Thursday, October 11, 2007

"It's all a bunch of circumstantial evidence" -- Actually, no it isn't

Allow me to waste a few words on one of my biggest pet peeves when it comes to mainstream media coverage of the judicial system, especially coverage of steroids controversies: The constant confusion and misunderstanding of direct and circumstantial evidence.

The latest offender is a "For the Record" piece in this week's Sports Illustrated about track star Marion Jones, who recently admitted to using steroids (to no one's great surprise) and pled guilty to making false statements to federal investigators. The story traces Jones' past denials that "rang hollow in the face of voluminous circumstantial evidence that included damning statements by BALCO chief Victor Conte and Jones' ex-husband, shot-putter C.J. Hunter."

Presumably, the statements by Conte and Hunter amounted to, essentially, "Marion Jones used steroids, which I knew because I either saw her use/helped her use/gave her to use/injected her with those steroids." That is not circumstantial evidence of steroid use; that is direct evidence of steroid use. Circumstantial evidence of steroid use is things such as a player's performance dramatically improving when, under ordinary laws of nature, it should be declining; circumstantial evidence of steroid use is a player's head increasing in size. Direct evidence is evidence that, if believed by the jury, tends to establish the fact to be proven without more ("I saw X;" if the jury believes that, it ends to prove X is true). Circumstantial evidence is evidence that, if believed, tends to establish the fact to be proven only by the use of some logical/common-sense/experiential inference. The category of "direct evidence" is not limited only to scientific evidence (such as a failed drug test) or admissions.

The problem arises from common confusion between the question of direct/circumstantial, as opposed to the question of reliability. Admissions and failed drug tests are deemed reliable, while testimony from Hunter and Conte is not (for good reasons). But reliability and evidence type are distinct concepts that really have nothing to do with one another. Direct evidence (eyewitness testimony, for example) may be highly unreliable, as recent studies have suggested, while circumstantial evidence may be highly reliable. And scientific evidence, which many in our CSI world now regard as the only acceptable means of proving anything, often operates circumstantially. For example, imagine Jones failed a drug test in 2006. Does that prove she was using steroids in 2000, when she won five Olympic medals? Only if you draw the inference that, if she is using now, she was using then. And that is circumstantial evidence--reliable circumstantial evidences, perhaps--but still circumstantial. This, by the way, is why Rafael Palmeiro could not be prosecuted for perjury for telling the House committee that he did not use steroids. He made his statements in March, he failed the drug test in May. The failed test does not, without more, tend to establish that he had taken steroids prior to his March statement; it does that only if you draw an inference that he probably also had taken in the past, especially in light of his late-career power surge.

So, when Jones denied steroid use in the face of statements by Conte and Hunter, she was not defending herself against circumstantial evidence. She was defending herself against direct evidence with her own competing direct evidence. None of this is circumstantial. The only question is which direct-evidence witness (Jones or Conte) you want to believe. But if that is the question, then say so.

I beg reporters to be careful in its terms, especially as to central concepts such as these. OK. End of rant.

0 comments:

Post a Comment