gardner and Erskine.
[6]	BRITTON, J. L., ed. (1992) Collected Works of A. M. Turing. Pure Mathematics North-Holland, Amsterdam. · Zbl 0751.01017
[7]	BUDIANSKY, S. (2000). Battle of Wits: The Complete Story of Codebreaking in World War II. The Free Press, New York.
[8]	BUDIANSKY, S. (2006). Colossus, codebreaking, and the digital age. In Colossus: The Secrets of Bletchley Park’s Codebreaking Computers (B. J. Copeland, ed.) 52-63. Oxford Univ. Press, Oxford.
[9]	COPELAND, B. J., ed. (2006). Colossus: The Secrets of Bletchley Park’s Codebreaking Computers Oxford Univ. Press, Oxford.
[10]	DEMING, W. E. and STEPHAN, F. F. (1940). On a least squares adjustment of a sampled frequency table when the expected marginal totals are known. Ann. Math. Stat. 11 427-444. · Zbl 0024.05502 · doi:10.1214/aoms/1177731829
[11]	FLOOD, J. L. (2011). Arthur Thomas Hatto: 1910-2010. In Biographical Memoirs of Fellows of the British Academy (R. Johnson, ed.) 10 172-198. Oxford Univ. Press, Oxford.
[12]	FRIEDMAN, W. F. (1922). The Index of Coincidence and Its Applications in Cryptography. Riverbank Laboratories, Geneva, IL. Department of Ciphers, Publication 22. Second ed., 1935, War Department, Office of the Chief Signal Officer, Government Printing Office, Washington, DC. Declassified and approved for release by NSA on 01-06-2014.
[13]	GOOD, I. J. (1950). Probability and the Weighing of Evidence. Charles Griffin & Company, Limited, London. · Zbl 0036.08402
[14]	GOOD, I. J. (1951). Random motion on a finite Abelian group. Proc. Camb. Philos. Soc. 47 756-762. · Zbl 0045.08103 · doi:10.1017/s0305004100027201
[15]	GOOD, I. J. (1953). The population frequencies of species and the estimation of population parameters. Biometrika 40 237-264. · Zbl 0051.37103 · doi:10.1093/biomet/40.3-4.237
[16]	GOOD, I. J. (1956). Some terminology and notation in information theory. Proc. Inst. Elec. Engrs. C. 103 200-204.
[17]	GOOD, I. J. (1958). The interaction algorithm and practical Fourier analysis. J. Roy. Statist. Soc. Ser. B 20 361-372. · Zbl 0086.12403
[18]	GOOD, I. J. (1960). Weight of evidence, corroboration, explanatory power, information and the utility of experiments. J. Roy. Statist. Soc. Ser. B 22 319-331. · Zbl 0122.36803
[19]	GOOD, I. J. (1961). Weight of evidence, causality and false-alarm probabilities. In Information Theory (Symposium, London, 1960) 125-136. Butterworths, Washington, DC.
[20]	GOOD, I. J. (1962). Analogues of Poisson’s summation formula. Amer. Math. Monthly 69 259-266. · Zbl 0106.02502 · doi:10.2307/2312938
[21]	GOOD, I. J. (1965a). A list of properties of Bayes-Turing factors. NSA Tech. J. 10 1-6. Issue 2 for 1965. Approved for Release by NSA on Appeal on 03-9-2011, FOIA Case #58820.
[22]	GOOD, I. J. (1965b). The Estimation of Probabilities. An Essay on Modern Bayesian Methods. MIT Press, Cambridge, MA. · Zbl 0168.39603
[23]	GOOD, I. J. (1967). On the principle of total evidence. British J. Philos. Sci. 17 319-321.
[24]	GOOD, I. J. (1968a). Corroboration, explanation, evolving probability, simplicity and a sharpened razor. British J. Philos. Sci. 19 123-143. · doi:10.1093/bjps/19.2.123
[25]	GOOD, I. J. (1969). Statistics of language. In Encyclopedia of Linguistics, Information and Control (A. R. Meetham, ed.) 567-581. Pergamon, London.
