“Facts are stubborn things, but statistics are more pliable.”
– Mark Twain
Yesterday’s post looked at Census data. We saw that the process of collecting and analyzing the data can’t really be separated from considerations of the end use. A short vignette shows how that same truth can play out, even in the arena of so-called pure science.
When I was a graduate student in physics at The University of Chicago, the department had a weekly seminar. One year Subrahmanyan Chandrasekhar, a faculty member, already a famous astrophysicist (he would go on to win the Nobel prize) and author of over 1000 publications, delivered a talk on his new theory of quasars. The next week’s speaker happened to be Tommy Gold, a Brit, a fellow of the Royal Society, and then on the faculty at Cornell. He laid out a competing theory. When Gold was done, Chandrasekhar was the first person to ask a question. Wearing his signature pinstripe suit and looking like he just stepped out of the pages of GQ, Chandrasekhar cut a dignified figure as he rose from his chair. “Surely you would agree…” he smoothly began…
Can’t remember exactly what he said next, but it was the equivalent of asking Gold whether he believed that 2+2=4. It was also the first step down a line of argument that would lead to Chandrasekhar’s theory. The room was packed, but you could have heard a pin drop. Gold looked at him for what seemed like forever; then finally said “No.”
Maybe for the more senior faculty this was just another day at the lab, but we students had never seen a scientist do that. Every person in the room knew the only right answer was “yes.” Most of us had also been present the week before. Chandrasekhar glowered, then silently took his seat. A serious chill set in the room. The Q&A went on for some time, but never fully recovered.
Tommy Gold knew that you couldn’t separate data and facts from considerations of the end. To defend his theory he was going to have to say “no” at some point, so he might as well do it early.
It was not science’s best day. But in that instant I took to heart – internalized in my gut, not just my head – that while science might be objective, the pursuit of science could be very political – and combative.
But back to the Earth, and back to the environment. Here’s an example taken from the National Acid Precipitation Assessment Program (NAPAP). I got to see this up close and personal from 1987-1993. At that time I was the NOAA Deputy Chief Scientist. The NAPAP Program office, an interagency group, was administratively housed in NOAA under the leadership of the Chief Scientist. The same argument was playing out, but now with institutions, not just individuals, as the main actors.
For some time, scientists had been noticing alarming ecological trends in the northeastern United States and Canada – acidification of lakes and soils, freshwater fish die-offs, and area-wide tree damage among other symptoms. Similar problems were surfacing in Europe and elsewhere around the world. Evidence started pointing to emissions of oxides of sulfur and nitrogen, resulting from the combustion of fossil-fuels, especially coal for electrical power, and gasoline for transportation, as a major culprit.
At the time, much of the data on these trends and their causes were being collected by the Department of Energy and the Environmental Protection Agency respectively. DoE was the agency responsible for keeping the lights on and vehicles moving, and cheaply. They didn’t seem to find much of a problem, or much reason to regulate. By contrast, EPA, the agency in the business of regulating pollution, found plenty to worry about, and wasn’t concerned about cost. Outsiders could see that the DoE and EPA conclusions were poles apart, and that they seemed to reflect agency self-interest (and the interests of their constituencies). However it was difficult for policymakers and the public to go about the real-world job of splitting the difference. The stakes were high. The environmental damage was extreme, and rising; the costs of regulation were going to be significant. The status quo? Unacceptable.
NAPAP was the result; an attempt to structure the discussion between EPA and DoE (and many other players: government, private, and academic, state-level, federal, and international), and reach a consensus on both the science and the action needed based on facts, not rhetoric – before the politicians and interest groups got involved. [By the way, the Canadians, who were experiencing much of the damage, were rather more cynical. They viewed NAPAP as nothing more than a delaying tactic by the U.S..]
WWFFD? What would the Founding Fathers do?
Remember, the Founding Fathers didn’t see this coming. They didn’t contemplate a time when much of policy debate and legislation would hinge on the basis of scientific data and findings. Today, virtually all policy formulation and all political debate take place in an ocean of data and statistical analysis.
