"I wish you'd take more ownership." "It's an attitude problem." "I'm not feeling much drive from you."

When I go into organizations to help with AI and DX, I often watch the explanation for why work isn't flowing land on words like these — words about "mindset" and "attitude." A problem that could be solved by changing a procedure or redividing responsibilities gets talked about as a problem inside someone's character. And, strangely, the person on the receiving end usually can't argue back.

They finish everything they were asked to do, and the reply comes anyway: "That's not the point — it's your attitude." They can point to deliverables; the other side questions their inner state. The two never meet. And somewhere in that mismatch, it is always the person who was criticized who ends up on the back foot.

This essay takes that mechanism apart. Why is a criticism aimed at attitude or mindset so strong, and so hard to rebut? And what is that strength doing — whether or not anyone intends it — inside an organization?

Let me put the skeleton of the conclusion up front. "Attributing outcomes to attitude" is not a mere misunderstanding or a turn of phrase. It has a structure that pushes the entire cost of rebuttal onto one side. That structure assembles itself almost automatically on top of a quirk in how we think. And under certain conditions, it runs harder still. I'll take it apart step by step.

One thing before we start. The goal of this essay is not to find the "bad manager" who talks about attitude and beat him with it. As we'll see in the second half, the moment you do that, you fall into the very error you were criticizing. The aim is to look at the structure — why this phenomenon happens so naturally, and so forcefully. I want that stated first.

The situation disappears, and only the character is left

The starting point is a quirk that social psychology has described for a long time: the fundamental attribution error, or FAE.

The definition is clean. When we explain another person's behavior, we underweight the situational and environmental factors and overweight the internal factors — their personality, disposition, attitude. That is the FAE. The term was coined by the social psychologist Lee Ross in 1977, and if you trace the idea to its source, the observation that people tend to read others' behavior as a reflection of their inner states goes back to the work of Heider, Jones, and others in the 1950s and 60s.

Here is a concrete case. Someone shows up late to a meeting. The observer often concludes, almost reflexively, "that person is sloppy," "they're the type who's loose with time." But in fact the cause may have been situational — a heavy workload, a problem that blew up right before, something in their personal life that had to be handled. The observer doesn't enter those external factors into the ledger, and from a single behavior — being late — reads off the person's character.

What matters is that this is not the special behavior of a malicious person. The FAE happens to everyone, as an entirely ordinary working of cognition. Why does it happen? A few reasons have been pointed to.

One is that we assign causes to whatever stands out. In the observer's field of view, the thing that stands out most is the acting person. The situation around them — workload, deadlines, how the organization is designed, pressure from others — melts into the background and is hard to see. Attributing a cause to what is salient is a natural tendency of cognition.

Another is that we want to file complicated things away simply. "The division of roles in the organization is vague, and no one has defined where responsibility begins and ends" is a correct explanation, but it is heavy and hard to handle. Compared with that, "he just doesn't take enough ownership" is light, nameable, and instantly sayable. Cognition often chooses lightness over accuracy.

For the record: how strong the FAE is, and how constant it is across cultures and situations, is itself debated within psychology. Recent work argues it "swings quite a bit depending on the situation." I'll treat the FAE here not as an ironclad law but as a strong tilt — something that happens easily if left alone.

What happens when those two combine? The hard-to-handle situational factors vanish from view, and only the easy-to-handle character factor is left. That is the FAE. And the instant a cause is attributed to character, the conversation moves into territory that is hard to verify. Workload and mishaps can be checked. "Attitude" cannot. This is the first mechanism.

Why "change your mindset" changes nothing

That attitude attribution is hard to rebut has a deeper implication. Let me step away from the workplace and look at it through the world of safety — a field that has, in fact, been studied closely in Japan too.

After an accident, organizations often call for "raising safety awareness," for "everyone taking safety as their own concern." They post slogans, chant them at morning meetings, demand a lift in awareness. And yet the people who study safety have handed down some harsh verdicts on this kind of campaign.

