Inference Pitfalls

In the intelligence community, it seems sometimes as if for every piece of tradecraft or analytic trick of the trade, there are about a dozen cognitive traps or pitfalls for the unwary. For those on the intelligence end of things, this is just one of life’s little annoyances; for those in the counterintelligence end of things, this is what makes life interesting, as a large part of counterintelligence tradecraft involved baiting one’s opponents into committing those dangerous analytic no-nos.

One of those potentially dangerous analytic tools is “inference”: deducing what isn’t seen, or known, from what is. It is, essentially, the codification of “common sense”, kept in check by tests for “plausibility” and “reasonability”.

During the cold war, in particular, inference was an important and often-used analytic tool, not because it is really a good method of analysis, but because it, by definition, is used in situations where there is insufficient information to educe facts. Everybody involved knew this, of course, and both sides, over the years, spent a lot of time and effort trying to mislead the other by baiting them into making (incorrect) inferences.

The Soviets did this by creatively numbering their most important military units: one of the most common forms of inference is assuming that because there is something numbered x - say, a bomber squadron - there are also examples of that same thing numbered 1 through (x-1), as well. (An example of this, in popular culture, can be seen in internet debates about the Half Life game universe: Half-Life 2 takes place in “City Seventeen”, and so far, no other cities have been mentioned in the games. People infer that there are also Cities One through Sixteen, because that’s logical, but they don’t actually know so.)

Sometimes, inference is right: in the absence of counterintelligence deception ploys, or incorrect, unchallenged assumptions, it’s a surprisingly good analytic tool. And, most of the time - particularly outside the military environment - when it’s wrong, it’s only slightly wrong (if a company is producing a McGuffin 3, you might infer that they had, in the past, produced a McGuffin 1 and 2, as well. If that’s not the case, they probably designed two previous doohickeys, even if they never made it into production). Sometimes, though, inference is not only wrong, but completely and inexplicably wrong.

Consider: there are seven “cardinal” sins: lust, gluttony, greed, sloth, wrath, envy, and pride. We describe envy as green - people are “green with envy” (though I usually only see people green with motion-sickness… but I digress) - and occasionally either lust or wrath are characterized as red. One could, from this, infer that the other sins also have colors associated with them, but this does not, strictly speaking, appear to be true. Oh, if you hunt through metaphysical books and websites long enough, you’ll eventually find colors for all these vices, but they’re never primary colors, and the sources rarely agree with one another. There are seven primary colors (red, orange, yellow, green, blue, indigo, and violet) and seven cardinal sins, so you’d think it’d be a no-brainer, at least at first. But then, if you’re like me, you’ll sit motionless for a few moments, trying to figure out which sin is “indigo”, and give the whole thing up as a bad idea. Thus, inference, at least on this occasion, is completely and utterly wrong. And, not only is it wrong, it’s wrong in a completely irrational way that cannot, is not, explained by logic or reason…

(Bonus material, for those college students in need of a subject for a research paper: Until the last twenty or so years, when environmentalism became fashionable, “green” was almost invariably used in a negative connotation: untrained and inexperienced employees were “green”, people were “green with envy”, or “green around the gills” if unwell. The usual explanation for green as a pejorative is from the lumber industry, where freshly-cut lumber - which is sometimes literally green - is basically useless until dried. Could it be, though, that using “green” in a negative connotation was a symptom of the institutionalized anti-Irish bigotry that prevailed in much of the western world until somewhat recently? Discuss amongst yourselves…)

Bookmark to: