Think Positive

Auguste Comte (1798-1857) seems a prime example of the sort of influential-at-one-time thinker that scarcely anyone feels guilty for not having read. You may have some sense of him as the originator of Positivism, without knowing what relation it has to the kind of positivism still referred to (often polemically) almost two centuries later. But chances are your curiosity on the matter is not burning.

Mine certainly wasn't when I fell down the Comtean rabbit hole a couple of weeks ago. My fascination was not with him as much as it was with the magnetism once generated by his rather dull books. In its (relatively) commonplace sense now, positivism is taken to identify knowledge strictly with a steady accumulation of verifiable facts from which general laws can be extracted. Comte would agree, but without claiming any originality for it, since Francis Bacon had understood the advance of knowledge in more or less the same terms a good 200 years earlier.

Comte's capital-P Positivism went further by insisting that knowledge would not just accumulate but consolidate into a sturdy and comprehensive system, with the various sciences as layers. Biology rested on chemistry, while chemistry was grounded in physics and so on, back to the perfect rigor and certainty of mathematics. The grand edifice was still under construction, of course, but Comte knew what would eventually be at the very top of it: a form of knowledge he called "sociology." It would both embody the culmination of human understanding and help to manage the social strains caused by progress and its aftershocks. Sociologists would be like Plato's philosopher-kings, but with a much larger database. Comte later carried things a few steps further by turning Positivism into a sort of humanist religion, with priests and rituals as well as saints (Bacon among them). Someone called it "Catholicism minus Christianity."

You can see how this vision of a cohesive, forward-moving body of knowledge, grounded in rational analysis, with an emphasis on reform, might would have a certain appeal. The young John Stuart Mill was strongly attracted to Comte's ideas, up to a point. He didn't go in for the religious part. But a few people around the world did. For a while there was positivist church in New York City, among other places, as well as a rather substantial presence in Brazil, memorialized by the words "Ordem e Progresso" ("Order and Progress"), still emblazoned on the country's flag.

But that might be one rabbit hole too many at the moment. Having decided to put Comte's quasi-messianic system aside for a while, one of the new books I read last month was Bobby Duffy's Why We're Wrong About Nearly Everything: A Theory of Human Misunderstanding (Basic Books). Only just now did the connection jump out at me. The author, who is director of the Policy Institute at King’s College London, draws on hundreds of public-opinion studies he and his colleagues have run on a variety of social and political questions and in a number of countries. The book is as positivistic as can be, in today's sense: the product of extensive fact gathering and statistical analysis, a search for verifiable trends in how people understand the world. Or rather, how we misunderstand it. Duffy believes that the spread of misinformation (accidental or deliberate) can be reduced, if not halted, through rational effort. That is a long way from Positivist social engineering, but Comte would approve.

"Almost all existing analysis," Duffy says, "tends to focus on one side or the other: on our fallible human brains or a manipulative information environment that leads us astray. This echoes our human need for simplicity and solutions: we want to see problems as caused by one thing or another, providing a clear focus for blame and a single answer." We are prone to delusions -- but also to having delusions about our delusions. In fact it is "myriad interactions and feedback loops" among our cognitive glitches "that together create a system of delusion," quite apart from any bogus information from people trying to take advantage of us.

One such glitch Duffy refers to is "what behavioral scientists call the ‘availability heuristic,’ a mental shortcut whereby we reach for information that’s readily available, even if it doesn’t quite fit the situation or give us the full picture." The classic example comes from a study in which some subjects were read a list of famous men and less famous women, while others heard a list where the proportions were vice versa. When asked later whether the list had more male or female names, subjects "were more likely to say that there were more of the gender from the group with more famous names."

Another tendency is to infer that the level of attention focused on something corresponds to the prevalence of the phenomenon itself. Duffy and his colleagues gathered data on both teen pregnancy and public estimates of its frequency. "Teen pregnancy across the 38 countries we studied is rare," writes the author, "with about 2 per cent of girls between 15 and 19 giving birth in any given year. On average, however, people estimated that 23 per cent of teenage girls were having babies each year." The figures for the United States corresponded closely to this average, while in Germany, respondents guessed that 16 percent of teenage girls were having babies when the real rate was 0.6 percent. Most of us "have no evidence to contradict the media's focus" on teenage pregnancy, Duffy says, "which makes [it] seem like a frequent occurrence."

The skewed perspective reflected in these figures exemplifies "emotional innumeracy": the tendency to "overestimate what we worry about as much as worrying about what we overestimate." The topic of immigration is another obvious example. The author reports that every survey of Americans or Britons he's done has shown they're convinced that the immigration rate is "roughly twice its actual level."

Then stir in the ego-fortifying penchant to overestimate what and how much we know. One survey found that a third of Americans said they believed that the "North Dakota Crash" was being covered up by the government, despite there being no such incident. Another study "asked respondents to rate their knowledge of 150 different topics," Duffy writes, "everything from Napoleon to ‘double entendre,’ but sprinkled in were completely made-up examples, such as ‘choramine’ and ‘El Puente.’ Respondents claimed to have at least some knowledge of 44 per cent of real topics, but also claimed to know something about 25 per cent of the made-up ones."

The mass media environment did not create any of these tendencies, although it isn't exactly helping. For Comte, progress implied the growth of ever certain and comprehensive knowledge. Duffy seems to share that optimism in principle, or least to want to feel it, while also implying that delusion and self-deception remain default modes of the adult mind.

One of the few really encouraging things he reports is a study at Stanford University of "how 10 Ph.D. historians, ten professional fact-checkers, and 25 Stanford undergraduates evaluated live websites and searched for information on social and political issues." The findings suggest that the crucial factor was not educational level, as such. Profs and undergrads alike "often fell for the easily manipulated features of fake websites, such as professional-looking logos. Even though these were well-educated groups, they tended to stay within websites." By contrast, the fact-checkers "took a much more lateral approach, opening multiple tabs to quickly gather external views of the veracity of the information … [and reaching] the correct conclusions in a fraction of the time it took the other groups."

As for the "theory of misunderstanding" promised by the book's subtitle, it seems to have been worked out in a passage Duffy quotes from Bacon in 1620: "The human understanding when it has once adopted an opinion … draws all things else to support and agree with it. And though there be a greater number and weight of instances to be found on the other side, yet these it either neglects or despises, or else by some distraction sets aside and rejects." Real progress means learning this lesson -- and revisiting it as necessary.