The Power and Perils of Incrementalism

Start small. When starting something new, it makes sense to cut it down into easily manageable chunks, from anywhere to 5 minutes to an hour, that can be completed with relative ease. It doesn’t matter whether you are trying to build up your exercise capacity, writing computer code, reading a textbook, learning some new skill, or whatever. Everything, in the beginning, benefits from making the task small and fun. Completing it gives you a sense of accomplishment, that you are capable of fulfilling the task that previously you did not think you could do.

And, once started, there’s momentum. In a piece of code, you may start off doing something badly, but it works. Then, you’ll see some small way to improve it. Then, another, and another. Eventually, you get to the point where it looks like you knew what you were doing all along, and the task helped you to learn your way there.

But, there is another side of this kind of incrementalism. Invariably, your learn enough that your initial ideas and effort weren’t the best place to start. Or, your goals change. Something tends to happen that makes you want to completely refactor what you have been doing into something new. You’ll want to rewrite the code or essay. You’ll decide, now that you can run, perhaps you should run a marathon, as a challenge.

Partly, incrementalism gets us to the point where we have a skill, and we want to challenge ourselves, to do something bigger than what we could have imagined before we started. This is great, when it happens.

But, another thing sometimes happens too. We get complacent. Rewriting the code is a lot of work, and incrementalism is all about work, but in small size chunks. But, getting yourself in a mental mindset to redo your incremental work is the same as when you start out trying to learn something you didn’t know before. Except, now you have a better understanding of how much work is required, and it will be harder to just want to do incremental changes. You’ll want to do more, because you have the capability to do more. However, this desire also has a tendency to cut into your enthusiasm.

Why refactor the code, when what we have is “good enough” for most of our purposes? The calculus of benefit tends to run this way. Further, the more people are involved, the more inertia will set-in. This is why revolutions always require vanguards because its at the vanguard that the enthusiasm for wholesale change is nurtured and acted upon.

People Aren’t Perfectable

“Progress is a possibility for the animal: it can be broken in, tamed and trained; but it is not a possibility for the fool, because the fool thinks he has nothing to learn. It is his place to dictate to others and put them right, and so it is impossible to reason with him. He will laugh you to scorn in saying that what he does not understand is not a meaningful proposition. ‘Why don’t I understand it, then?’, he asks you, with marvelous impudence. To tell him it is because he is a fool would only be taken as an insult, so there is nothing you can say in reply.”

-Eliphas Levi

“Telling the truth to someone who can’t understand it is tantamount to telling that person a lie.”

-Eliphas Levi

I’d go further than Eliphas and say this isn’t a problem just of fools. It’s a problem for the vast majority of humanity. It’s a rare person that is prepared to hear anything different than what they already think they know.

One of the things that most people seem to believe in is progress. In it’s most generic format, it’s doing well in school, getting a job, getting married, having children and so forth. If you look at it as a sequential timeline with markers to be hit, then it looks like progress. But, is it?

Let’s say we change the measuring stick to include various types of intelligence: analytical, emotional, social, psychological, et al. Are people that are older more advanced in these modes of intelligence?

Perhaps in some ways. It is probably true that, as we age, we refine out mental models for how the world works because we’ve done a lot of reality testing of different models and have found some that work much better than others. It’s probably also true that different social experiences have provided some depth in our ability to be graceful in a social setting, at least most people.

But, at the same time, it is also clear that there is accumulated damage that works against this notion of progress. Clearly, as we age, we become subject to a whole host of ailments that could probably be described as a general decline for most people after their twenties. For example, 20% of Americans aged 65 or older don’t have a single tooth. Or, 40% of Americans aged 80 or older have some level of dementia. Clearly, aging is not physical progress.

Does aging allow for other types of development? For example, does cognitive decline open up some other previous state, A Flowers for Algernon in reverse, where someone burdened with worries becomes care free? I’m sure it happens, just as I’m sure it happens only rarely.

When I think about the people I know, I don’t see strong signs of progress. I see a give and take, where certain qualities tend to be at a set point, which can be influenced, up and down, by effort, circumstances and other factors. But, I think the pattern I most commonly see is that people rarely change dramatically.

And, if we cannot make progress, what then of perfection? If only we were 10 pounds thinner, finally learned calculus, got in tune with our emotions, got the nose job or mole removed, found the “perfect” mate, won the lottery – you pick it, but it seems that a lot of people have notions about what is missing in their lives, and if they only had those things, their lives would be perfect.

Progress or perfection is probably a bad mythology to undergird our perceptions of the world with. Perhaps, it is time to accept the fact that we got dealt a hand, and it’s only going to get worse from here, a regression model, if you will. Then, we might think more in terms of maintenance, rather than perfectability, which strikes me as a better mental model.