[26]	GOOD, I. J. (1971). 46656 varieties of Bayesians. Amer. Statist. 25 62-63. Letter in the December 1971 issue.
[27]	GOOD, I. J. (1973). The joint probability generating function for run-lengths in regenerative binary Markov chains, with applications. Ann. Statist. 1 933-939. · Zbl 0269.60005
[28]	GOOD, I. J. (1975). Explicativity, corroboration, and the relative odds of hypotheses. Synthese 30 39-73. · Zbl 0307.62003
[29]	GOOD, I. J. (1976). Early work on computers at Bletchley. Natl. Phys. Lab. Rep. Com. Sci. 82. Updated and reprinted in Annals of the History of Computing 1 (1979), 38-48.
[30]	GOOD, I. J. (1979). Studies in the history of probability and statistics. XXXVII. A. M. Turing’s statistical work in World War II. Biometrika 66 393-396. · doi:10.1093/biomet/66.2.393
[31]	GOOD, I. J. (1980). The contributions of Jeffreys to Bayesian statistics. In Bayesian Analysis in Econometrics and Statistics: Essays in Honor of Harold Jeffreys (A. Zellner, ed.) 21-34. North-Holland, Amsterdam. · Zbl 0444.62002
[32]	GOOD, I. J. (1982). A report on a lecture by Tom Flowers on the design of Colossus. Ann. Hist. Comput. 4 53-59.
[33]	Good, I. J. (1983). Good Thinking: The Foundations of Probability and Its Applications. Univ. Minnesota Press, Minneapolis, MN. · Zbl 0583.60001
[34]	GOOD, I. J. (1988). The interface between statistics and philosophy of science. Statist. Sci. 3 386-412. · Zbl 0955.62512
[35]	GOOD, I. J. (1992a). Introductory remarks for the article in Biometrika 66 (1979). In The Collected Works of A. M. Turing. Pure Mathematics (J. L. Britton, ed.) 211-223. North-Holland, Amsterdam.
[36]	GOOD, I. J. (1992b). The Bayes/non-Bayes compromise: A brief review. J. Amer. Statist. Assoc. 87 597-606.
[37]	GOOD, I. J. (1993). Enigma and Fish. In Codebreakers: The Inside Story of Bletchley Park (F. H. Hinsley and A. Stripp, eds.) 149-166. Oxford Univ. Press, Oxford. Corrected paperback edition, 1994.
[38]	GOOD, I. J. (2000). Turing’s anticipation of empirical Bayes in connection with the cryptanalysis of the Naval Enigma. J. Stat. Comput. Simul. 66 101-111. · Zbl 1054.62004
[39]	GOOD, I. J. (2006). From Hut 8 to the Newmanry. In Colossus: The Secrets of Bletchley Park’s Codebreaking Computers (B. J. Copeland, ed.) 204-222. Oxford Univ. Press, Oxford.
[40]	Good, I. J. and Toulmin, G. H. (1956). The number of new species, and the increase in population coverage, when a sample is increased. Biometrika 43 45-63. · Zbl 0070.14403 · doi:10.1093/biomet/43.1-2.45
[41]	GOOD, I. J. and TOULMIN, G. H. (1968). Coding theorems and weight of evidence. IMA J. Appl. Math. 4 94-105. · Zbl 0212.23402
[42]	GREENBERG, J. (2014). Gordon Welchman: Bletchley Park’s Architect of Ultra Intelligence. Frontline Books, London.
[43]	HALL, H. S. and KNIGHT, S. R. (1891). Higher Algebra: A Sequel to Elementary Algebra for Schools, 4th ed. Macmillan, London.
[44]	HATTO, A. T. (1965). The Nibelungenlied. Penguin Books Inc, New York.
[45]	HILTON, P. J. and WYLIE, S. (1967). Homology Theory: An Introduction to Algebraic Topology. Cambridge Univ. Press, Cambridge. · Zbl 0163.17803
[46]	HINSLEY, F. H. (1979-1990). British Intelligence in the Second World War. Her Majesty’s Stationery’s Office, London. Four volumes, Volume 3 issued in two parts.
