By the time four or five books have been published on a topic, you're tempted to think of it as yesterday's news, a wrap, case closed. But just as I was beginning to dismiss the subject of leadership in precisely this manner, Fortune magazine arrived on my desk.

Corporate America, Fortune informed me, is searching for leaders. Real leaders, not just managers. Men and women (mostly men, I bet) who have vision, who seek change, who can transform their organizations. Looking to instill these traits in their executives, big firms have sent them on all sorts of nutty training programs. In one such, teams from General Foods Corp. found themselves trying to build rafts and sail down a river, first under nit-picking managerial inspection and then on their own. Which rafts sank and which ones floated? No -- you tell me.

Maybe executives in big companies need all the help they can get (anyway, being picked for a sexy new leadership training program sure beats working). But what cooled the cockles of my heart about the new search for leadership was precisely what had turned me off about the aforementioned books. Both treat leadership as something like painting by the numbers. Learn the rules, and you too can be a corporate Rembrandt.

Take The Leadership Challenge, by James M. Kouzes and Barry Z. Posner (Jossey-Bass, 1987), one of the latest entries in the leadership library. Kouzes and Posner asked more than 550 "leaders" (they don't say who) to fill out questionnaires about "personal best" cases, or situations where the respondents thought they had accomplished something significant through leadership. A nice idea -- and potentially a source of some fascinating stories. But the authors keep the juicy specifics to themselves, offering us instead some bland commonalities. Sample: leaders "envision the future." Leaders "search for opportunities."

If the world were only so simple, we could all be leaders, even without building a raft. Trouble is, the qualities that seem so important in one context are irrelevant (or downright harmful) in another. And the most important trait -- good judgment -- can't be taught at all. Look again at a few of the qualities Kouzes and Posner say you need.

Leaders have a vision. Who could disagree? Show me an entrepreneur without a vision and I'll show you an entrepreneur who needs a new PR firm. Just don't try to put your vision in the bank. Nolan Bushnell, creator of Chuck E. Cheese and Pizza Time Theatres, was long on vision. Unfortunately he was short on customers.

Leaders "strengthen others: sharing power and information." Jan Carlzon, president and chief executive officer of Scandinavian Airlines System (SAS) and by all accounts one of the most imaginative executives in his industry, tells in his book Moments of Truth (Ballinger, 1987) how he "flattened" his organization's pyramid, giving "more responsibility to the frontline personnel." He even began calling first-line employees "managers" -- a reminder, he writes, "that their roles [had] undergone a fundamental change." Great stuff -- and exactly what Donald Burr, founder of People Express Airlines Inc., did with his company. One succeeded, the other didn't.

Leaders "experiment and take risks." Remember a few years back, when Amdahl Corp. founder Gene M. Amdahl left to form a new company called Trilogy Systems Corp.? Trilogy would engineer a great leap forward in computers based on developing an advanced new chip. Alas, the company has vanished, and Amdahl has embarked on yet another venture. Maybe people are acclaimed as leaders when the risks they take pay off.

Right there, of course, is one reason leadership can't be reduced to a set of skills or personality traits, to be acquired on weekend retreats or learned from reading books. Plenty of people have possessed all the attributes the books ascribe to leaders, but they were sorely lacking in judgment. Remember General Custer? Gary Hart? In business, an executive's judgments about the marketplace, people, and a myriad of other variables are more important than abstract "leadership" qualities in determining a company's success or failure. What undermined Burr wasn't his innovative approach to managing people, it was the fact that he tried to expand his company too quickly.

The real trouble for the theorists of leadership, though, is that yesterday's leaders may be today's goats -- even though they're still the same people with the same virtues. Steven Jobs, founder of Apple Computer Inc., created a new industry with his vision of inexpensive, easy-to-use computers; passionate and dedicated, he inspired his corporate troops to build something approaching a model company. Then, pffft. Suddenly Jobs's vision and dedication were liabilities, and new chief John Sculley was ushering him out the door. But Jobs was hardly the first world-class leader to blow a big one. Edwin H. Land built Polaroid Corp. on the basis of a stunning new invention, instant photography. Then he staked Polaroid's future on an equally stunning instant-movie system, and very nearly drove it down the tubes. Remember Polavision? Not many other people do either -- except Polaroid's shareholders. Writing off the ill-starred system cost the company $68.5 million.

Seen in this light, leadership is in large part the ability to make good judgments -- and to do so again and again. But it's also the ability not to let leadership traits themselves get in your way. Land and Jobs accomplished what they did by imagining new things and working hard to implement them, disregarding naysayers along the way. People Express grew at a breakneck pace because Donald Burr was willing to take big risks. In the end, it was precisely these traits that turned out to be tragic flaws. But who could have known? The difference between confidence and overconfidence is clear only in retrospect.

The unsettling truth is that leadership, like fine art, is too complex to be reduced to a set of rules. Does this mean that it can't be written about, studied, even taught? No -- only that it has to be learned and practiced in real situations, where real outcomes are factored into the equation. The difference between leaders and everybody else isn't what kind of people they are, it's what they accomplish -- and whether they can avoid blunders as big as their victories. Many CEOs can cite instances in which they and their organizations outdid themselves, achieving what Kouzes and Posner call personal bests. The question is how much time separated those occasions from the personal worsts.

Artists know their work is good if it sells, or if other knowledgeable people respect it. Business leadership -- not quite so rarefied a commodity as artistic talent, thank goodness -- is even easier to judge. If you want to know how you're doing as a leader, don't ask yourself whether you have 20-20 vision or whether you're taking enough risks. Instead, just ask yourself how your company is doing -- and how you're viewed by the people who work for and with you. Where leadership is concerned, a little time spent investigating those questions, in their many facets, is worth a lot of trips down the river on a raft.