Think about it: have you ever met someone who genuinely believes that safety is not an important part of the job? Almost no one. Nobody thinks safety is unimportant. Which means "raise your safety awareness" is stating a proposition that has no opponents. Everyone already agrees. And precisely because everyone already agrees, it moves nothing. Reducing accidents requires more than changing awareness — redesigning procedures, retrofitting equipment, reworking the physical flow, remaking the situation itself. The moment you call it "a problem of awareness" or "a problem of culture," you get to skip that troublesome remaking.

This is not my own bright idea. In Managing the Risks of Organizational Accidents, James Reason divided how we face accidents into a "person approach," which blames the individual, and a "system approach," which fixes the mechanism. The person approach locates the cause in individual inner states — carelessness, inattention — and tries to deal with it through warnings and reprimands. But that doesn't reduce accidents; the cause, in most cases, lies in the mechanism behind the person. That is Reason's central claim. In Japan too, Shigeru Haga has written repeatedly about how awareness-raising campaigns spin their wheels, and Toru Nakata about how exhortations like "do it properly" and "be more mindful" are not countermeasures at all. "Awareness doesn't reduce accidents" is, in this field, close to settled.

The very same structure plays out, unchanged, in the workplace's "ownership" and "drive." No one publicly declares that they'd rather not take ownership. So "take more ownership" is a hollow proposition that cannot be opposed. Because it is hollow, it is unbeatable as a criticism. The person criticized has no move but to agree. All they can say is "Yes, I will." And when the same situation comes around again, the same criticism flies again. Nothing changes; only the criticism repeats.

Here there is an important distinction we must not blur. The FAE is a cognitive quirk that happens inside an individual's head. That it then becomes institutionalized inside an organization as a way of talking — "it's a problem of awareness" — into a mechanism that excuses structural improvement: that cannot be explained by the cognitive quirk alone. That requires another force — who holds the authority to change the situation, whose responsibility gets to go unquestioned, what incentives are at work in the evaluation. The cognitive quirk only supplies the fuel; what assembles the device itself is the structure of authority and incentives. I'll keep these two stages from being spoken of as one.

With that in place: why does the "problem of awareness" way of talking survive so well in organizations? This is where the essay meets the arguments of Yuki Honda on the exploitation of "meaningfulness" and Ryosuke Ishii on psychological safety. Honda pointed out that a way of talking that pins labor problems on the worker's "motivation" or "sense of purpose" conceals the responsibility the employer side was supposed to carry. Ishii argues that the focus of management should be placed on "behavior," not on "what's in the heart" (drive, confidence). The direction is the same in each case: a way of talking that locates the cause inside a person is convenient for the side that holds responsibility for changing the situation.

But what this essay wants to say lies beyond that. These arguments have carefully drawn the picture up to "awareness won't change it" and "look at behavior, not inner states." Yet why that "problem of awareness" criticism is so strong in the moment of argument, why it bounces back every rebuttal — the nature of that strength itself, I feel, has not been put into words much. That is where we go next.

Why it becomes "an unbeatable criticism"

The peculiar strength of attribution to attitude comes into focus cleanly if we borrow one tool from the philosophy of science: the idea of falsifiability.

Karl Popper held that a condition for a claim to have meaning is being able to say what would be observed if it were false — that is, being falsifiable. Conversely, a claim that no possible fact could ever refute looks strong at first glance, seems able to explain anything, and is in fact hollow. Being able to explain anything is the same as predicting nothing.

A criticism of "attitude" has exactly the shape of this unfalsifiable claim. When you are told "you lack ownership," how would you go about denying it? Answer "I do have it," and it ends with "do you, really?" No matter how much you deliver, the reply comes: "you produced results, but there's a problem with your posture." There is, in principle, no fact that can break this criticism. So in an argument it looks unbeatable. But by Popper's criterion, it is not unbeatable — it is empty. Because it states nothing, there is nothing to refute.

I don't want to settle this by saying "inner states can't be observed by anyone, so they can't be falsified." That is too strong. If you commit to that, then the explanation this essay itself uses later — "the observer had negative feelings" — is also a claim about someone's inner state, and becomes equally unfalsifiable. So I'll put the point in a much more modest place.

