Deadlines on Scientific Integrity II: Inter-Disciplinarity in the Real World

Science communication is more than just cramming a lot of facts together. NEPA agencies should invest in the state of that art.

Our last post explained the Office of Inspector General’s conclusions surrounding Lease Sale 193’s hurry-up supplemental environmental impact statement (SEIS).  In that report, OIG found numerous instances where staff specialists at the Bureau of Ocean Energy Management (BOEM) were strong-armed by their superiors as the narratives they were offering for inclusion in the SEIS were cut short or excluded.  A few cried foul.  And we wondered whether some set of practicable principles might help improve what is becoming an all-too-common phenomenon in government’s use of multiple forms of technical expertise to inform what are irreducibly political choices.  The Obama Administration promised to outdo its predecessors when it came to “scientific integrity” and transparency.  And, yet, it seems to be embroiled in exactly the same messes those others were.

Toward Neutral Principles?

Perhaps the foremost principle should be: get over the desire for unitary, all-encompassing first principles.  The problem is too complex and the strategic value of allegations of perversion too high to think we can articulate some unifying, principled set of norms which are both specific enough to make a practical difference and reflect our common values.  This is not surrender. The obvious alternative is to start working in the specific contexts where these issues arise so regularly.  “Scientific integrity” in the evaluation of chemicals for health and safety purposes sounds a lot like scientific integrity in drafting and using an SEIS for an oil lease in the Chukchi Sea.  But the two involve different threats to integrity, different uses of “science” and scientific personnel, and different procedural and agency methods.  Still more diversity arises in the science-funding agencies and their interventions in the scientific world.

Toward Some Context-Driven NEPA Principles

What do we seek of our science/policy collaborations?  Truth, justice and the American way, of course!  But in a sober light, we want scientific people doing the science and policy people making the policy decisions.  Trouble is, that boundary is permeable, shifting, and of questionable relevance in the real world—where science is much less a “truth machine” than it is a method or way of proceeding which is, at times, terribly inconvenient for “policy” peoples’ political parties.  As Professor Doremus has observed,

[s]ound choices depend not on certainty but on awareness of uncertainties and their possible consequences.  Scientific integrity in the development, gathering, presentation, and interpretation of evidence helps ensure that decisions are supported by the best current understanding of the facts.

What it should exclude, thus, is a political person’s use of scientific data or otherwise interpretable evidence merely to justify a foregone conclusion, to battle back or confuse external pressures to rethink or re-examine a policy, or to commandeer the reputational strengths of specialists who are attuned to the evidentiary uncertainties of their particular domains without the corresponding empowerment of those specialists in the policy dialogue.

Especially in the NEPA context—where action agencies like BOEM are synthesizing disclosure documents that others (like EPA, CEQ, states, etc.) will use to gage the policy choices—incorporation of someone’s expertise should mean the fullest amount of their perspective practicable is heard.  What is that threshold?  In an overstuffed EIS like the SEIS for Lease Sale 193, that threshold is necessarily diminished.  And that may be the NEPA interdisciplinary problem right there.

Professor Eric Biber made a fundamental point to be kept in mind: scientific domains often overlap and it is in the selection of specialists and their techniques where much of policymaking discretion is exercised.  Choosing one kind of specialist to get at a question over other, perhaps equally qualified specialists can be the most significant choice.  See Eric Biber, Which Science? Whose Science?  How Scientific Disciplines Can Shape Environmental Law, 79 U. Chi. L. Rev. 471 (2012).

Biber reviewed a controversy that engulfed global fisheries a decade ago when Boris Worm published high-profile papers in Nature and Science attempting to show undeniable and steady declines in essentially all of the world’s major fisheries, blaming mismanagement.  Worm’s thesis was predicated on a very difficult statistic to gather, express, and analyze: catch-per-effort.  Other, more senior fisheries scientists were quick to attack Worm’s work and allege that the data he’d collected and the story he was telling proved nothing of the kind—that, in effect, Worm had inferred too much from extrapolations and other techniques needed to fit disparate bits and pieces of evidence together.  Worm’s statistical techniques were foreign to the traditionalists.  And Biber reviews a number of other like conflicts where “extraevidential considerations” and “metaphysical preferences” end up arbitrating between different, plausible constructions of available data and the nature of the relevant uncertainties.  Biber, 79 U. Chi. L. Rev. at 481.

