In the Shadow of Mt. Hollywood

The Theory of the Lead Narcissist

In trying to make some overall sense of what happened in the Bob Willis story, I re-read C.S.Lewis's essay "The Inner Ring", which I'm very happy to see is on line. Lewis begins by quoting a page from Tolstoy's War and Peace, in which a young lieutenant, Boris, sees to his initial puzzlement an old general who is acting deferential to a captain, putting up with the captain's discourtesy in interrupting him to talk to Boris. Lewis observes:

[T]he young second lieutenant Boris Dubretskoi discovers that there exist in the army two different systems or hierarchies. The one is printed in some little red book and anyone can easily read it up. It also remains constant. A general is always superior to a colonel and a colonel to a captain. The other is not printed anywhere. Nor is it even a formally organized secret society with officers and rules which you would be told after you had been admitted. You are never formally and explicitly admitted by anyone. You discover gradually, in almost indefinable ways, that it exists and that you are outside it; and then later, perhaps, that you are inside it.

This is one feature of the Willis story that puzzled me more as I wrote it down than it did when I lived through it -- two fairly high-level managers (though, in Cadovra's highly stratified system, neither a Vice President), Fred Feebles and Vijay Patel, wound up spending considerable time and effort catering to, making allowances, and covering for Bob Willis, whom they outranked. Throughout Willis's career, the organization adjusted to his various personal quirks and irresponsibilities, if possible simply by telling complainers to suck it in, but if not, by shuffling him around without visible penalty. I witnessed and heard directly about several such episodes, but there had been a considerable history of similar happenings in the past.

I had suspected, in fact (probably with some justification), that he had never had his disastrous interpersonal skills discussed on an annual appraisal, and likely had never been held accountable for the various lacunae in his professional stewardship that resulted in his transfers around the organization. I think the dual-hierarchy insight from Tolstoy and C.S.Lewis goes some way to explaining why this happened.

"Inner rings", according to Lewis, exist for two reasons; the one I'm most interested in here is to exempt the initiates from ordinary rigors of their disciplines or professions. Willis was tolerated, it would seem, until his self-exemptions threatened to become something his higher-ups could no longer brush off on their own -- for example, abusiveness toward female subordinates that could potentially bring in Human Resources, the company lawyers, and even outside authority. The situation, I think, is roughly as Lewis describes:

Obviously bad men, obviously threatening or bribing, will almost certainly not appear. Over a drink or a cup of coffee, disguised as a triviality and sandwiched between two jokes, from the lips of a man, or woman, whom you have recently been getting to know rather better and whom you hope to know better still-just at the moment when you are most anxious not to appear crude, or naif, or a prig-the hint will come. It will be the hint of something which is not quite in accordance with the technical rules of fair play: something which the public, the ignorant, romantic public, would never understand: something which even the outsiders in your own profession are apt to make a fuss about: but something, says your new friend, which "we"-and at the word "we" you try not to blush for mere pleasure-something "we always do."

The actual factors that make someone eligible for high status in the "dual hierarchy" are inscrutable. Whatever they are, Willis had them. I think almost everyone who knew him would (at least in unguarded moments) describe him as unbalanced and corrupt -- I suspect there's a relationship.

Willis served a valuable function in Cadovra's "second hierarchy". He basically showed how far you could go if you were in it, and by implication he showed the penalties for not being in it. If you were in it, any of the silly business on the appraisal form about leadership skills and the like was so much twaddle. If you weren't in it, of course, they could make those words mean anything they wanted to be sure you got a lower rating. To try to make someone like Willis answer to the same requirements as any old programmer, as Fred Feebles half-heartedly did, was to invite disaster. To be sure, Feebles and Patel were fully aware of their corner offices and all their other perks and position on the org chart, but they tacitly accepted some implicit standing Willis had in the "real" order of things -- something likely no one in the organization would ever verbalize. (I can imagine what some lady in Human Resources would do if I started explaining to her about dual hierarchies, too.)

