"It keeps me in line"

This is the belief that holds the rest in place. You may dislike the commentary, but you think you need it. It's your quality control. You're a bit lazy and careless underneath, and only the threat of a roasting keeps you performing. If you turned it off, you'd stop trying, the forms would never go in, and the figures would all be wrong.

It's a reasonable theory, and it can be tested. If harsh self-criticism drives achievement, then people who do more of it ought to get more done.

Theodore Powers, Richard Koestner, David Zuroff and their colleagues have tested that repeatedly. In a set of five studies, they measured how self-critical people were and then followed their progress on goals they'd set themselves, across different areas of life, including academic work, weight loss and music. The pattern was the same in all five. The more self-critical people were, the less progress they made towards their goals.

The same studies contain a finding that I think is even more useful. The researchers also measured how high people's standards were and how much they strove for excellence. High standards and harsh self-criticism often go together, which is why people assume they're the same thing. When the researchers separated the two statistically, high standards on their own went with more progress, and self-criticism went with less. It was the standards that were helping, and the self-criticism was holding people back.

These are observational studies, and I should be careful. It might be that people who are failing at their goals become more self-critical, and not the other way round. The researchers measured self-criticism first and progress later, which helps, but it doesn't settle the matter. What the studies do show is that the discipline story's central prediction fails. If self-criticism were driving achievement, the most self-critical people would be the ones getting furthest. In study after study they were the ones falling behind.

Why might attack hold a person back? This is my interpretation, and I think most people will recognise it. When you know that any failure will be followed by a savaging, failure becomes something to avoid at almost any cost. The surest way to avoid it is not to attempt anything you might fail at. So you put off the difficult task, because starting it means risking the commentary. You don't apply for the job or ask the question, and you leave the draft unshared. When something does go wrong, you're tempted to hide it, from others and from yourself, because looking at it squarely means another savaging. People don't learn well under threat. They get by through defending themselves and concealing things.

Think about how this works with anyone else. Imagine a manager who responded to every slip with "You idiot, what's wrong with you?" Some of her staff would work hard, out of fear, for a while. They'd also hide their mistakes, avoid anything risky, stop making suggestions, and leave when they could. Nobody thinks that's good management. You've been managing yourself that way for thirty years, and you've given the method credit for everything you've achieved.

Look at that Wednesday again. The form was late and you found it. What got it sorted out was a ninety-second phone call. The commentary contributed nothing to that call. It delayed it by a minute or two while you sat in the car with your heart going, and it carried on afterwards when there was nothing left to fix. You're conscientious, and that's why the call got made. The commentary played no part in it.