Junking Junk Science
The label "junk science" has been one of the most powerful tools in ensuring that political and legal decisions are taken based
on only the soundest of footings. Alarmism, hype and scaremongering have all been avoided by scrutinizing scientific data and evidence to check that it conforms to good scientific practice. If it doesn't, it is junk science. Over the past decade, we have seen more and more safeguards put in place to stop junk science influencing political or judicial decisions. Now, however, the very concept is under attack. And the tactic the interest groups are using in their onslaught is to cry "censorship."
The tactic is quite recent. It began a month or so ago when questions were raised in certain publications such as Harper's and In These Times over the effects of the Federal Data Quality Act (FDQA). Enacted in December 2000, it requires that data used to support laws and regulations should conform to strict scientific standards. One of the first targets under this piece of legislation was the thoroughly discredited U.S. National Assessment on Climate Change, which relied for much of its alarmism on two climate models which were proven to have no more predictive power than tables of random numbers. As Pat Michaels of the University of Virginia put it, the assessment "breaks the cardinal rule of science: If a hypothesis doesn't work, throw it out. The Assessment can't pass the simplest of scientific tests."
Yet for reasons that are hard to fathom, the administration continues to disseminate the discredited document. The questions in the publications alluded to above focused on continued attempts to get the executive branch to conform to the FDQA. Both Harper's and In These Times suggested that the attempt to expunge junk science from the record was aimed at censoring science.
This is essentially arguing that black is white. If science does not conform to basic scientific standards, it isn't science. Censorship just isn't the issue. What is at issue is the assurance to taxpayers that policies they pay for are based on the soundest scientific basis.
Yet the argument is not only used in relation to the FDQA. We saw it again on June 20, when the New York Times ran an editorial about the controversy over the White House's changes to the recent EPA report on the State of the Environment. Once again, the National Assessment was at issue. The White House, perfectly reasonably, asked for references to this junk science to be removed. The entire basis of the EPA's case on climate change therefore collapsing, the agency removed all reference to climate change, leading the Times to deploy once again the accusation of censorship, arguing that the administration "wants to bury any research findings that global warming may be a threat to human health or the environment."
Editorials all over the country took up the cry. The Hartford Courant said "government policy on this issue ought to be based on the best scientific evidence, not politics." The Atlanta Journal Constitution charged "the White House slaps down science in favor of its corporate oil friends." The Idaho Statesman, home paper of Gov. Dirk Kempthorne, the favorite to succeed Christie Todd Whitman as the head of EPA, said, this "is an administration that seems determined to shape science around policy, not the other way around."
Now even the judicial safeguards against junk science are under attack for the same reasons. Ten years ago, on June 28 1993, the Supreme Court handed down its ruling in Daubert v. Merrell Dow Pharmaceuticals, Inc. In that ruling it established that expert scientific testimony should be subject to a set of rules for admissibility. The testimony should be based on a testable theory or method that had passed peer review; it should possess a known error rate and/or standards; and it should reflect "generally acceptable" science. The new standards stopped large numbers of charlatans and crackpots who had previously posed as experts from testifying in court, where they might have been able to convince a jury that they knew what they were talking about. Instead, it was judges who now had the responsibility of deciding whether scientists possessed enough credentials to lay their evidence before a jury.
Yet now the junk scientists are fighting back. After years of seeing their claims of environmental health risks being ruled inadmissible owing to lack of statistical significance or other sound reasoning, they have taken to calling the Daubert ruling scientific censorship, with the added twist that they claim it aids "polluters." The Wall Street Journal's Science Journal column took their claims at face value on June 27, under the headline, "'Junk Science' Ban Also Keeps Jurors from Sound Evidence."
The allegations against Daubert are summarized in a report available at the ironically titled Defendingscience.org. The scientists concerned make three basic charges against Daubert. First, that exclusions of "expert" testimony from the courtroom rose significantly after Daubert. This is, of course, exactly what Daubert was designed to do. Secondly, that defending against challenges to scientific reliability is "chilling" and puts scientists off testifying. But if their science meets basic standards, they will meet the required tests. Thirdly, there appears to be a disparity in Daubert's application between civil and criminal cases (in criminal cases, neither side can afford to challenge admissibility). This may be true, but it is a red herring, failing to establish any reason why Daubert itself is a bad thing.
The charges of 'scientific censorship' are disingenuous. What they
are really aimed at is abolishing the concept of junk science. Without
rules protecting us from junk science, the scaremongers, alarmists
and trial lawyers will have a field day. Junk science rules, far from
censoring science, champion it.
In accordance with Title 17 U.S.C. Section 107, any copyrighted work in this message is distributed under fair use without profit or payment for non-profit research and educational purposes only. [Ref. http://www.law.cornell.edu/uscode/17/107.shtml]