What Is Ours to Change

Take stock of a life that needs adjusting and you'll probably arrive at the same short list everyone arrives at, without ever being taught it. After a bad year, or at the start of a new one, four things show up as the levers within reach: what I do, how I take things, where I am, and who I'm around. Behavior, attitude, environment, people. It doesn't feel like a theory. It feels like a description of the controls—as if the self came with a small panel of switches, and these are the ones that aren't painted over.

The intuition is old. The Enchiridion of Epictetus opens by sorting all of existence into two bins—what's up to us and what isn't—and tells you to spend your energy entirely on the first. The sort has survived two thousand years because it's broadly correct. But the folk version most of us carry—the New Year's-resolution version, the self-help version—is dangerous in a precise and mostly invisible way. The danger isn't that the sort is wrong. It's that we almost always run it in the wrong direction: turned outward, onto other people, where it quietly converts their circumstances into their character.

So this is about how the four levers reveal who we are—and about the exact moment that revelation turns into a weapon.

Two Different Claims You Keep Confusing

Start with a distinction the folk version collapses. There's what a person can actually change, which is a claim about the world. And there's what a person believes they can change, which is a claim about that person's mind. These aren't the same, and almost everything interesting lives in the gap.

A belief that tracks reality reveals one kind of interior. The exact same belief, held against reality, reveals something else—often the opposite. So every lever has to be read twice: once for the person whose sense of their own power is accurate, and once for the person whose sense of it isn't.

Now lay the four levers out by how much control they actually give you. They don't come out flat.

Behavior: Real, but Slippery

Behavior and attitude sit closest to the self. Bandura's work on self-efficacy—your belief in your own capacity to carry out an action—is basically a theory of the behavior lever, and it explains why this one feels so immediate. A discrete act is under near-instant command. You can stand, refrain, speak, begin.

The catch is as old as Aristotle's worry about akrasia, weakness of will. Patterned behavior—the habit, the compulsion, the thing you swore off Tuesday and resumed Wednesday—answers to the will far less reliably than the single act does. So the lever is real but slippery: high command over the deed, much lower command over the groove the deeds have worn.

Attitude: Sovereign in One Direction, Useless in the Other

Attitude is the lever the Stoics prized most, and modern cognitive therapy turned their insight into a procedure: it's not the event but your appraisal of the event that does most of the damage, and the appraisal can—with effort, over time—be revised. That's the engine of CBT, and in a different key of acceptance and commitment therapy, which swaps "change the feeling" for "change your relationship to the feeling." Viktor Frankl gave it the limit case: writing from the far side of the camps, he insisted the last freedom no one can strip from you is the freedom to choose your stance toward what's happening.

Here's the subtlety the Stoics overran. This lever is powerful in one direction and feeble in the other. No external force can stop you from reframing—in that sense attitude is more sovereign than behavior. But no act of will flips a mood on contact. The immediate spark, driven by biochemistry and history, isn't on the panel at all. You govern the slow interpretive layer, not the fast one.

Environment: The Lever That Isn't the Same for Everyone

Environment is the first lever that points outward, and its real reach varies more wildly between lives than any other. Someone with money and mobility commands an enormous amount of their surroundings. Someone in poverty, or in a cell, commands almost none. Rating this lever the same for both is the first quiet falsehood the framework tells.

But for those who can reach it, environment carries a hidden multiplier—the one Thaler and Sunstein's work on choice architecture makes explicit. The most reliable way to change behavior usually isn't to summon more willpower. It's to redesign the context so the willed thing becomes the easy thing. Environment is, among other things, how you move the behavior lever without grinding your gears against it.

People: The Lever That Splits in Two

The fourth lever—the people around you—is the most revealing of the set, and it splits the moment you press on it.

In one sense it means transforming the people themselves: changing who they fundamentally are. Here the control available is close to zero. You cannot will another adult into a different character, and the belief that you can is one of the most dependable sources of human misery—the engine of codependency, of the rescuer who'll fix him, of the controller who mistakes proximity for permission. You can influence: model, persuade, change the incentives, set boundaries that change what behavior pays. But influence is not transformation, and confusing the two is the whole error.

In the other sense, the lever means selection: not changing who people are, but changing which of them you keep close. That version is genuinely within reach, and it's the high-agency move—more so given what the Harvard Study of Adult Development, the longest study of its kind, keeps finding: the quality of your relationships is among the strongest predictors of how your life goes.

So "I can change the people around me" is either one of the healthiest things a person can believe or one of the most corrosive—depending entirely on which of the two it means.

