What a self-taught miller can tell us about what actually separates successful researchers from everyone else.
Nottingham, sometime around 1823. A miller named George Green climbs the stairs of his father’s windmill after a full day’s work, hauling sacks, tending the grindstones, breathing flour dust, and sits down under a candle with a borrowed mathematics journal he can barely translate. He has had, in total, about a year of formal schooling in his life, most of it before age nine. There is no one in Nottingham who can check his work. No advisor, no seminar, no colleague down the hall to tell him whether the strange new object he keeps running into, something he starts calling a “potential function,” is real mathematics or a mistake he doesn’t yet know how to see. He does this for years, alone and unpaid, with nothing but his own doubt to referee the work.
In 1828 he pays out of his own pocket to print copies of an essay containing what we now call Green’s theorem, a result so central to physics and engineering that you cannot get through a course on electromagnetism or fluid dynamics without meeting it. Almost nobody reads it. It sits forgotten for over a decade until a young William Thomson, later Lord Kelvin, stumbles across a reference to it, hunts down one of the few surviving copies, and realizes what Green had actually built, alone, with no one watching.
We know exactly why most people don’t do what Green did. It’s genuinely unpleasant to work for years on something with no external confirmation you’re not wasting your time. It’s unpleasant to study material alone, with no one to ask, and to grind through the tedious verification of your own reasoning because you’re the only check you have. Most people want at least occasional evidence they’re not fooling themselves, a grade, an advisor’s nod, a paper accepted. Green had almost none of that for the better part of two decades.
In 1940, an insurance executive named Albert Gray gave a talk trying to name the one thing shared by every successful person he’d studied. He landed on something almost too plain to be satisfying: successful people form the habit of doing the things that failures don’t like to do. Not that they enjoy those things more, Gray is emphatic on this point. They dislike them exactly as much as everyone else. The difference is that they do them anyway, repeatedly, until doing them stops requiring fresh willpower every day and becomes closer to automatic.
Research mathematics is an unusually pure test of this idea, because the unpleasant tasks are so specific and so hard to avoid. There’s the discomfort of sitting inside real confusion, not the kind that resolves in twenty minutes, but the kind that might not resolve for a year, with no proof along the way that you’re moving toward anything. There’s the tedium of grinding through small cases by hand to check a claim that “obviously” holds, because obviousness isn’t proof and the whole discipline runs on that difference. And there’s, perhaps hardest of all, the willingness to throw away weeks of work on an approach you’ve grown attached to, because it isn’t working, when starting over is always available and always the thing you’d rather not do.
None of this is really about who has more talent. Among people who already have enough of it, the difference tends to be who can tolerate this kind of discomfort on a schedule, daily, not once, until the tolerance itself becomes unremarkable, the way Green’s climb up the windmill stairs must have stopped feeling like an act of heroism and started feeling like Tuesday.
It’s worth being honest, too, about what this isn’t: it isn’t a claim that patience is unlimited, or that institutional pressure doesn’t matter. A postdoc with a two-year contract, or a professor buried in committee work, doesn’t have Green’s freedom to let a question sit for a decade. The point isn’t to wait indefinitely, it’s not to mistake the ordinary discomfort of real work for evidence that something’s gone wrong.
Gray adds one more piece that matters as much as the habit itself: it only survives attached to a purpose strong enough to outlast the discomfort, and he insists that purpose has to be emotional, not merely logical. “I need publications” is logical. It collapses exactly when the work gets hard, because a purely logical need stops pushing you the moment it’s minimally satisfied. Green’s purpose was closer to needing to know whether the object he’d found was real, a question that apparently mattered more to him than the total absence of anyone to share it with. That kind of purpose doesn’t get satisfied by minimal effort. It keeps pulling.
This reframes a gap that gets talked about constantly in academic life and rarely examined honestly: the one between “talented” and “productive.” Talent gets you noticed early. It doesn’t, by itself, get you through the eleventh unproductive month on a hard problem, or the fortieth rejected draft, or years of work nobody outside your own head has confirmed is worth anything. What gets you through that is closer to what Green had, not less discomfort, but a habit of moving through it anyway, in service of a question he actually needed answered.
Green never saw most of his own influence. He died in 1841, thirteen years after printing an essay almost no one read, five years before Thomson would find it and understand what it was. He wasn’t waiting for that confirmation. Whatever purpose had him climbing those stairs each night, it was strong enough on its own.