I think this is also an answer to David Foster's comment that in a corporation, subordinates would complain if a manager didn't fill a hiring authority, and superiors would expect it to be filled. In addition to the fact that some creative procrastination -- excuses for which are always legion -- can allow the open authority to be overtaken by some other event like a hiring freeze, the fact is that the "second hierarchy" -- the real chain of command run by the entrenched narcissists and cube weasels -- is exempt from the complaints of subordinates or the normal expectation of the "red book" system. Willis treated his subordinates like crash dummies and suffered no real penalty -- his superiors covered for him. And the whole point of the "second hierarchy", as Lewis sees it, is to exempt its members from ordinary professional expectations. Willis's superiors covered for him there, too, and insofar as Feebles didn't do it, he suffered.

So we're back to where I started: the narcissist-survivor is, I think, often more in charge than you would think looking at the org chart. David Foster is right that, if you look at the org chart, some low-level supervisor who won't fill a hiring authority ought to be in deep trouble. But in the real world, that's not necessarily how it goes. Lewis repeatedly makes the point that the "inner ring" that exempts itself from the ordinary requirements of the organization is a feature of all groups much beyond the size of a string quartet, so I don't believe it's an answer simply to call Cadovra an "outlier" or a unique case. Willis served an important function in the "unofficial" hierarchy as a "lead narcissist", setting very important expectations for self-aggrandizing behavior that other members of management had to observe. I strongly suspect this kind of situation is as common as dirt.

Lead Narcissist -- Aftermath and Reflections

Following his transfer and demotion, Willis nevertheless made it all the way through to retirement. Demoted on the organization chart, he retained the salary he'd drawn as a manager, though by Cadovra policy, it was "redlined", meaning that he couldn't receive salary increases as long as he was doing the work of a lower pay grade. Nevertheless, since he wasn't able to handle such work, he came out extremely well -- able to finish out a 30 or 35 year career in a well-paid sinecure, then getting a pension payment based on his manager's salary. (The pension, of course, would be shared with two ex-wives.) Willis, it seems to me, is an archetypical narcissist survivor, whose abilities consisted entirely of keeping his job. He had neither technical nor people skills otherwise, and in fact his effect on the organization was destructive in nearly all respects.

The data base programmer who was exchanged for Willis and put into Willis's manager job, while definitely a narcissist (and likely a subject for future discussion), didn't come out as well. He and I were laid off nearly the same day, in one of Cadovra's rounds of downsizing. A vendor I frequently worked with told me, when she'd learned of my layoff, "I'm sorry to hear that, and that's from my viewpoint. We'd all rather work with customers who know what they're doing, and I don't think they're going to replace you with someone who knows what they're doing."

In fact, I learned gradually over the years that the opinion within Cadovra was, with little dissent, that what had happened to Willis was an injustice, and I was the one responsible. The truth was, of course, that my efforts to secure redress against him were completely ineffectual, and the actual reason for his transfer and demotion was the women in his area who'd complained. When Feebles left Cadovra, I asked him if he'd be a reference should I need one in the future, and he agreed -- but once I needed one, he reneged. Apparently on reflection, he'd decided I was also the problem as far as Willis went, and I suspect he blames me for the outcome in his case as well, though since he had a good job to return to, it's hard to see much injustice there, either.

This brings me to the original reason for this series of posts, which was David Foster's comment on my earlier theory about narcissistic survivors and their impact on corporate hiring (I will certainly welcome David's reactions). First, it seems to me that from Cadovra's perspective, Feebles's original error was in filling the hiring authority he'd been given -- or at least in trying to assure himself that it would be filled with someone who was worth the hire. Amid all the subsequent politics, the original reason for the hiring authority -- to bring in someone to handle the password backlog (and other undone tasks) was simply lost, and we may assume it was not important. In terms of the organizational turmoil that resulted, it was far more important that existing lines of authority be maintained, and the status of incumbent managers be validated.

Feebles's error was simple: his job was to keep the lid on, and he couldn't do it. One of David Foster's key comments was that if a hiring authority is made, the higher-level managers expect the authority to be filled. Actually, my experience of corporate life is that such things as hiring authorities are fairly easily revoked in hiring freezes, traded around, or otherwise vitiated by budget cuts or other adjustments. I would guess that it's not a real problem for an organizational adept to procrastinate until the matter is overtaken by events. Feebles, it seems to me, had two choices. One would have been to recognize that Willis was the actual problem, and deal with Willis before filling any hiring authority, since it shouldn't have been too difficult to predict how things would play out with Willis still in the catbird seat The other course would have been to procrastinate on the hiring authority, as seems frequently to happen in the real world.

