“Declustering”

Mix and match good ideas.

separator-purplish-thinnest-separator-line

When we assume that certain ideas must or should occur together, we “cluster” them. But most clusters are not inevitable, and we may inadvertently constrain our thinking by treating them as permanent, inevitable fixtures. To “decluster,” it helps to think like a skeptical statistician or historian, or a good psychotherapist.

When two things repeatedly occur together, we may mistakenly infer that they must always occur together — that correlation necessarily implies causation, or some sort of logical dependency. For example, tall people are, on average, heavier and stronger than short people. But that doesn’t mean that being tall is the same thing as being heavy. Similarly, when we think of famous people, we usually assume they’re rich. But the two don’t have to go together. Mother Theresa was famous, but not wealthy. And many millionaires live out of the limelight.

We can also think of clustering as a psychological side-effect of the human learning process. As a child, if a dog bites you once, you’re apt to fear dogs forever. From your experience with one ferocious monster, you may overgeneralize and assume that all dogs bite, as if “person-biting” is an essential characteristic of dogs.

Clusters can change. Ideas that have occurred together historically can be separated, and ideas that have never been identified with each other may someday be clustered together. For example, patriotism in America is today more often associated with conservative politics than liberal politics. Why? Liberals can love America just as much as conservatives.

To decluster an assumption that certain ideas must occur together, we must first recognize the clusters; we can then analyze why the clusters exist and disentangle unnecessary associations.  Try our tips for critical thinking and unpacking assumptions.

In politics, political parties generally have an incentive to maintain consistent positions in order to maximize their brand name and achieve long-term results. Unfortunately, rigid clustering of the major parties’ ideas has greatly increased polarization. People who affiliate with a political party can ask themselves whether they agree with the full cluster of policy choices of their party. If not, they can question their colleagues about whether the party’s cluster must stay as it is.

But more fundamentally, we should question clustering wherever it occurs to see whether it unnecessarily constrains our thinking.