Law of the Instrument

The Law of the Instrument

Author Abraham Kaplan describes what he calls the law of the instrument: “Give a small boy a hammer, and he will find that everything he encounters needs pounding.” This is not unlike the old proverb that says that when all you have is a hammer, every problem looks like a nail.

The implication is obvious: If you only have one framework through which to view the world, you will find that every problem you encounter seems to fit into the framework somehow. The world will narrow to fit your limited conception of it. And as Feynman described, if you simply never occupy a certain mental framework, you may live forever without knowing what a certain problem might have looked like through that lens.

Every model is a simplification—it crunches reality down so we can more readily manage it. But this is not a problem so long as we are aware of the limits of the models we’re using. We need to always be cognizant of the fact that we are using mental models in the first place, that these models are by their nature limited, and that we can always abandon our model for another one if it serves our needs better.

Remember the paradox of expertise? It may be that the deeper into one field our expertise grows, the more at risk you are of “confusing the map for the territory,” or assuming that your particular skill set is the answer to every problem you encounter. You apply it to everything. Confirmation bias allows you to remember all the times it works, so your assumptions get further entrenched, and when it doesn’t work, you conclude the problem is unsolvable. In other words, you have a hammer, and you go around bashing everything. One day you are challenged to do open heart surgery, but you kill the patient with your hammer and then say sadly, “Well, you can’t save everyone. There are limits to what medical science can do!”

Your mistake is to think that your model equals medical science. Your mistake was to not have a range of different tools at your disposal. Here, “tools” does not mean a flashy new app or gadget, an updated theory, or a machine that can do more than you can. This is just more of the same, a little like becoming really, really good at using a hammer. What you want to do is look beyond the hammer entirely—not just inventing literal new tools, but theoretical and philosophical frameworks that spawn the very idea of those new tools in the first place.

In the movie inspired by the life of the mathematician John Nash, A Beautiful Mind, Nash is attempting to hold a lecture while construction workers are making a noise outside. Nash closes the windows, but the students complain of getting hot. One of the students then gets up to open the window and politely asks the workers to take a break until the class is over. Nash then says, “Well, as you will find with multivariate analysis, there are a number of solutions to any given problem.”

Regardless of how accurately the movie depicts mathematical genius (probably not very well), the point is neatly made. One way of looking at the problem is to imagine it mathematically: The window can be open or closed, and there are two distinct problems with each possibility. But, this “either/or” mental model is not the only one we can use. The student shows that there is another one, where the construction workers outside are not taken as a fixed variable.

What’s more, the solution of the mathematician (close the window, get hot) is not necessarily the solution of someone with better social skills (negotiate and cooperate). The question of whether the window should be open or closed is most satisfactorily resolved when one relinquishes the mental model that says that open and noisy versus closed and hot are the only options possible.