Shermer: Science Friction

From Scienticity

Scienticity:
Readability:
Hermeneutics:
Charisma:
Recommendation:
Ratings are described on the Book-note ratings page.

Michael Shermer, Science Friction : Where the Known Meets the Unknown. New York : Henry Holt and Company, 2005. 296 pages, with notes and index.

This is a nice collection of essays by Shermer, who writes a column for Scientific American; however, all of these essays originally appears elsewhere, several in Skeptic magazine. As I often find with such collections, some of the essays are stronger than others, but overall I found it stimulating and worthwhile to read.

Shermer's theme, to the extent that there is a theme, is skepticism and its central role in science. He has a fondness for debunking pseudoscience that makes of think of Martin Gardner, but this collection goes well beyond that and presented enough insight and creative analysis to keep me engaged.

Here are two representative excerpts. The first is from "Shadowlands: Science and Spirit in Life and Death", a personal essay on existential questions that meshes with thoughts about his mother's death.

Sixty thousand years ago, in a cave 132 feet deep cut into the Zagros Mountains of northern Iraq, 250 miles north of Baghdad at a site called Shanidar, the body of a Neanderthal man was carefully buried in a cave, on a bed of evergreen boughs, on his left side, head to the south, facing west, and covered in flowers, so identified through microscopic analysis of the surviving pollens. Already in the grave were an infant and two women. The flowers were from eight different species and the arrangement was not accidental. There was a purpose to the burial process. It is the earliest memorial celebration of life and mourning of death of which we know.

Now that Neanderthals are extinct, we are the only species who is aware of its own mortality. Death is an inescapable end to life. Every organism that has ever lived has died. There are no exceptions. Behind every one of the 6.2 billion people now living lie seventeen others in the ground, for 106 billion, according to the demographer Carl Haub, is the total number of humans who have ever lived. Our future is sealed by our past.

Thus, we are faced with the existential question that has haunted everyone who has thought about this uncomfortable fact of life: why are we here? People throughout the ages and around the world, in all cultures and communities, have devised a remarkable variety of answers to this question. Indeed, anthropologists estimate that over the past ten thousand years have created roughly ten thousand different religions, the wellspring of which may be found in the answers they have offered to that soul-jarring question: Why are we here?

I started thinking hard about this question in 1992, when my mother started acting strange.... [pp. 101—102]

The essay that I found most compelling, oddly, was the preface to the book, titled "Introduction: Why Not Knowing", in which Shermer thinks about the amazing capacity the human mind has for finding patterns in nature, and speculating on how easily that propensity can lead our best critical reasoning astray at times. He seems to be getting closer to identifying the "why" that he first addressed in an earlier book, Why People Believe Weird Things. This is an excerpt.

In the end, the confirmation bias won out and Dow Corning had to pay $4.25 billion to settle tens of thousands of claims. The only problem was, there is no connection between silicone breast implants and any of the diseases linked to them in these trials. After multiple independent studies by reputable scientific institutions in no way connected to either the corporation or any of the litigants, the Journal of the American Medical Association, the New England journal of Medicine, the Journal of the National Cancer Institute, the National Academy of Science, and other medical organizations declared that this was a case of "junk science" in the courtroom. Dr. Marcia Angell, the executive editor of the New England Journal of Medicine, explained that this was nothing more than a chance overlap between two populations: 1 percent of American women have silicone breast implants, 1 percent of American women have autoimmune or degenerative tissue diseases. With millions of women in each of these categories, by chance tens of thousands will have both implants and disease, even though there is no causal connection. That's all there is to it.

Why, then, in this age of modern science, was this not clear to judges and juries? Because [Francis] Bacon's idols of the marketplace dictate that scientists and lawyers speak two different languages that represent dramatically against one another. There will be a winner and a loser. Evidence is to be marshaled and winnowed to best support your side in order to defeat your opponent. As an attorney for the prosecution it doesn't matter if silicone actually causes disease, it only matters if you can convince a jury that it does. Science, by contrast, attempts to answer questions about the way the world really works. Although scientists may be competitive with one another the system is self-correcting and self-policing, with a long-term collective and cooperative goal of determining the truth. Scientists want to know if silicone really causes disease. Either it does or it does not.

Marcia Angell wrote a book on this subject. Science on Trial, in which she explained how "a lawyer questioning an epidemiologist in a deposition asked him why he was undertaking a study of breast implants when one had already been done. To the lawyer, a second study clearly implied that there was something wrong with the first. The epidemiologist was initially confused by the line of questioning. When he explained that no single study was conclusive, that all studies yielded tentative answers, that he was looking for consistency among a number of differently designed studies, it was the lawyer's turn to be confused." As executive editor of the New England Journal of Medicine, Angell recalled that she was occasionally asked why the journal does not publish studies "on the other side," a concept, she explained, "that has no meaning in medical research." There is "another side" to an issue only if the data warrant it, not by fiat. [pp. xxiv—xxv]

-- Notes by JNS