Take food safety: levels of pesticides and bacteria in food. Think about public health: regulations with respect to vaccinations, mammogram coverage in health plans, code-orange ozone days, and so much more. Or focus on the environment: admissible levels of toxins, bacteria, etc., in water supplies; carbon dioxide emissions; and the like. Or consider the Endangered Species Act. In each arena, and many others, policy is supported and opposed, and made, and modified, on the basis of data and findings. Even policies of longstanding concern, antedating this trend — such as full-employment measures, price supports for agriculture, public education, etc. – are all informed today by data, statistics, scientific understanding.
But if they were alive today, and the Constitutional deliberations were still in play, the Founding Fathers would see a problem. They would accept that DoE and EPA would each require in-house data collection and science in order to support their mission. But they might not stop there. They might determine that an additional governmental body would also be needed to collect data, and do science, and in the process provide a check and a balance on more narrowly focused agencies and the Congress.
Today, failing explicit Constitutional provision, government uses several options in practice. One is the judiciary. Our courts are indeed shouldering much of this load today. Think of Massachusetts v. EPA, 549 U.S. 497 (2007), in which the Supreme Court forced EPA to regulate carbon dioxide and other greenhouse gases as pollutants. But the courts are overtaxed already. Judges by training are not necessarily well-equipped to deal with such cases; they have to do a lot of boning up on the scientific particulars. The result? Slow, intermittent progress, applied haphazardly across the issues. Some receive considerable attention. Others much less.
A second option is special attention in the executive branch. The history of the federal government offers several examples. Here are a few. When atomic energy was no longer a laboratory curiosity but had grown commercially viable, the Nuclear Regulatory Commission had to be split off from the Atomic Energy Commission (now DoE), so that industry regulation would be separate from atomic-energy advocacy. The National Transportation Safety Board was originally established within the Federal Aviation Administration, but it was soon recognized that the NTSB had to be an independent agency if its findings were to be unbiased – and unflinchingly address any shortcomings in FAA performance. In a similar way, EPA itself had to be an independent agency, so that it could comment in a detached way on the activities of other federal agencies.
And let’s not forget the Census Bureau. The parent Department of Commerce houses many other agencies responsible for making complex measurements (and letting the chips fall where they may). Take the Economic Statistics Administration. It compiles statistics for the national income accounts (GDP, balance of trade, etc.). Users are free to use their findings to place bets on the financial markets, argue the success or failure of the administration, etc. NOAA measures fish stocks, average global temperatures, etc. Think of the National Weather Service. When it issues a tornado warning, its automatic messages don’t add, “and anyone in the warning area not in his/her basement for the next half hour will be fined $100 per minute.” The public is free to hunker down, to evacuate, or grab the videocam and rush outside to get footage they can put on YouTube.
A third option is the Congress. Here the results have been checkered. Within the Congress, the Office of Technology Assessment once performed some of these functions but was disestablished. The Congressional Research Service and the Congressional Budget Office perform some of these functions and still survive. All three agencies have played to mixed reviews.
But instead of allowing these patchwork approaches to crop up (and the use of these promises to be increasingly extensive and cumbersome in the future), the Founding Fathers could have chosen to stand up a fourth branch of government, charged with independently developing science and information on which an entire range of national decisions could be based.
The pros? Such a fourth branch of government could be of great value in structuring policy formulation on a wide range of national issues. The cons? To see those, all we have to do is look at the woes affecting the current federal judiciary. It’s dependent on funding and appointments from the legislative branch. This has been a contentious, almost dysfunctional arena in recent years. The judiciary is frequently accused of being elitist – high-handed, unresponsive, out of touch. It’s criticized for what some see as attempts to intrude – illegally – on the powers of the executive and legislative branches. All these criticisms would quickly be leveled at the new entity. And, in the nature of our world, some would be undeserved, but others would have merit.
Perhaps the Founding Fathers would have also asked Hamilton, Madison, and Jay (“Publius”) to write a few new additions to the 85 extant Federalist papers (probably the closest the early 1800’s came to blogposts).
and the Fisheries Management Councils themselves represent an interesting case study in the conflict-of-interest issue.