Say, it’s only a paper moon
Sailing over a cardboard sea
But it wouldn’t be make-believe
If you believed in me
—“It’s Only A Paper Moon” (1933).
As all of us sublunaries knows, we now live in a technological age where high-level training is required for anyone who prefers not to deal methamphetamine out of their trailer—or at least, that’s the story we are fed. Anyway, in my own case the urge towards higher training has manifested in a return to school; hence my absence from this blog. Yet, while even I recognize this imperative, the drive toward scientific excellence is not accepted everywhere: as longer-term readers may know, last year Michael Wilbon of ESPN wrote a screed (“Mission Impossible: African-Americans and Analytics”) not only against the importation of what is known as “analytics” into sports—where he joined arms with nearly every old white guy sportswriter everywhere—but, more curiously, essentially claimed that the statistical analysis of sports was racist. “Analytics” seem, Wilbon said, “to be a new safe haven for a new ‘Old Boy Network’ of Ivy Leaguers who can hire each other and justify passing on people not given to their analytic philosophies.” But while Wilbon may be dismissed because “analytics” is obviously friendlier to black people than many other forms of thought—it seems patently clear that something that pays more attention to actual production than to whether an athlete has a “good face” (as detailed in Moneyball) is going to be, on the whole, less racist—he isn’t entirely mistaken. Even if Wilbon appears, moronically, to think that his “enemy” is just a bunch of statheads arguing about where to put your pitcher in the lineup, or whether two-point jump shots are valuable, he can be taken seriously if he recognizes that his true opponent is none other than—Sir Isaac Newton.
Although not many realize it, Isaac Newton was not simply the model of genius familiar to us today as the maker of scientific laws and victim of falling apples. (A story he may simply have made up in order to fend off annoying idiots—a feeling with which, if you are reading this, you may be familiar.) Newton did, of course, first conjure the laws of motion that, on Boxing Day 1968, led William Anders, aboard Apollo 8, to reply “I think Isaac Newton is doing … the driving now” to a ground controller’s son who asked who was in charge of the capsule—but despite the immensity of his scientific achievements, those were not the driving (ahem) force of his curiosity. Newton’s main interests, as a devout Christian, were instead about ecclesiastical history—a topic that led him to perhaps the earliest piece of “analytics” ever written: an 87,000-word monstrosity the great physicist published in 1728.
Within the pages of this book is one of the earliest statistical studies ever written—or so at least Karl Pearson, called “the founder of modern statistics,” realized some two centuries later. Pearson started the world’s first statistics department in 1911, at the University College London; he either inaugurated or greatly expanded some half-dozen entire scientific disciplines, from meteorology to genetics. When Albert Einstein was a young graduate student, the first book his study group studied was a work of Pearson’s. In other words, while perhaps not a genius on the order of his predecessor Newton or his successor Einstein, Pearson was prepared to recognize a mind that was. More signifcantly, Pearson understood that, as he later wrote in the essay that furnishes the occasion for this one, “it is unusual for a great man even in old age to write absolutely idle things”: when someone immensely intelligent does something, it may not be nonsense no matter how much it might look it.
That’s what led Pearson, in 1928, to publish the short essay of interest here, which is about what could appear like the ravings of a religious madman, but as Pearson saw, weren’t: Newton’s 1728 The Chronology of Ancient Kingdoms amended, to which is prefixed: A Short Chronicle from the First Memory of Things in Europe to the Conquest of Persia by Alexander the Great. As Pearson understood, it’s a work of apparent madness that conceals depths of genius. But it’s also, as Wilbon might recognize (were he informed enough to realize it) it’s a work that is both a loaded gun pointed at African-Americans—and also, perhaps, a very tool of liberation.
The purpose of the section of the Chronology that concerned Pearson—there are others—was what Pearson called “a scientific study of chronology”: that is, Newton attempted to reconstruct the reigns of various kings, from contemporary France and England to the ancient rulers of “the Egyptians, Greeks and Latins” to the kings of Israel and Babylon. By consulting ancient histories, the English physicist compiled lists of various reigns in kingdoms around the world—and what he found, Pearson tells us, is that “18 to 20 years is the general average period for a reign.” But why is this, which might appear to be utterly recondite, something valuable to know? Well, because Newton is suggesting that by using this list and average, we can compare it to any other list of kings we find—and thereby determine whether the new list is likely to be spurious or not. The greater the difference between the new list of kingly reigns and Newton’s calculations of old lists, in short, the more likely it is that the new list is simply made up, or fanciful.
Newton did his study because he wanted to show that biblical history was not simply mythology, like that of the ancient Greeks: he wanted to show that the list of the kings of Israel exhibited all the same signs as the lists of kings we know to have really existed. Newton thereby sought to demonstrate the literal truth of the Bible. Now, that’s not something, as Pearson knew, that anyone today is likely much to care about—but what is significant about Newton’s work, as Pearson also knew, is that what Newton here realized was that it’s possible to use numbers to demonstrate something about reality, which was not something that had ever really been done before in quite this same way. Within Newton’s seeming absurdity, in sum, there lurked a powerful sense—the very same sense Bill James and others have been able to apply to baseball and other sports over the past generation and more, with the result that, for example, the Chicago Cubs (managed by Theo Epstein, Bill James’ acolyte) last year finally won, for the first time in more than a century, the final game of the season. In other words, during that nocturnal November moonshot on Chicago’s North Side last year, Sir Isaac Newton was driving.
