American Scientist

To the Editors:

In their Macroscope column “To Throw Away Data: Plagiarism as a Statistical Crime,” Andrew Gelman and Thomas Basbøll relate how the “When you are lost, any old map will do” meme originated in a passage lifted without attribution by Karl Weick from a poem by Miroslav Holub. Weick wrote in Sensemaking in Organizations that “in an equivocal, postmodern world, infused with the politics of interpretation and conflicting interests and inhabited by people with multiple shifting identities, an obsession with accuracy seems fruitless, and not of much practical help, either.” It is hardly surprising, then, that he and his adherents would dismiss an accusation of plagiarism as mere pedantry. Weick actively rejects the standards Gelman and Basebøll apply to him: Small wonder that he turns out to have violated those standards.

Gelman and Basbøll argue that Weick’s failure to cite a source for the example throws away data about the population from which it came. Holub’s poem relates a military anecdote from the point of view of a Hungarian officer and attributes it to Albert Szent-Györgyi, a Hungarian who served in the First World War.

War poetry’s a large genre, and contains other examples of “sensemaking” in a situation of confusion or uncertainty. If Weick denies the primacy of facts on grounds of postmodernism, perhaps literature is an appropriate substitute. Alfred Lord Tennyson’s “The Charge of the Light Brigade,” for example. At Balaclava, the troops followed a plan that was obviously wrong, but the Earl of Cardigan’s confidence convinced them. Should managers emulate him?

Siegfried Sassoon’s “The General” demonstrates that he understood that when he led a charge the troops had to believe the implicit “I know what I’m doing!” or else they’d hit the dirt and stay down:

“‘Good Morning, good morning!’ the general said / When we met him last week on our way to the line.
Now the soldiers he smiled at are most of ‘em dead / And we’re cursing his staff for incompetent swine.
‘He’s a cheery old card,’ grunted Harry to Jack / As they slogged up to Arras with rifle and pack....
But he did for them both by his plan of attack.”

Conclusion: “When you are lost, any old map will do. Except when it won’t.”

Scott Cooper
Cleveland, OH

To the Editors:

I was reading with great interest the May–June Macroscope column by Andrew Gelman and Thomas Basbøll titled: “To Throw Away Data: Plagiarism as a Statistical Crime,” when, into about the second page, the article started to strike me in a different way than I thought was its original intent. I went into it expecting to hear about quantitative data analysis and how plagiarizing can often produce statistical errors, such as when data are manufactured or copied by a scientist in order to reach a predetermined conclusion. However, I see nothing statistical at all about what is, essentially, a grousing piece. The column has a brief introduction that connects plagiarism to statistics in a very natural way. But then it devolves into accusing language that comes off as a complaint about plagiarism committed by one specific individual, namely one Karl Weick, throughout the rest of the article. I was expecting the Weick story to be one of many examples. The diatribe was bookended with fluff that is unrelated to what I thought was the original topic. The piece may as well have been titled, “Karl Weick Is a Bad Person,” if the bookends had been removed.

Brian Hart
San Diego, CA

Drs. Gelman and Basbøll respond:

We have no particular interest in Karl Weick as a person and we are puzzled by the suggestion that our article’s main message can be taken as a judgment about his character. In fact, it was precisely to avoid the more familiar moralizing about the plagiarist that we compared hiding a source to throwing away data rather than stealing someone’s property. In our article, we discuss the ways in which the conclusions we can draw from a story can be distorted when its source is obscured. The importance of correctly and honestly describing the data-generating process is well known in quantitative statistical analysis, and our key point was that these concerns arise in qualitative inference as well. Plagiarism, we want to emphasize, is not merely an offense against the person whose work has been misappropriated; it also does harm to the reader trying to make sense of a story and fit it into the larger world.