(pure) managers don't attempt to solve technical problems because such action usually obstructs individual contributors. What's apparent here is the substantial difference in perspective between managers and ICs. Managers tend to see social equilibria and the network effects of tech problems. ICs can see much, much more deeply into specific (and usually key) aspects of tech problems. A lot of classic conflicts arise when ICs-turned-managers overestimate the depth and completeness of their comprehension and try to lead the technical effort rather than compliment it. Hence a lot of managers are 'uninterested in problems.'

> They think that if someone doesn't have a manager telling them what to do then they won't do anything

As someone who has moved from being an IC to being a manager, I can assure you that's not true. What we do think is that an engineer won't necessarily work on the right things. Engineers have a tendency to work on something until they've solved it, which isn't the same things as finishing it. This is why personal projects usually end up being unfinished and unpolished. Engineers also sometimes let the perfect become the enemy of the good. The oft-cited Malcolm in the Middle episode (http://www.youtube.com/watch?v=RHpJFROEOmg) illustrates this well.

As a manager, I don't see it as my job to keep my team busy or even productive...they do that on their own. I see it as my job to keep them focused, particularly on the tasks that will provide the most benefit for the business.

I don't think you are talking about the introspective capabilities of a whole class of employees (managers) but rather describe the percieved dynamic of an employee-manager relationship in a particular company culture.

Lots of industrial management theory offer this strategy oriented top-down command and control structure as the ideal dynamic for a BigOrg but my personal view is that it is the one that is most easy to describe academically and is probably best (best application for model, not the best model for...) for doing scalable things in a technologically fairly static setting.

If the company is structured around this top-down command tree then I imagine it's not as much necessarily about how the manager internally models peoples motivations but rather about the management culture of his colleagues and superiors within which he works.

This is the problem: manager types

There is no such thing, it is an absurd invention.

There is management - when that occurs, and only after, do you have a manager.