EPA: Fracking’s lack of effect on water isn’t something we can’t not negate

BEATRICE: O, on my soul, my cousin is belied!
BENEDICK: Lady, were you her bedfellow last night?
BEATRICE: No, truly not; although, until last night, I have this twelvemonth been her bedfellow.
LEONATO: Confirm’d, confirm’d! O, that is stronger made which was before barr’d up with ribs of iron!
— William Shakespeare, Much Ado About Nothing, Act IV, Scene 1

The EPA opted to play politics with its final report on whether hydraulic fracturing (fracking) contaminates groundwater. Media have opted to go along. Some headlines:

So … the findings from 2015 have been erased? Something new suddenly bubble up? What?

Politics. The draft report stated that there was “no evidence that fracking systemically contaminates water.”

This was a mainstream finding. Long-term academic and government studies have repeatedly made the same finding. Fracking is an intrinsically safe process.

Environmentalists howled at it. Pressure was brought, letters were written, hands were wrung. (How strange that is, considering that because of fracking, we have seen actual, significant reduction in greenhouse gas emissions!)

So that sentence was struck from the report. In its place, the EPA report said the fracking process “can impact drinking water resources under some circumstances” that can “range in frequency and severity, depending on the combination of hydraulic fracturing water cycle activities and local- or regional-scale factors.”

Possibilities for which the EPA cites … no evidence:

However, significant data gaps and uncertainties in the available data prevented us from calculating or estimating the national frequency of impacts on drinking water resources from activities in the hydraulic fracturing water cycle.

It was simple rephrasing. Instead of saying there was “no evidence” that fracking contaminates water, it made it sound as if the reality of fracking contaminating water couldn’t be shown yet owing to “significant data gaps and uncertainties.”

The rephrasing worked, at least for the media — who, in fairness, probably can’t be expected to understand the concept of a null hypothesis or of begging the question.

Leonato’s fallacy

bigfootLike Leonato, they jumped at lack of confirmation of the negative (Hero isn’t cheating on Claudio) as proof of the positive (Hero is “an approved wanton”). They were willing to believe that a lack of overturning the null, as opposed to outright confirmation of the hypothesis, was … confirmation of the hypothesis.

But it’s essentially the same finding as before.

Consider proving the existence of Bigfoot in the EPA manner. There’s no definitive proof there isn’t Bigfoot, based on the difficulty of conclusively proving a negative. But there’s no definitive proof that Bigfoot exists, which is the issue.

We could state the latter and be scientific about it. Or, to mollify Bigfoot enthusiasts, we could say there are significant data gaps and uncertainties in the available evidence for Bigfoot.

Stated that way, the argument is not inaccurate. But rhetorically it assumes Bigfoot exists although good evidence for him has yet to be found.

Jon Sanders / Research Editor and Senior Fellow, Regulatory Studies

Jon Sanders studies regulatory policy, a veritable kudzu of invasive government and unintended consequences. As director of regulatory studies at the John Locke Foundation, Jo...