With that example in mind, however, it might be difficult to see just why a technique, or method of thinking, that allows a historic underdog finally to triumph over its adversaries after eons of oppression could be a threat to African-Americans, as Michael Wilbon fears. After all, like the House of Israel, neither black people nor Cubs fans are unfamiliar with the travails of wandering for generations in the wilderness—and so a method that promises, and has delivered, a sure road to Jerusalem might seem to be attractive, not a source of anxiety. Yet, while in that sense Wilbon’s plea might seem obscure, even the oddest ravings of a great man can reward study.
Wilbon is right to fear statistical science, that is, for a reason that I have been exploring recently: of all things, the Voting Rights Act of 1965. That might appear to be a reference even more obscure than the descendants of Hammurabi, but in fact not so: there is a statistical argument, in other words, to be derived from Sections Two and Five of that act. As legal scholars know, those two sections form the legal basis of what are known as “majority minority districts”: as one scholar has described them, these are “districts where minorities comprise the majority or a sufficient percentage of a given district such that there is a greater likelihood that they can elect a candidate who may be racially or ethnically similar to them.” Since 1965, such districts have increasingly grown, particularly since a 1986 U.S. Supreme Court decision (Thornburg v. Gingles, 478 U.S. 30 (1986)) that the Justice Department took to mandate their use in the fight against racism. The rise of such districts are essentially why, although there were fewer than five black congressmen in the United States House of Representatives prior to 1965, there are around forty today: a percentage of congress (slightly less than 10%) not much less than the percentage of black people in the American population (slightly more than 10%). But what appears to be a triumph for black people may not be, so statistics may tell us, for all Americans.
That’s because, according to some scholars, the rise in the numbers of black congressional representatives may also have effectively required a decline in the numbers of Democrats in the House: as one such researcher remarked a few years ago, “the growth in the number of majority-minority districts has come at the direct electoral expense of … Democrats.” That might appear, to many, to be paradoxical: aren’t most African-Americans Democrats? So how can more black reps mean fewer Democratic representatives?
The answer however is provided, again perhaps strangely, by the very question itself: in short, by precisely the fact that most (upwards of 90%) black people are Democrats. Concentrating black voters into congressional districts, in other words, also has the effect of concentrating Democratic voters: districts that elect black congressmen and women tend to see returns that are heavily Democratic. What that means, conversely, that these are votes that are not being voted in other districts: as Steven Hill put the point for The Atlantic in 2013, drawing up majority minority districts “had the effect of bleeding minority voters out of all the surrounding districts,” and hence worked to “pack Democratic voters into fewer districts.” In other words, majority minority districts have indeed had the effect of electing more black people to Congress—at the likely cost of electing fewer Democrats. Or to put it another way: of electing more Republicans.
It’s certainly true that some of the foremost supporters of majority minority districts have been Republicans: for example, the Reagan-era Justice Department mentioned above. Or Benjamin L. Ginsberg, who told the New York Times that such districts were “‘much fairer to Republicans, blacks and Hispanics” in 1992—when he was general counsel of the Republican National Committee. But while all of that is so—and there is more to be said about majority minority districts along these lines—these are only indirectly the reasons why Michael Wilbon is right to fear statistical thought.
That’s because what Michael Wilbon ought to be afraid of about statistical science, if he isn’t already, is what happens if somebody—with all of the foregoing about majority minority districts in mind, as well as the fact that Democrats have historically been far more likely to look after the interests of working people—happened to start messing around in a fashion similar to how Isaac Newton did with those lists of ancient kings. Newton, remember, used those old lists of ancient kings to compare them with more recent, verifiable lists of kings: by comparing the two he was able to make assertions about which lists were more or less likely to be the records of real kings. Nowadays, statistical science has advanced over Newton’s time, though at heart the process is the same: the comparison of two or more data sets. Today, through more sophisticated techniques—some invented by Karl Pearson—statisticians can make inferences about, for example, whether the operations recorded in one data set caused what happened in another. Using such techniques, someone today could use the lists of African-American congressmen and women and begin to compare them to other sets of data. And that is the real reason Michael Wilbon should be afraid of statistical thought.
Because what happens when, let’s say, somebody used that data about black congressmen—and compared it to, I don’t know, Thomas Piketty’s mountains of data about economic inequality? Let’s say, specifically, the share of American income captured by the top 0.01% of all wage earners? Here is a graph of African-American members of Congress since 1965:
And here is, from Piketty’s original data, the share of American income captured etc.:
You may wish to peruse the middle 1980s—perhaps coincidentally, right around the time of Thornburg v. Gingles both take a huge jump. Leftists, of course, may complain that this juxtaposition could lead to blaming African-Americans for the economic woes suffered by so many Americans—a result that Wilbon should, rightly, fear. But on the other hand, it could also lead Americans to realize that their political system, in which the number of seats in Congress are so limited that “majority minority districts” have, seemingly paradoxically, resulted in fewer Democrats overall, may not be much less anachronistic than the system that governed Babylon—a result that, Michael Wilbon is apparently not anxious to tell you, might lead to something of benefit to everyone.
Either thought, however, can lead to only one conclusion: when it comes to the moonshot of American politics, maybe Isaac Newton should still—despite the protests of people like Michael Wilbon—be driving.