The Thankless Task of the Political Fact-Checker

One overarching theme of the 2012 election season was a struggle with truth. Both campaigns were accused of serial falsehoods, and continuing to spread incorrect information after the truth was reported. The utility of fact-checking in the election was also questioned: David Carr argued that the efforts of fact-checkers were irrelevant to the conduct of the campaigns, while the parsing of endless minutiae fed partisan bickering; Bill Adair of PolitiFact responded that the purpose of fact-checking was to inform voters so that they could better judge the candidates, not to moderate the statements of candidates directly.

As a lawyer who has defended more than a few defamation cases, all of this led me to consider the difficulties faced by those who attempt to analyze the truth of even seemingly simple statements. It sometimes surprises those not familiar with media law that courts in libel cases are not generally tasked with deciding issues of ultimate truth. Rather, judges and juries are responsible for determining whether a defendant's statements are probably true or false, to a degree of probability that allows imposition of penalties in a manner consistent with society's level of comfort. After all, jurors are not witnesses and have no direct knowledge of events; their verdict represents the opinion of a small sample of the public about the veracity of actual witnesses and conclusions as to how piecemeal information fits together. Such a decision might be reliable enough for an award of damages, but it would seem strange to present a jury's determination to the world as a binding declaration of indisputable fact.

And all of this presumes that a statement is capable of being shown to be true or false through the presentation of evidence. The legal analysis of a particular allegedly defamatory statement is not a simple dichotomy of right and wrong; rather, it quickly devolves into distinctions between fact and opinion, material and immaterial errors, predictions about the future and declarations about the present, and implications and explicit statements. In many cases, courts do not even reach the tricky question of whether a statement is true or false, finding that even if a statement might be false the defendant lacks culpable fault for its publication.

In other words, the issue in a defamation case is rarely a simple matter of whether the publisher of a statement got it "right" or "wrong," but whether the statements were (1) correct enough that it is improper to impose liability, (2) subjective enough that the question of truth has no practical meaning, (3) innocuous enough that we will not consider the plaintiff to be harmed, or (4) innocent enough that we will not punish the speaker.

It is tempting to dismiss the parsing of speech in defamation cases as the byzantine working of a lawyer's mind at its most arcane, but it is basically the same task that political fact-checkers must undertake when evaluating a candidate's claims. Political speech is routinely highly charged; even the simplest statements can be rife with implications as well as explicit claims, and contain a tumultuous mixture of factual allegations, unspoken assumptions, and opinions based upon disclosed or undisclosed facts.

For example, suppose candidate John Smith were to claim that his opponent, incumbent Jane Doe, "would be in favor of budget measures that would cost individual taxpayers up to $3,000." It might seem relatively straightforward to determine the truth of this statement, but evaluating the statement will raise a host of questions, such as:

• What does it mean to be "in favor of" budget measures?
• Is the statement with that Doe "would be in favor of" particular measures an opinion based on Doe's general policy platform, or is it based on actual statements or votes she has made?
• Which "budget measures" is Smith grouping together as a basis for this claim?
• What calculation is Smith using to reach the projected figure of $3,000?
• Is the calculation based upon assumptions as to future events that are incapable of verification?
• Does the calculation have any internal flaws?
• Is the "cost" to individual taxpayers limited to tax increases, or are there ancillary costs?
• Are the "costs" attributable to the budget measures, or do they arise out of preexisting circumstances such that they would be incurred even if an approach preferred by Smith were implemented?
• Does "individual taxpayer" refer to all taxpayers, or just certain taxpayers?
• Does the $3,000 figure represent a figure for each taxpayer, or the total effect on all affected taxpayers?
• What does "up to" $3,000 mean? How much less could