“Lies, damned lies, and statistics”

The popular expression I employ as a title for this post is often invoked to criticize the use and abuse of arguments supported by statistics. The quote has been attributed to various characters, including Mark Twain and Benjamin Disraeli. I do not intend to wade into that debate. I feel confident, however, that Shakespeare is not responsible for the utterance.  I am, however, prompted to say a word about statistics on this blog about Shakespeare and law because of a statistically significant confluence of events in the last 24 hours.  While studying a fairly recent statistical analysis of reversal rates and reasons for those reversals in appeals to the Texas intermediate appellate courts, I was thinking about how my own recent attempt to rely on statistics failed to find a receptive ear. Then I received an informative comment from James S. Ferris on my last post, wherein he refers me to his statistical analyses of Shakespeare’s canon supporting the theory that Edward de Vere, 17^th Earl of Oxford, was a likely author of works we attribute to “Shakespeare.”

When do statistics add to rhetorical force and when are they fairly brushed aside?

Among serious-minded people, the value of empirical data is not really subject to debate. Empiricism is the foundation of science; it is an essential tool for helping us break out of our narcissistic bubbles made of mere anecdotal “evidence,” hunches, coincidence, prejudice, and superstition. Statistics—the medium whereby empirical data is rendered accessible—reflect both hard and soft knowledge. Statistics involve hard numbers and yet presume to do no more than capture trends. Statistics permit us to make more informed predictions; but these predictions must always be characterized as reflecting probabilistic, not apodictic, certainty.

Because of the inherent modesty of statistics, someone can always decide to reject even the most damning statistics by concluding that the current situation constitutes an exception to a given statistical trend. And such a decision does not necessarily constitute intellectual dishonesty or cowardice—because part of what makes the science of statistics sound is that statistics do not presume to speak to individual instances. Just because a decision-maker has rejected x claim by y kind of person each and every time x has been presented to him over a multi-year period, that does not mean that the decision to reject this particular y’s x claim on this particular occasion was unfounded.

But if the proffered reasons given for rejecting a particular y’s x claim are not supported by legitimate evidence, then the statistics showing that z routinely rejects x claims from the likes of y should take on heightened significance. That is when statistics should resonate as authoritative.

Therefore, when someone presented with that situation decides to reject y’s x claim and the statistical argument showing that it was not given fair consideration that decision can seem like an exercise of raw power or the product of political preference. And that is certainly a circumstance that Shakespeare understood.

So I end by quoting Harry Hotspur, aka Henry Percy, from Henry IV, Part 1.  Hotspur chides his kinsman, the Earls of Worcester and Northumberland, who, after having played an instrumental role in furthering a political plot, are upset that the man whose ascension they enabled seems to have forgotten to elevate them too now that he holds the reins of power:

Shall it for shame be spoken in these days,

Or fill up chronicles in time to come,
That men of your nobility and power
Did gage them both in an unjust behalf,
As both of you—God pardon it!—have done,
To put down Richard, that sweet lovely rose,
And plant this thorn, this canker, Bolingbroke?
And shall it in more shame be further spoken,
That you are fool'd, discarded and shook off
By him for whom these shames ye underwent?
(I.3)

Hotspur was not surprised by abuses of power or that those who had obtained power unjustly would wield it in unjust ways; but that lack of surprise did not prevent him from railing against it nevertheless.