Granularity: Think as big as possible but no bigger

Granularity: Think as big as possible but no bigger

Print Friendly, PDF & Email

ShellI’ve argued before that there are three types of guidance or wisdom. The first and simplest is the “always do X” type. For example, always think positive thoughts, always follow your father’s advice, or always do what the Bible says.

The second is the “a time for X; a time for Y” type, the wise recognition that you have to do different things in different situations. It’s the wisdom of saying “well, it depends” – without specifying what it depends upon. In the Psalms, it’s a time to reap; a time to sow. In the serenity prayer, it’s praying for the wisdom to know the difference between what you can and can’t change.

The third type of wisdom is the most difficult and open-ended. It’s the potentially endless pursuit of the specifics by which to know the difference between when to reap and when to sow, when to have courage and when to have serenity, when to do which.

How then to balance between these three orders of wisdom? Einstein said, “A theory should be as simple as possible but no simpler.” That, by the way, is wisdom of the second type. It translates as “Sometimes you have to simplify and sometimes you have to complicate. It depends on the situation.”

Normal Audio : Download | Embed | Play in Popup

150% Speed Audio : Download | Embed | Play in Popup

Taking Einstein’s advice, we should prefer the first kind of wisdom wherever the situation permits it. If you find an all-weather, all-purpose truth – one that leads to success, whatever the situation – by all means adopt it. Candidates include “breathe” or “keep yourself alive,” though even these have their exceptions. “Breathe unless you’re underwater.” “Keep yourself alive except to save children’s lives.”

It’s disappointing how few cases actually call for wisdom of the first type. It would be nice if we had more all-weather, all-purpose truths. For lack of them, we’re forced to say “it depends,” which opens a bottomless can of worms. What things depend upon is a more or less infinite set of conditions, exceptions to rules that are themselves exceptions to other rules. Start specifying exceptions and you end up chopping your laws into ever smaller bits. “When you’re not underwater, breathe.” “When you don’t have access to clean air, breathe.” “When you’re not underwater or surrounded by cyanide gas without access to clean air, breathe.”

Cyanide gas is rarely a factor in our environment, but you get the point. Try applying this type of rule to more pressing circumstances and the problem becomes obvious: “Always stay with your job/partner/career/campaign/role/doctor, unless . . . ” If you had time to think it through, the list of exceptions could go on and on.

As a practical matter, we get by just fine most of the time. Intuitively, we accumulate long mental lists of general rules (sometimes people go to bed early, so don’t call after 9 p.m.) and particular exceptions (you can call Mike up to 11 p.m., because he never goes to bed before then) and access them intuitively as situations arise.

And if you set aside the question of how your gut negotiates among clusters of rules, each particular rule can be thought of as a rule of the first type. “Always feel free to call Mike till 11 p.m.,” for example – the rule may always be true; it just doesn’t apply widely. Theories should be as general as possible but no more general.

We can think of the process of rule evolution as a sequence whereby you start with a type one, general rule to always do X. An exception comes up and you embrace the type two rule that doing X depends upon the situation. You use type three reasoning to come up with some new specific “in situation A do X” rule that becomes your new type one rule at a level of greater specificity.

Depending on where the exceptions arise in our life experiences, we end up with a range of generality from the coarse- to the fine-grained, anywhere from the very general (always trust God) to the very narrow (always feel free to call Mike after 11 on Mondays).

So when gut-crafting laws to live by, how specific should one be? Following Einstein, as general as possible but not more general. Possible for what? For success, of course.

The problem with using success as the deciding factor is that success is in the future, and you need to decide now. Einstein’s wisdom is good for pointing out the problem, which is designing rules that lead to future success, but not for saying what cues to go by today.

I have a fairly general rule I follow when driving. At green lights, go. A few months back, I was hit by a car running a red light. I needed a more specific rule at that moment that my gut didn’t come up with. “Always go on green except when a car is going to run a red light.”

Rules that are too simple can kill you. Rules that are too specific can too. And it’s easy to make mistakes, because only with future success or failure can you look back and see whether your rule was too narrow or too general, too fine-grained or too coarse-grained.

We debate granularity. You know what it’s like when someone rolls their eyes in disgust at what they consider to be too fine-grained an analysis. “Oh, c’mon, get a life. Quit micromanaging every detail!” And there’s the opposite as well: “Don’t generalize. The devil’s in the details!”

As an aside, it’s interesting that we have the pejorative term micromanagement, but you never hear talk about the perils of macromanagement.

We have intuitions about this granularity issue. For example, corporate decision-making procedures suggest the intuitive ranking:


There’s also a host of recipes for handling these in a simple top-down manner. Always start with your mission. Once you have it firmly in mind, then move on to your goals, and so on.

Sounds good on paper – but in practice, the process is more complex than that. After all, your mission is shaped by your capabilities, which are shaped by your actions.

I hope this one doesn’t strike you as too dry, because it has struck me in the past week that the hierarchy of perspectives from fine- to coarse-grained is the biggest “it depends” there is. Sometimes you should think more generally; sometimes more specifically – it depends.

In coming weeks I’ll show how fundamental this issue of granularity is. It’s at the heart of all that meta stuff I talk about, at the heart of all tough judgment calls, at the heart of faith versus reason, letter versus spirit, certainty versus doubt, me versus we. You’ll see.