[47]	HINSLEY, F. H. and STRIPP, A. (1993). Codebreakers: The Inside Story of Bletchley Park. Oxford Univ. Press, Oxford. Corrected paperback reprinting, 1994. · Zbl 0858.94002
[48]	KAHN, D. (1967). The Codebreakers: The Story of Secret Writing. Macmillan, New York. Second edition, 1996, Macmillan, New York.
[49]	KAHN, D. (1991). Seizing the Enigma: The Race to Break the German U-Boat Codes \(1939-1943\). Houghton Mifflin, Boston, MA. Second edition, 1998, Barnes and Noble, New York; 2012 edition, Naval Institute Press, Annapolis.
[50]	KAHN, D. (2010). How I discovered World War II’s greatest spy. Cryptologia 34 12-21. Reprinted in How I Discovered World War II’s Greatest Spy, and Other Stories of Intelligence and Code, CRC Press, Boca Raton, FL, 2014, Chapter 1.
[51]	MAHON, A. P. (1945). The History of Hut 8. HW 25/2, UK National Archives, Kew, London.
[52]	PLUMB, J. H. (1950). England in the Eighteenth Century. Pelican Books, London.
[53]	REEDS, J. A. et al. (2015). Breaking Teleprinter Ciphers at Bletchley Park. Wiley-IEEE Press, Piscataway, NJ.
[54]	REJEWSKI, M. (1981). How Polish mathematicians deciphered the Enigma. Ann. Hist. Comput. 3 213-234. Includes discussion at the end by Cipher A. Deavours and I. J. Good.
[55]	SEBAG-MONTEFIORE, H. (2000). Enigma: The Battle for the Code. Wiley, Hoboken, NJ.
[56]	SINGH, S. (1999). The Code Book: The Science of Secrecy from Ancient Egypt to Quantum Cryptography. Doubleday, New York. Very readable, but to be consulted with caution; see the review by Jim Reeds in the Notices of the American Mathematical Society 47 (2000) 369-372.
[57]	TURING, S. (1959). Alan M. Turing. Heffer, Cambridge.
[58]	TURING, A. M. (2012a). The applications of probability to cryptography. Unpublished paper, c. 1941, UK National Archives, HW 25/37, released 2012 on the centenary of Turing’s birth.
[59]	TURING, A. M. (2012b). Paper on the statistics of repetitions. Unpublished paper, c. 1941, UK National Archives, HW 25/38, released 2012 on the centenary of Turing’s birth.
[60]	TURING, D. (2021). X, Y & Z, 2nd ed. The History Press, Cheltenham, UK. First edition, 2018. The author, Sir John Dermot Turing, is a nephew of Alan Turing, and has—in addition to a very successful career in his own right—recently written several books on the history of cryptography in WWII..
[61]	WALD, A. (1945a). Sequential method of sampling for deciding between two courses of action. J. Amer. Statist. Assoc. 40 277-306. · Zbl 0063.08128
[62]	WALD, A. (1945b). Sequential tests of statistical hypotheses. Ann. Math. Stat. 16 117-186. · Zbl 0060.30207 · doi:10.1214/aoms/1177731118
[63]	Wald, A. (1947). Sequential Analysis. Wiley, New York. · Zbl 0029.15805
[64]	WELCHMAN, G. (1982). The Hut Six Story: Breaking the Enigma Codes. McGraw-Hill, New York.
[65]	WINTERBOTHAM, F. W. (1974). The Ultra Secret. Weidenfeld and Nicolson, London.
[66]	ZABELL, S. (2012). Commentary on Alan M. Turing: The applications of probability to cryptography. Cryptologia 36 191-214.
[67]	ZABELL, S. (2015). Statistics at Bletchley Park. In Breaking Teleprinter Ciphers at Bletchley Park (J. Reeds et al., eds.) lxxv-ci. Wiley-IEEE Press, Piscataway, NJ