(Bloomberg) -- As misinformation weapons go, fake news is sort of like a cannon: noisy and provocative. Innuendo is like a dirty bomb — invisible, toxic and lingering. I became more aware of the misleading uses of innuendo after I spoke with linguistics professor Andrew Kehler during the run-up to the 2016 election.
Kehler studies something called pragmatic enrichment of language — the way we leave gaps in our utterances which listeners will fill in, allowing us to converse without being impossibly wordy. But by the same token, speakers who want to mislead without literally lying can nudge people to fill is such gaps with their own faulty assumptions.
This happened in the presidential debates, and it happens in advertising and other forms of persuasion. For example, a television commercial might promote Brand X vitamins as having twice the iron as a competitor’s, he said. That may be true. But it implies that more iron will make you healthier, which is likely to be false. (Recent data show more Americans get too much iron than too little). “We’re always taking more information away from utterances than what is said, and we don’t realize how we are manipulated this way,” he told me.
Identifying this kind of innuendo could help with the problem of polarization in the media. We can all learn from the viewpoints of people who are more liberal or conservative than we are. But there’s no understanding to be gained from arguments that try to flatter or fool readers. A good example comes from a recent Wall Street Journal opinion piece titled “How Bad is the Government’s Science?” by Peter Wood and David Randall. They are, respectively, president and director of research the National Association of Scholars.
I was eager to read this piece. I wanted to know how the U.S. government is using shoddy research to inform policy. I’d never heard of the National Association of Scholars, but that’s okay. Scientists don’t have to belong to the prestigious National Academy of Sciences to put the time into a good investigation of bad government policy.
The authors started by explaining something I’ve been writing about since 2012 — a rash of irreproducible results in social science and medical research, sometimes collectively dubbed the “replication crisis.” They point out that surveys have revealed a big portion of published papers in medicine, social science and economics are based on experiments that can’t be replicated.
The authors then cite some examples, including an infamous claim that a body language technique called power posing can make you feel more powerful and cause real physiological changes. I kept reading, eager to find out how such discredited results are being used in the creation of public policy. But what came next was this:
“The economics research that steers decisions at the Federal Reserve and the Treasury Department needs to be rechecked. The social psychology that informs education policy could be entirely irreproducible. The whole discipline of climate science is a farrago of unreliable statistics, arbitrary research techniques and politicized groupthink.”
It may be that the government isn’t cautious enough about using results from the economics or social psychology journals, but it’s hard to tell since there were no specific examples on offer. The statement about climate science is a non sequitur, but readers may assume that its inclusion means climatology is deeply afflicted with reproducibility problems.
There’s been a lot of research into the replication crisis, and despite some overblown headlines implying the whole of science is going down the drain, the crisis seems to be limited to social science (including economics) and medical research. The problems in these areas no more negate the value of climate science than they suggest NASA should stop using physics to calculate the trajectories of government spacecraft.
Of course, scientists in any field can screw up through error, wishful thinking or, more rarely, cheating. But the physical sciences have the advantage of a well-established set of laws, which are subject to change but only in light of multiple lines of evidence. That’s why cold fusion and other erroneous results in physics tend to get exposed before they become established knowledge.
Climate science is rooted in basic chemistry and physics, going back to the 1800s, when scientists first hypothesized that gases in the atmosphere kept our planet warm. The American Institute of Physics offers some good material about this history. Historian Spencer Weart explains how Irish physicist John Tyndall created a source of long-wavelength infrared radiation, equivalent to sunlight that’s been absorbed and re-emitted by the earth. He found it passed through oxygen, nitrogen and other gases but was absorbed by carbon dioxide and water vapor.
This is backed by independent lines of evidence. The subsequent development of quantum mechanics can be used to calculate how different wavelengths of light interact with different molecules of gas. In a similar way, Darwin’s idea about the evolution of life from a common origin is backed by the fossil record and by analysis of DNA.
In 1896, Swedish chemist Svante Arrhenius published the first climate model, estimating that if coal burning doubled atmospheric carbon dioxide, the global average temperature would rise by about 5 degrees Celsius. Since then, the understanding of clouds and other complicating factors have led to a slightly lower estimate, and scientists have developed more complex models giving a range of predictions. None of this makes for a “farrago of unreliable statistics, arbitrary research techniques and politicized groupthink,” as Wood and Randall write.
And now, the online magazine Undark, produced by the Knight Science Journalism program, has picked up on the National Association of Scholars’ contention, balancing it with other voices but never directly challenging the assumption that climatology is caught up in the replication crisis. Instead of cleaning up the toxic waste of innuendo, such attempts at balance are just spreading a diluted form of contamination.
©2018 Bloomberg L.P.