In the NEPA context, this can be especially hard to detect, manage and communicate.  In the Lease 193 SEIS, for example, the projections being made about the region for the scenario of oil and gas development planned varied tremendously in their epistemic bases.  BOEM assumed (in the last SEIS) that 4.3 billion barrels of oil and 2.2 trillion cubic feet of gas would be recovered from 589 wells over a period of 70+ years and then had its specialists scheme that out from archaeological, sociological, economic, biological, ethical and other “perspectives.”

The SEIS was at least forthright about the risks should a “very large” spill occur.  In the Executive Summary, the area in question was described quite clearly:

Primary productivity . . . in the Chukchi Sea shelf region is considered the highest of any shelf region in the world due to the influence of several ocean currents. The Chukchi Sea is relatively rich in benthic faunal resources as compared to other Arctic shelves. Many species of fish are present here. Essential Fish Habitat (EFH) has been designated for all five species of Pacific salmon as well as for Arctic cod, saffron cod, and opilio crab. An abundance of marine mammals use the Chukchi Sea, most notably the bowhead and beluga whales, polar bears, Pacific walrus, and ice seals. Several species are listed as Endangered or Threatened or classified as candidate species under the Endangered Species Act (ESA). The region is also important to a wide variety of marine and coastal birds, including several ESA-listed species. Onshore, caribou and other terrestrial mammals inhabit the predominately tundra and wetland environment.

SEIS ES at 2.2.

But were some of the specialists who were asked to forecast the socioeconomic effects on the North Slope Borough “muzzled” and kept from saying what they truly believed would result?  Those were the allegations.  OIG found that nothing had been doctored or tampered with (the SEIS did conclude across each of the dimensions named above that the effects of the preferred alternative were “potentially” “major,” even without a huge spill).

That raises a hard question, though. CEQ’s NEPA regulations state that EISs “shall be prepared using an inter-disciplinary approach which will insure the integrated use of the natural and social sciences,” 40 C.F.R. § 1502.6 (quoting NEPA § 102(2)(A)), and that EISs “shall be prepared early enough so that it can serve practically as an important contribution to the decisionmaking process and will not be used to rationalize or justify decisions already made.”  Id. at § 2502.5.  When following these norms yields a 700+ page EIS wherein the ambiguities and uncertainties are edited out, do the necessities of “integration” leave any room for the “integrity” principle articulated above?

Under-determination, Politics and Principled Science Communications

The ideal at stake in the NEPA context is that science actually inform policy choices.  That means the decision cannot be a foregone conclusion.  But it also means that inferences from inconclusive evidence not be enshrouded in mystery, misdirection or deflections.

In the context of environmental science . . . political actors can often develop new arguments to “explain away” inconvenient new factual data through claims about, for example, the exceptional circumstances of the time period in which the data were collected, alternative causal mechanisms, claims as to bias or error in the collection of data, and so forth.

Biber, 79 U. Chi. L. Rev. at 481.  Deflections of the kind, though perhaps increasingly common, are at odds with NEPA’s ideals and with our own hopes for science in the crucible of public policy-making.

Newton’s Cradle by Dominique Touissant

If we accept that that is the trendline, it is no big leap to conclude that overstuffed EISs are, themselves, a threat to NEPA’s ideals.  If domain expertise is going to be wound up into the super collider that Lease Sale 193’s SEIS became, it will inevitably devolve into a sham.  It enhances neither transparency nor policymaking integrity for scientific domains to be crammed together so tightly that comparing their certainties and uncertainties becomes an exercise only a Nobel laureate could attempt.  NEPA’s repeat-play agencies like BOEM have an obligation to do better, to find the “Newton’s Cradle” out there which saves people from having to read the PrincipiaScience communication has made vast leaps like that a lot lately.  BOEM and others need not reinvent that wheel.  But they ought to get moving with it.

{Image: Arctic Sea by NOAA}

I teach environmental, natural resources, and administrative law at Penn State Law. Before teaching I was an enforcement lawyer at U.S. EPA. Along the way I've done work for environmental nonprofits and written a fair bit about NEPA.
No Comment

Leave a Reply

RELATED BY