I still haven't outlined my Theory of the Lead Narcissist, so I will continue this tomorrow.

The Real St. Patrick

Over the years I've gotten very tired of the usual green-beer-and-parades routine for St. Patrick's day -- possibly because it's taken me a long time to recognize and appreciate my own Irish heritage, which, like my wife's, is non-stereotypical. Both of us have Irish forebears, hers somewhat more identifiable than mine. The best I can tell from a family history written by my uncle is that a great-grandfather on my father's side, a William John Bruce (I am John William Bruce), who was born in Scotland in 1860, apparently moved (according to another family tradition) to Ulster, where he was a mason or stonecutter of some sort building railroad bridges, but by 1889 he was in Philadelphia, where my grandfather was born.

He apparently was neither Catholic nor Protestant, but Anglican, as were my wife's forebears. There is a later, perhaps apocryphal, story of William John Bruce working on St. Patrick's cathedral in New York, where the bishop paid him in whiskey. On not getting his preferred payment one day, he hammered the nose off the image of the bishop that was to be placed on the cathedral wall. William John Bruce died on the operating table in 1910, suffering from cancer of the tongue, probably brought on by heavy smoking. I can't envision him drinking green beer or marching in a parade wearing a funny hat. He sinned, I suspect, much more boldly, and I look forward to meeting him. You may argue that William John Bruce doesn't sound very Irish at all, but I will reply that neither was St. Patrick.

The real St. Patrick, once we try to peel away the layers of saccharine-sentimental Irish-American nostalgia and equally goopy hagiography, comes through as remarkably human (as many saints do when you stop listening to what well-meaning priests try to tell you). Patrick lived in the fifth century, still the late Roman period, and was a near-contemporary of St. Augustine of Hippo, another very human saint. One of the best discussions of Patrick's life, in fact, is in Stanley Coren's book The Intelligence of Dogs: "Consider Patrick MacAlpern, later Saint Patrick, whose life was strangely entwined with dogs."

Around A.D. 400, at age sixteen, Patrick [who had grown up in Wales or Cornwall] was abducted by Irish marauders. He was enslaved and kept as a shepherd for six years, his sole companion being a dog. In response to a dream, he made his way some two hundred miles to the coast, where he found the ship that the dream foretold would return him to his own land.

The ship was from Gaul, and the master had put into Irish waters in order to get a cargo of hunting hounds, which were bringing fabulous prices on European markets. Not surprisingly, as a penniless runaway slave, Patrick was received rather unsympathetically when he tried to gain passage. However, just as he was leaving, he was suddenly called back. It seems that, to maximize his profit, the captain had opted for stealing, rather than purchasing, his cargo of dogs. Over one hundred great Irish wolfhounds now packed the holds and filled the deck of the ship. Taken from their masters and their familiar surroundings, the giant dogs were frantic and furious, ready to savage anyone who came near. Some of the sailors had noticed that during Patrick's brief visit to the ship, he had spoken with some of the dogs and seemed to have a calming effect on them. Therefore, in exchange for his services -- which would involve feeding, cleaning up after, and otherwise caring for the dogs -- Patrick received passage to the continent.

The ship was badly underprovisioned and reached a ruined and deserted section of Gaul with its stores exhausted and nothing left to feed dogs or men. Because the dogs were worth more than the ship, the crew took the animals, abandoned the ship, and set off on foot, heading inland. Finding no inhabitants or food in the area, the dogs and men were soon all in jeopardy of dying of starvation. The shipmaster, who had learned that Patrick was a Christian, turned to him and in a taunting manner said, "If your god is so great, then pray to him to send us food." Patrick did so, and, the story goes, a miracle occurred. A herd of wild pigs appeared, seemingly from nowhere. Instead of bolting and running, as one might have expected, the swine stayed within reach long enough for the starving men, with the assistance of the dogs, to kill a number of them, providing meat for all.

When he was about 30, Patrick returned to England, studied for the priesthood, and entered a monastery in Brittany. It wasn’t until then that he realized his major goal in life, to return to Ireland and covert the pagans there. And it wasn’t until he was over 40, after struggling with his superiors’ worries that he didn’t have the education, that he was appointed Second Bishop of Ireland and began his 30-year mission. Bits and pieces of Patrick's attitudes come through in his autobiographical Latin fragments -- I'll bet his struggle for promotion wasn't an easy one, and I'll bet he knew some characters in the Church that would seem much like the compulsive climbers and cube weasels we know now in the business world (and likely still in the Church as well).