The problem is not the philosophical question of whether something is observable. It is that the labor of rebuttal is loaded unfairly onto one side only. For a criticism about facts, the one making it also bears a duty to show "is that fact true?" But for a criticism about attitude, that duty moves wholesale onto the criticized. You are told "produce evidence that you don't have it," when neither having it nor not having it can be shown by evidence. The proof of a negative is always shouldered by the same side. That is the nature of the strength. You can't win — not because the other side's claim is correct, but because the distribution of the burden of proof was tilted from the start.

It runs hardest against someone you dislike

So far I've drawn the FAE as an "almost automatic quirk that happens to everyone." That is a deliberate framing. The frightening thing about the FAE is that even a non-villain commits it. Again: this is not a story about "a mean-spirited boss."

But there is a condition I have to add honestly. There are conditions under which the FAE runs harder. One of them is when the observer already holds negative feelings toward the other person.

Psychology has a concept called motivated reasoning. Organized by Ziva Kunda in 1990, it holds that when people have a conclusion they already want to reach, they tend to select the evidence that supports it and downgrade the evidence that doesn't. If the pure FAE is "carelessly overlooking the situation," motivated reasoning is "conveniently overlooking the situation." The same attribution error, but at a different temperature.

An observer who already harbors negative feelings toward someone is more likely to have a motive to subtract the situational factors from that person's behavior. If they can conclude "that behavior isn't due to the situation, it's due to their character," their own negative feelings are justified. And here, attribution to attitude — precisely because it is hard to rebut — offers just the right shape for justifying an emotion. Even when the negative appraisal can't be sustained in the realm of facts, move it into the realm of attitude and it can never be refuted. Throw the unverifiable appraisal "but there's a problem with your posture" against the verifiable fact "the role is being fulfilled," and you can deduct points without limit.

But here, to protect the axis of the whole argument, there is something I must strongly restrain.

The moment you conclude "this happened because that boss had bad intentions," you yourself — the one who reached that conclusion — are committing the exact same FAE.

This is the biggest pitfall a writer on this subject falls into. "Because the boss is mean," "because that person hated me" — those explanations are themselves textbook FAE: ignoring situational factors and attributing the cause to the other person's character. You think you're criticizing the device, and you set the same device running.

So I treat motivated reasoning strictly as "a general condition under which the bias grows stronger." I will not descend to the individual character verdict "this boss held malice." If anything, the accuser too should turn an eye toward the situation. That boss may himself lack the authority or the vocabulary to change the situation. He may be exposed to pressure from further up, taking someone else's FAE in turn. This essay places its weight on writing the defense of the criticized side, so the description naturally leans toward the subordinate. But that asymmetry is a deliberate choice — it does not mean that the species called "boss" is excused from the situation. When emotion comes first, the attribution error grows stronger: I stop at the altitude of that general rule.

Not everything is a device — drawing the boundary

Here, so that this essay itself does not fall into the same trap, I have to draw a boundary.

Read this far, it might sound like the conclusion is "any mention of attitude is an unjust trap." But that goes too far. If I declared "every criticism of attitude is an FAE device," then this essay would itself become an unfalsifiable claim — able to answer any counterexample with "that too is part of the device." I would fall into the structure I criticized.

In reality, dispositions and tendencies do exist, and they are useful for prediction. Sometimes the phrase "lacks ownership" correctly points at a real behavioral pattern — never raising a problem, doing the bare minimum, leaving those around to keep picking up the slack. Not all candid feedback given to help someone grow is an evasion device.

So when does a mention of attitude become a problem? The line can be drawn like this: when there are means actually available to change the situation, and instead of using them one escapes into talk of attitude, and moreover that criticism is used in an unfalsifiable form. Only then does attribution to attitude become a device of evasion. Conversely, if you name specific behaviors, show when and what and how something fell short, and think through the fix together, that is no longer talk of attitude — it is talk of facts and behavior. With the very same word, "ownership," one use becomes defense and the other becomes development. The dividing line is whether it has been put into a verifiable form.

"Highly conscientious" and "lacking the right attitude" are two sides of one coin

Let me draw one more line to widen the view. The FAE is said to have cultural variation.

