Lies, Damned Lies, and Statistics Misquotations

What may be the most famous statistics quotation of all time is commonly attributed to Mark Twain: “There are three types of lies: lies, damned lies, and statistics.” But Twain attributed the saying to British Prime Minister Benjamin Disraeli:

Figures often beguile me, particularly when I have the arranging of them myself; in which case the remark attributed to Disraeli would often apply with justice and force: `There are three kinds of lies: lies, damned lies, and statistics.’ (Twain, 2010, p. 228).

But did Disraeli originate the quote? The late Peter Lee, in a highly entertaining exploration of the origin of the quote, wrote that the attribution to Disraeli was probably false. He found that numerous persons wrote or said the phrase in the nineteenth century, including Walter Bagehot, Arthur James Balfour, Cornelia Crosse, Charles Dilke, and Carroll D. Wright, but most of them said that the quote was “well known.”

“Lies, damned lies, and statistics” is a real quote, in that someone said it, but the attribution is questionable. Let’s look at some other statistics aphorisms that are misquoted or attributed to the wrong person. The last saying, in fact, was likely intended as an attack on Winston Churchill but ended up being attributed to him.

Statistical thinking will one day be as necessary for efficient citizenship as the ability to read and write.

Attributed to: Science fiction author H.G. Wells. Samuel S. Wilks (1951, p. 5) attributed this quote to Wells in his American Statistical Association presidential address about undergraduate statistics education. Since then the quote has appeared on many websites and in statistical talks, papers, and textbooks.

But Wells did not write the words “quoted” by Wilks. Wells (1903, p. 204) actually wrote: “The great body of physical science, a great deal of the essential fact of financial science, and endless social and political problems are only accessible and only thinkable to those who have had a sound training in mathematical analysis, and the time may not be very remote when it will be understood that for complete initiation as an efficient citizen of one of the new great complex world-wide States that are now developing, it is as necessary to be able to compute, to think in averages and maxima and minima, as it is now to be able to read and write.”

Wells’s verbose quote, in the chapter on Schooling, concluded a paragraph about the importance of teaching mathematics. He did not reference “statistical thinking” or “statistics” in the Schooling chapter. Elsewhere in the book, though, Wells demonstrated statistical thinking when he examined statistics of infant mortality. Wells also exhibited statistical thinking when he questioned the quality of statistics that had been collected about the relative achievements of only, elder, and younger children.

In God we trust: All others must bring data.

Attributed to: Statistician and quality expert W. Edwards Deming. This is a popular quote on blogs, including those from IBM and the UK Industry Forum.

I have read almost everything Deming ever published, however, and have never encountered this quote in his writings. Mary Walton (1988, p. 96) begins Chapter 20 of her book about Deming with the quote “In God we trust. All others must use data.” but she does not attribute the quote to Deming.

The saying was, however, found in an article by Brian Joiner (1985, p. 226), who was the first person selected to give the American Statistical Association’s Deming Lecture (in 1993) and who worked closely with Deming in Wisconsin. Did Deming say it? Maybe, but I cannot find proof that he did.

I only believe in statistics that I doctored myself.

Attributed to: Winston Churchill. The statistical mystery novel Data Games begins with this quote (Weisberg, 2020). The quote is attributed to Churchill in multiple places on the Internet, including Goodreads Quotes and AP News.

I enjoyed reading Data Games, but this did not sound to me like something Churchill would have said. Would the man who said he had “nothing to offer but blood, toil, tears and sweat” and exhorted Harrow students to “never give in, never give in, never, never, never, never—in nothing, great or small, large or petty—never give in except to convictions of honour and good sense,” really brag about doctoring statistics?

In fact, one of the first things Churchill did upon being reappointed First Lord of the Admiralty in 1939 was to establish an “S-Branch” (S for statistics) to help with prosecuting the war (Larson, 2020). Upon being told that there were already statisticians in all departments, Churchill wrote: “Surely the account you give of all these various disconnected Statistical Branches constitutes the case for a central body which should grip together all Admiralty statistics.… I want to know at the end of each week everything we have got … the progress of all vessels … all munitions affecting us, …our merchant tonnage, together with losses and numbers of every branch of the R.N. [Royal Navy]” (Nahum, 2022).