The Goal Isn't Belief in Your Power. It's Accuracy About It.

Notice what the gradient implies. The aim isn't to believe maximally in your own power. It's to believe accurately in it. The operative virtue is calibration—the match between your felt sense of control and the real distribution of it.

Covey's image is the same idea in a popular register: a Circle of Concern holding everything you care about, and a smaller Circle of Influence holding what you can actually affect. Maturity is the steady migration of effort from the outer ring to the inner one.

Read this way, the framework becomes a diagnostic—and a set of recognizable portraits falls out of it, each one a particular mis-aiming of agency:

• The one who pours belief into transforming other people while neglecting their own behavior and attitude—agency inverted, spent on the least controllable lever, condemned to a frustration that's structural rather than incidental.
• The one who believes only attitude is changeable and treats everything else as fixed—running the Stoic move so far it becomes quietism, reframing a situation they could simply have left, calling avoidance wisdom.
• The willpower moralist, who thinks behavior yields to nothing but force of character, white-knuckles every change, and reads each failure as a verdict on his worth.
• The serial restarter, forever sure the next city or job is the one, working the environment lever obsessively and never noticing the constant across every fresh start is himself.
• The fatalist, who believes nothing on the panel responds at all.

It's an elegant instrument. You can hold almost anyone up to it and get a reading. And that's exactly where the trouble starts.

The Cruelty in the Machine

Watch what the framework licenses once you treat it not as one lens among several but as the truth of the matter. Every one of those portraits is a diagnosis. And a diagnosis of someone who believes the wrong things about their own power sits one short, frictionless step from a verdict on someone whose life just went badly.

The fatalist "has a victim mentality." The person stuck in a punishing environment "could leave if he really wanted to." The exhausted person who's stopped trying "lacks self-efficacy." Held as bedrock, the map performs a silent alchemy: it turns circumstance into character. It hands the comfortable a clinical-sounding vocabulary for reading the constrained—and it does it wearing the dispassionate face of psychology.

That's not a reason to throw the framework out. It's a reason to find out what it can survive. There are three serious objections to the whole agency-centered apparatus. The honest move isn't to fend them off—it's to let each one in at full strength and see what shape the framework is in when they're done.

Objection One: There's No Control to Calibrate

The most abstract attack is the determinist's. If behavior and attitude are themselves outputs of genes, upbringing, and a long chain of prior environments—if the person doing the deliberating is as caused as everything else—then "control" is a folk fiction at every level. The whole controllability axis is smuggled metaphysics in a lab coat. There's no panel of switches. There's only physics, and the feeling of choosing is what the machinery feels like from inside.

The answer is the compatibilist one, worked out by Hume and sharpened in our time by Dennett. Control is a real feature of the world at the level where it actually lives: deliberating agents choosing among options—not particles obeying laws. Both descriptions can be true at once. That a choice is caused doesn't make it not a choice, any more than a calculation running on caused circuitry makes it not a calculation. The framework never needed libertarian free will, the spooky kind that breaks the causal chain. It only needed deliberation to make a difference—and it manifestly does. The person who weighs the options ends up somewhere different from the person who doesn't.

So determinism doesn't dissolve the map. It relocates it: the map is drawn at the scale of persons, where talk of control is the right vocabulary, not at the scale of mechanism, where it's the wrong one.

But the objection leaves a mark, and the mark points forward. If your very capacity to deliberate well was handed to you by a childhood you didn't choose, then the credit you take for being well-calibrated—and the blame you assign others for being poorly calibrated—are both shakier than they looked. The determinist doesn't destroy the framework. He installs the first dose of humility in it. That humility matters enormously by the end.

Objection Two: The Self in the Framework Doesn't Exist

The second attack comes from the relational traditions, and from the Southern African philosophy gathered under the word ubuntu—rendered by Mbiti as "I am because we are," given its best-known voice by Desmond Tutu: a person is a person through other persons.

The framework, this objection runs, assumes a bounded, separate self—one that can draw a clean line between "my attitude" over here and "the people around me" over there. But if the self is constituted by its relationships rather than merely surrounded by them, that line is an artifact. You can't fully own an attitude your community placed in you before you were old enough to consent. The first two levers and the fourth aren't separate controls. They're aspects of one woven thing.

The right response is to concede this completely—and then watch the objection strengthen the framework instead of breaking it. If the self is co-authored, two things follow that the framework should have said louder from the start.