People from so-called individualistic cultures — which prize individual achievement and autonomy — are said to be more prone to the FAE. The tendency to seek the cause of behavior in the person's own inner state is stronger. People from cultures that prize the group and relationships, by contrast, enter more of the situational factors into the ledger, and are said to be less prone to the FAE. These are findings shown by Miller in 1984 and by Morris and Peng in 1994, each comparing the United States with other cultures. That said, how strong this cultural difference is, and how far it generalizes, is also debated.

On that basis, from here on I write as my own interpretation. Many of the attitude words — "ownership," "give first" — were born and polished in individualistic cultures, the very soil that is prone to the FAE. When such words are imported and transplanted into a workplace with a different context, it may be that not only the words but the very framework of attributing behavior to individual inner states is carried in along with them.

If so, then the circulation of attitude words is not merely a matter of adding more encouraging vocabulary. It roots a habit of attribution — seeking the cause in the individual rather than the situation — as the workplace's common language. The praise "highly conscientious," "takes ownership," and the reprimand "lacks the right attitude" are two sides of one coin. Both stand on the same premise: that the cause of behavior lies in the individual's inner state, not the situation. In a workplace where that premise has become the common language, the FAE is no longer an individual's quirk but the group's mode of thought.

The ground of facts, so you don't get off the platform

So what can the criticized side do against this structure?

First, let's confirm what must not be done. You must not answer attribution-to-attitude with attribution-to-attitude. Told "that person has a bad attitude," to fire back "no, you're the one with the bad attitude" does not stop the device — it just builds a second one. The instant it becomes character against character, both parties sink into an unrefutable swamp. And in the swamp, the weaker party usually loses.

The foothold is staying on the ground of facts. Attitude can't be checked, but behavior and facts can. Was the required role fulfilled? Was the agreed deliverable shipped? Was the defined scope met? All of these can be shown as fact. If the other side talks attitude, you keep pulling the point back into checkable territory: "this isn't about posture — it's about what was delivered and to what extent, right?"

But honestly, this counter has conditions under which it works. Those are when the relationship is at least somewhat equal, or when a neutral third party — HR, records, contracts — is present. In a relationship where one side unilaterally holds the power of evaluation, you can line up all the facts and the appraisal or promotion decision still gets pushed through as it was. More troubling still: the very act of pulling the point back — "let's return to the facts" — can itself be reinterpreted as "that's defensive; that's proof of your lack of ownership." The counter gets turned into fuel for the accusation.

So let me estimate honestly what staying on the ground of facts is worth. Before an unconstrained power of evaluation, it is not a means of protecting the outcome. What it can protect is your own understanding, and a trail left in the record. Not losing your own coordinates — "I fulfilled my role; this was merely swapped out for a story about attitude." And keeping the exchange as a record. That, in itself, guards at least your own footing against being unfairly and repeatedly docked in a realm that can't be falsified.

One more thing. If you ever stand on the attributing side — that is, when you feel irritated by someone's behavior — there is a question worth stopping to ask. "How much of this person's situation can I actually see right now?" What stands out is always the acting person; the situation behind them melts into the background and is hard to see. The single extra effort of trying to see what is hard to see is the only thing that keeps you from becoming the one who sets this device running.

The moment you start hunting for a culprit, you fall in

Attribution to attitude is hard to rebut because attitude lives in the inner state, and its truth can't be settled by facts. This impossibility of settlement produces two consequences: the criticized side can't rebut, and the criticizing side never gets rebutted. This asymmetry in the distribution of responsibility is what makes attitude words a powerful tool inside organizations.

But there is one thing I want to confirm one last time. To understand this structure is not to find the "bad someone" who uses it. Before the FAE is a villain's tool, it is a cognitive quirk everyone carries. When it ties in with the organization's language, takes on fuel from emotion, and rides the structure of authority and incentives, it becomes institutionalized as a device of evasion — seeing that connection is the goal.

The moment we start hunting for a culprit, we fall into the very quirk we were criticizing. At the point where you attribute the cause to character — "that boss is the bad one" — you have set one more FAE running, ignoring the situation to blame the individual. So I keep the axis, to the end, on structure rather than character. Not who is at fault, but why this error happens so naturally, and so forcefully. Continuing to look at that structure is, perhaps, the only way to keep from becoming a part of it yourself.

Main sources