Scientific American's predictions weren't predictions (but they were wrong anyway)

Sarah Zhang’s post on Gizmodo, "Why Scientific American's Predictions from 10 Years Ago Were So Wrong,” leaves me with mixed feelings. First, I think she did a good job with it. She very capably reviews the current state of the assorted technologies mentioned in that feature from December 2005 and, I think, fairly evaluates how the excitement or hype that surrounded them at that time has or has not yet been realized. 

Moreover, her intentions in writing the piece are dead-on: getting infected with the optimism that surrounds new technologies and discoveries is distressingly easy, and science journalists need to be ever on guard against it. Follow-up articles like this one are therefore sorely needed as a counterweight to the wide-eyed coverage that too often appears, floating on some combination of public-relations puffery, reporter credulity, and simple error. I’m a big advocate for such reality checks: witness the workshop on “Viewing Futurism Skeptically” that I led at the recent Northeast Conference on Science and Skepticism, and my past calls that science news focus less on what’s new and more on what seems to hold up as true and valuable.

As the editor who oversaw the creation of that section, I also appreciate that Sarah isn’t beating on Scientific American specifically; she’s using it as a convenient example of a wider journalistic tendency, as she notes in the comments. 

I hope it doesn’t seem overly defensive, though, if I nitpick that the Scientific American 50 wasn’t a list of predictions, notwithstanding the slugs in the Gizmodo story. The introduction to the section states that it “recognizes people, teams and organizations whose recent accomplishments… demonstrate leadership in shaping both established and emerging technologies.” It continues: “In naming the winners of 2005, the magazine’s editors and their expert advisers identified noteworthy trends related to technology ranging from polymer memory chips to a technique for regenerating damaged heart tissue.” 

Thus the awards were primarily about leadership in emerging or ongoing trends. Leadership can happen many ways and mean many things: a general can lead an army; the fastest runner can lead a pack of sprinters; a figurehead can lead a ship. The SciAm 50 didn’t try to be too narrow in its definition of leadership because there wasn’t just one that would work for all the researchers, entrepreneurs, policymakers, and organizations it aimed to cover. No doubt that points to a weakness of the central conception, and I’ll cop to that. But Sarah’s exclusion of the business and policy components tends to obscure what the section was about.
It’s not a coincidence that the introduction avoided the word predictions and instead discussed trends. Trends come and go; sometimes they mature into something and sometimes they don’t, but they rarely last.

One of the ideas I discussed in my NECSS workshop on skeptical futurism was that predictions are different from forecasts and projections. Predictions make specific, falsifiable claims. Forecasts and projections discuss probable outcomes under contingent circumstances: if these conditions continue, then this outcome is likely. At Scientific American, we would not have stuck out our necks on a foolish claim that all or even necessarily most of the discussed trends would continue. Rather, the SciAm 50 represented a kind of encouragement for work that seemed promising and desirable.

But in offering that objection, I also concede the spirit of Sarah’s critique: the up-with-science sunniness of the Scientific American 50 and all the talk of trends can be mistaken easily for a set of predictions—and if all the things on the list had come to pass in the past decade, no doubt we would be congratulating ourselves now on our shrewd foresight. Enjoy Sarah’s article, take its reappraisals of once promising technologies to heart, and watch out for the distorting effects of uncritical optimism in the next science journalism you read.