But England made it clear that the department was strongly committed to the principle of pay-for-performance, which was a key impetus for the reform effort - and not just on the grounds of fairness. Indeed, England painted a picture in which improved management of Defense operations would flow naturally from the closer link between compensation and job performance.

"If you're going to have pay for performance, then you have to have measurable objectives," England told employees. "So, ahead of time, I'll tell you what's great about this system: It's going to force managers to manage better. It will force managers and supervisors ahead of time to sit down with you at the beginning of the year and say, 'This is what we want to accomplish. Here are the objectives.' They'll make sure that they can be achieved between the two parties."

Forgive me a bit of skepticism about this view. For while it's certainly true that discussions between managers and workers about goals and objectives will be crucial to the success of the new National Security Personnel System, the mere creation of the system does not by itself ensure that managers will take the time and effort to set reasonable targets, communicate them to employees and evaluate workers based on how they perform relative to the criteria.

In fact, there's only one way that will happen -- if Defense leaders embark on an aggressive strategy to train civilian managers in a very different way of doing business and hold their feet to the fire in follow- ing through.

England told Defense workers that in his private-sector experience, "I've never been in a system [where I] wasn't paid for performance. I mean, my whole career has been that way. [You] sat down, you actually had to find objectives. They were written out with dates they were expected to be accomplished. You got graded - and not at the end of the year, but throughout the year. You know, the boss would say, 'Gordon, that was great, but you didn't do this very well, you ought to be doing this, and there's something that didn't work out.' And I'd decide, well, I don't have that skill, so I would go and take a class."

After 20 years in the private sector, I'm willing to bet that England's experience is the exception, not the rule. Many companies lack formal personnel systems with clear standards and goals, and even those that have them often fail to establish clear links between pay and performance, because top executives don't take steps to ensure that managers set appropriate standards and judge employees fairly.

But while we're pointing the finger of blame at executives and managers, let's take a moment to cut them just a little bit of slack. Setting standards for employees is far from easy, especially in this day and age. In both the private sector and in federal agencies, much of what workers do now is difficult to quantify and measure.

And under the new Defense approach, that problem will be compounded by the fact that civilian managers are accustomed to having personnel offices take on the complex task of setting objectives for specific jobs on their behalf. Now the managers will have to take on this role, in addition to doing the full-time jobs they already have.

That grumbling sound you're already beginning to hear is managers realizing just how much more will be demanded of them under the new approach.

Deep down, these managers know that it is in their own interest to communicate better with employees about expectations. But we live and work in the real world. And let's face it, some supervisors simply won't do the right thing -- unless they have to.

England may well be right when he says that with a pay-for-performance system across the Defense civilian workforce, "the whole department will operate better, because people will sit down and actually talk about jobs that need to be accomplished, and what's the schedule to do that, and what's the need." But the system itself won't make this happen, regardless of how well it is structured. Only people can do that.