Scalia's Poker: Puzzles and Mysteries in Constitutional Interpretation
O'Neill, Timothy P., Constitutional Commentary
Howard Hawks had some trouble directing The Big Sleep, the 1946 Humphrey Bogart/Lauren Bacall film noir based on Raymond Chandler's classic novel. He once convened a story conference with the film's three screenwriters, a group headed by Nobel Prize-winner William Faulkner. The four were having a terrible time trying to make sense of Chandler's labyrinthine tale. Flummoxed by one plot twist, Hawks fired off a telegram to Chandler reading "Who killed chauffeur?" The next day he received Chandler's response: "Damned if I know." (1)
Such an answer would have been inconceivable from a writer such as Agatha Christie. Her books are elaborately plotted, with every detail accounted for. There are no loose ends.
Christie and Chandler were both fine writers. But they worked in very different genres: Christie created puzzles, while Chandler created mysteries.
This point was brought home in a completely different context by Gregory Treverton, a Senior Consultant at RAND, in his book Reshaping National Intelligence for an Age of Information. (2) He traces the shift in the roles of the United States intelligence community from the Cold War to the present. During the Cold War, Treverton notes, the most pressing questions facing the intelligence community were "puzzles," i.e., questions "that could, in principle, have been answered definitively if only the information had been available." (3) He offers examples of Cold War puzzles: How big was the Soviet economy? How many missiles did the Soviet Union have? How much steel did the Soviet Union produce during the previous year? (4) These were all questions that could be definitively answered if one only had the right information.
Treverton differentiates these Cold War puzzles from what he calls "mysteries." A mystery is "a question that cannot be answered with certainty even in principle." (5) And he contends that today "most of the critical questions facing American foreign policy are mysteries." (6) He offers these examples: Will China continue to grow rapidly or will it fragment? Will reform and democracy take hold in the former Soviet Union? Where is South Africa headed? (7) These are mysteries because no one knows for certain what the answers will be. Unlike puzzles, which cannot be solved because of a lack of information, many mysteries these days ironically suffer from a surfeit of information. The problem is determining which parts of the mountain of available information are truly relevant.
This leads to another distinction: "Mysteries also differ from puzzles in that, by definition, puzzles have already happened," i.e., the Soviet steel has already been made and the missiles have already been built. (8) Mysteries are more subtle. Not only are they unknowable at this time, but their eventual answer is intertwined with events which have not yet occurred, such as what U.S. government policy will be next year.
Not everyone is equally adept at solving puzzles and mysteries. Malcolm Gladwell addressed this issue in a recent New Yorker article in which he characterized the Enron scandal as a mystery rather than a puzzle. (9) Gladwell notes that "Mysteries ... are a lot murkier [than puzzles].... [S]ometimes the question itself cannot be answered. Puzzles come to satisfying conclusions. Mysteries often don't." (10) He quotes Yale law professor Jonathan Macey that puzzles are "transmitter-dependent;" that is, their solutions turn on what information we are provided. Mysteries, on the other hand, are "receiver-dependent;" their solutions turn on the skills of the listener. (11)
What does all of this have to do with law? I would suggest that the distinction between puzzles and mysteries may describe a significant dichotomy in constitutional interpretation. Some justices view interpreting the Constitution as a puzzle--they seek a definite answer from the past by looking at information that will yield an objective answer. They see constitutional interpretation as "transmitter-dependent. …