First, the people lever was never really about reaching into separate others and rearranging them. It was always about the web you're suspended in—which is exactly why selection is the powerful move and the fantasy of transformation is not just low-yield but conceptually confused.

Second, and bigger: if your attitude is partly given to you, then the operative virtue can't be raw will. It has to be calibration—the accurate, hard-won sense of which part of a disposition you were handed is actually yours to revise, and which part is load-bearing structure you didn't build. The relational critique, far from refuting the idea that calibration is what matters, turns out to be the strongest argument for it. A self partly made of others needs, more than anything, an honest accounting of where the seams are.

Objection Three: The Map Convicts the People It Can't See

The strongest objection is the structural one, and it's strong enough to be the hinge the whole thing turns on. The framework, this attack holds, responsibilizes individuals for conditions that aren't individual at all.

The evidence is in the very literature the framework leans on. Rotter's research on locus of control did find that an internal locus—the sense that outcomes follow from your own actions—correlates with better results. But critics have pressed for decades on what runs through that correlation. The materially secure tend both to develop an internal locus and to enjoy good outcomes, for the same underlying reason: their environment is genuinely responsive to their actions. Push on the world and it reliably moves, and you learn two things at once—that pushing works, and the rewards pushing earns.

For someone whose environment is genuinely unresponsive—by poverty, by discrimination, by confinement—an external locus isn't a cognitive distortion to correct. It's an accurate reading of a world that does not, in fact, answer when they push. The diagnostic move that says "underestimation equals learned helplessness" can pathologize correct perception—scolding someone for seeing their situation clearly.

Sen and Nussbaum's capability approach gives the point its sharpest form. What matters for a human life isn't formal agency in the abstract but real capability: the actual set of things a person is in a position to do and to be, given the resources and freedoms genuinely available. Two people can hold identical beliefs about control and stand in front of radically different capability sets. Telling the second of them they have the same four levers as the first is simply false—the levers are bolted to a panel they can't reach.

And taking their failure to pull a lever they don't have as evidence about their character runs two of social psychology's best-documented errors at once. The just-world hypothesis—Lerner's finding that we're powerfully motivated to believe people get what they deserve. And the fundamental attribution error—our standing tendency to explain other people's outcomes by their dispositions while explaining our own by our circumstances. We blame the constrained person's plight on who they are; we'd blame our own identical plight on what happened to us. And we make the error asymmetrically—most readily, and most cruelly, against the people who have the least.

This is the cruelty from earlier, now named precisely. It isn't a flaw at the edge of the framework. It's what the framework becomes the instant you apply it across a difference in power without first asking the only question that matters: what was actually in this person's reach?

What the Map Is For

So where does the apparatus stand once all three objections have done their work? Intact—but demoted, and re-pointed.

Demoted from bedrock to instrument: not the truth about persons, but a tool with a proper use and a corresponding misuse. And re-pointed, because the proper use was always the first-person one. The dichotomy of control was built as a discipline for the self—a way for a single person to ask, of their own life, whether their felt sense of their power is calibrated to their real situation, and to stop spending finite agency on what won't move. Used that way, it's one of the most useful things you can carry.

The misuse is the turn outward as an audit. The moment you lay the map over a second person—they just won't take responsibility, they could change if they wanted to—it stops describing and starts convicting, and it convicts hardest exactly where it knows least.

This is the deepest reason the founding distinction isn't a technicality. It's the entire safeguard. The framework rests on separating the perceived from the real. Applied to yourself, you have privileged access to both: you can feel your sense of your own power, and you can test it against a situation you know from the inside. Applied to anyone else, you have access to neither. You see their behavior but not their capability set. You see their attitude but not the environment that pressed it into them. You see the output and almost none of the constraints. Run the diagnostic on a stranger and you're filling in everything you can't observe with assumptions—and the just-world reflex guarantees what those assumptions will be.

Epictetus, worth remembering, was born a slave and wrote as a freedman. He knew the boundary between what the will can reach and what it can't—not as a thought experiment, but from the inside, which is the only place the boundary can be honestly drawn. The error was never in sorting the world into what's ours and what isn't; that sort is the start of any serious life. The error is forgetting that the sort can only be done truthfully from within a single life—and that when we run it on others, we're almost always sorting their circumstances into their character, then calling the verdict justice.

The four levers are real. They're just, and importantly, only ever ours to count from where we're standing.

This week: Catch yourself running the audit outward—"he could change if he wanted to," "she just won't take responsibility"—and ask the one question the verdict skips: what was actually in that person's reach? Then turn the same four levers on your own life, where you can answer honestly.