Churchill later wrote that statistics were essential for winning the Battle of the Atlantic: “It did not take the form of flaring battles and glittering achievements. It manifested itself through statistics, diagrams, and curves, unknown to the nation, incomprehensible to the public” (Churchill, 1949, p. 598-599). Churchill clearly viewed having accurate statistics as essential for science and governance.

I can find no evidence that Churchill ever spoke or wrote about doctoring statistics. The Consortium for the Advancement of Undergraduate Statistics Education stated that scholars at the International Churchill Society cannot find this quote in any of Churchill’s writings or speeches (although CAUSE does not give a reference for this assertion). Werner Barke, a researcher in the statistical office for the state of Baden-Württemberg, concluded after an extensive investigation that nothing speaks for the accuracy of this quote and everything speaks against it (Barke, 2004). The statistical office’s investigation found that every article attributing the quote to Churchill gave either no source or a second-hand source.

If the quote isn’t from Churchill, who said it and who attributed it to him? Barke (2004) suggested that the quote is entirely consistent with writings of Joseph Goebbels, minister of propaganda for the Third Reich, who repeatedly attempted to portray Churchill as a liar who exaggerated German losses. Goebbels began a November 1941 essay in his weekly newspaper Das Reich with the paragraph: “The British Prime Minister Winston Churchill, as is well known, is a close friend of alcohol. His relations to the truth are a bit more strained. He has been on a war footing ever since his entry into political life. He is one of the world’s best known liars. Not only do those in neutral and enemy nations smile when he says something, even knowledgeable circles in England cannot repress a grin. Everyone knows how he adds or subtracts, for example. At the moment he divides figures that are unfavorable for England by three, and multiplies the favorable ones by the same figure” (quoted from Bytwerk, 1998).

Goebbels doctored statistics all the time himself, and likely could not fathom that other people would not do the same thing. If the quote really was fabricated as part of Goebbel’s propaganda, the lie seems to have succeeded, as the quote is still regularly attributed to Winston Churchill.

Copyright (c) 2023 Sharon L. Lohr

References

Barke, W. (2004). “Ich glaube nur der Statistik, die ich selbst gefälscht habe.” Baden-Württemberg Statistisches Landesamt Veröffentlichungen.

Bytwerk, R. (1998). Goebbels on Churchill (1941). In German Propaganda Archive, Calvin University. Translation of “Der tönende Koloß” (the pompous giant) by Joseph Goebbels.

Churchill, W. (1949). The Second World War: Their FInest Hour. Boston: Houghton Mifflin.

Edgerton, D. (2011). Britain’s War Machine: Weapons, Resources, ad Experts in the Second World War. Oxford: Oxford University Press.

Joiner, B. (1985). The key role of statisticians in the transformation of North American industry. The American Statistician, 39(3), 224-

Larson, E. (2020). The Splendid and the Vile: A Saga of Churchill, Family, and Defiance During the Blitz. New York: Crown.

Mukerjee, M. (2010). The most powerful scientist ever: Winston Churchill’s personal technocrat. Scientific American, https://www.scientificamerican.com/article/the-most-powerful-scientist-ever/

Nahum, A. (2022). The PM and the boffins: Churchill and his scientists. The Finest Hour, 195.

Tankard, J. W. (1979). The H. G. Wells quote on statistics: A question of accuracy. Historia Mathematica, 6, 30-33.

Twain, M. (2010). Autobiography, Volume I. Berkeley, CA: University of California Press. For an online source, see Chapter 20 of Chapters of my Autobiography, which contains the text Twain dictated in Florence in 1904 and published in North American Review in 1907.

Walton, M. (1988). The Deming Management Method. New York: Penguin.

Weisberg, H. I. (2020). Data Games. Causalytic Books. See an interview with the author in Significance magazine.

Wells, H. G. (1903). Mankind in the Making. London: Chapman & Hall.

Wilks, S. S. (1951). Undergraduate Statistics Education. Journal of the American Statistical Association, 46(253), 1-16.