Keeping up the pace
In the wonderful book Inventors at Work, there’s an interview with Wilson Greatbatch, who invented the first practical implantable pacemaker. After describing the hours that he spends tinkering at his workbench—“I put something together, test it out on the scope, and see what it does”—Greatbatch is asked whether carefully planned experiments are more important than sudden intuitive leaps. He responds:
I’m a proponent of the big jump. I like to throw something together, see if it works, and go on from there. Later, I might go back and fill in the gaps, but I might even let someone else do that.
One problem I have with the people I work with at universities is that they like to work step by step by step. And in their view, you don’t start one step until you’ve finished the last one. But there’s usually a place in any project where you can say, “Well, maybe if I just built this thing, I could jump way over here.”
In the case of the pacemaker, Greatbatch continues, “the big jump” was a matter of throwing a bunch of parts together and verifying that you could touch the wires to a dog’s heart to make it beat. (It’s also worth noting that his first key insight was the result of an experimental accident.) He goes on:
After that jump, I could go back and fill in different details. What kind of materials can be used in the body? What kind of circuitry can be used? A pacemaker is nothing more elaborate than a flasher that you see on a highway construction site. But you’ve got to redesign that flasher so that it will work off its battery for ten years instead of only a few nights. You’ve got to wrap it in something that the body won’t reject, like silicon rubber or platinum or stainless steel. You have to find what’s right and what isn’t.
All those details are important, but if I’ve at least made the big jump, I know that I can make the heart go with a tiny pulse of electricity.
And it’s revealing that Greatbatch contrasts this approach with the one that he observes in his academic counterparts: “They essentially lay out a research project ahead of time and call it an experiment.” Most extended projects of any kind consist of two alternating phases. The first consists of large conceptual leaps that focus on the big picture and can’t be predicted in advance; the second of the long stretches that are less a matter of transformation than of aggregation, as you fill in the gaps and gather the data. Both types of work are necessary, and there’s a very real sense in which each one depends on the other. Without a few intuitive jumps, you won’t know what you’re supposed to be researching in the first place, but those insights usually come into being only after you’ve laid some of the boring groundwork, and they often emerge in the act of gathering the raw material itself. The danger is that the aggregation phase can last forever. There’s something oddly hypnotic, even seductive, about the act of research. You can see that you’re making progress; you’re adding to your collection of information; and there’s always more to be done. It’s incontestable that you’re moving forward, however incrementally, and there’s no pressing reason to stop. If you’re mostly interested in the work for its own sake, you can continue it indefinitely—which is why it’s possible for a graduate student to spend a decade or more on a doctoral thesis that could have been polished off in a couple of years.
So how do you prevent yourself from falling into an intellectual holding pattern? The obvious answer is to come up with a big conceptual breakthrough that will allow you to shape or circumvent some of the research, guiding you toward promising paths and cutting off other avenues before they take up too much of your time. The trouble, of course, is that you can’t will yourself into having a transformative insight, although you can certainly foster the conditions in which such ideas are more likely to occur. This points to a second strategy that I’ve come to regard as essential: you set a deadline for when you think you’ll be done with your research, and then you move it forward by a third. For example, if you think you’ll need about six weeks to finish your background reading, give yourself four weeks, and stick to it. The actual schedule doesn’t matter as much as the fact that it needs to hurt a little. Once you force yourself to think big, you’ll generally find that you didn’t need all that research time anyway, and if you do need to fill in some of the blanks, you have a much better sense of what actually matters. At the end of this week, for instance, I’m forcing myself to make a hard stop on the research for my nonfiction book Astounding and start the outlining phase. It stings, but I also know that I’ll be better off if I just throw together what I have now and see how it holds up. Maintaining that steady pace is important—but sometimes you need to jolt yourself into a new pattern.