Coren continues, St. Patrick's association with dogs did not end in Gaul. Many years later, after a number of adventure, he returned to Ireland. This time it was of his own free will, and his goal was to preach Christianity. It is on his return that his rapport with dogs came to the fore again. It seems the news that a strange ship had just landed, from which had emerged white-robed men with clean shaved heads who chanted in a strange tongue, prompted an Irish prince named Dichu to go to the coast to investigate the situation. He was accompanied by his favorite large hunting hound. Observing St. Patrick's missionary group, Dichu decided that the best course was to kill these odd clerics and be done with it. With a wave and a shout, he set his dog at Patrick. The dog leapt forward in full fury, but when Patrick uttered a short, one-sentence prayer, the dog halted, grew quiet, and then approached Patrick and nuzzled his hand.

I've come up empty trying to Google another aspect of Patrick's life, the so-called "St. Patrick's measure", which came about, at least according to one story, when Patrick visited a tavern and found the barmaid pouring short measures of whiskey. He remonstrated with the barmaid, who became repentant on the spot, and thereafter poured extra measures of whiskey. My regret in telling this story is that my own capacity for whiskey is steadily diminishing.

So my St. Patrick is maybe a little like my own not quite Irish ancestor, someone who's likely sinned more than a little, had both struggles and disappointments, someone who, in the end, nevertheless comes off as pretty tough, someone you'll likely not encounter singing sappy songs in the beery sentimentality of the stereotypical Irish holiday. But by all means, if you're in fact into green beer and sappy songs, enjoy the day as you please.

Lead Narcissist -- VI

One day I was checking Vijay's appointment calendar on line -- I'd found that browsing the calendars of the high-level managers was a good way to find out what was happening. His secretary had made another mistake, and she hadn't made a series of meetings confidential. A whole parade of women who worked for Bob Willis was coming in to talk to Vijay about him, one at a time. I hadn't heard anything about this.

Over the next several days, I began to learn what had happened: I wasn't the only one having conversations with Willis suddenly veer into arguments with his wife. It turned out that all the women who worked for him had experienced episodes where he'd start arguing with them about some divorce-related issue and become abusive, as if the women were temporarily his estranged wife. It was happening enough that one or two had gone to Vijay about it, and then he called the rest in.

In fact, I'd gone to Vijay myself for the same reasons I went to Fred Feebles -- including the strange outbursts -- but all that happened was I'd learned it was useless to complain to Vijay. I didn't know him very well, but I came out of that meeting realizing he was very top-down. My job was to do what my manager told me. If there were problems with Willis, it was because I wasn't doing what he told me to do. Period. The memo about the travel authorization he'd told me to fill out so he could complain about it to Vijay was probably just part of the groundwork Willis had already laid.

I suspect the difference between me going to Vijay and all the women in Willis's area going to him was that any of the women might turn it into a sexual harassment case. At that point, it would simply be out of Vijay's hands, and Vijay himself would be in hot water, so he had to do something. Vijay got with Feebles and arranged another trade. One of Feebles's data base programmers who wasn't working out would take Willis's job, while Willis would become a data base programmer, where I already knew he wouldn't work out. Everything would be even. It was all under Vijay's budget, so nobody else would gain or lose from the trade, which made it even better.

Feebles had, of course, apparently taken a mildly rigorous tack with Willis when I'd brought the problem to him -- Willis had suddenly ended his little ploy of not quite being the manager but still sorta-kinda giving orders, and he'd had to back off the data base boondoggle, at least as long as he still worked for Feebles. The outcome with Willis, it seemed to me, should have vindicated both Feebles and me. The guy shouldn't have been a manager.

In fact, Feebles was on his way out. Unlike most of the other people at Cadovra (but like me), Feebles wasn't a lifer. He'd only been there six or seven years, and he'd come from an aerospace company where he'd taken early retirement. He pulled some strings back at the aerospace company and got another job back there, under some pressure, he mentioned to me at one point, to do so. For a long time, I thought there may have been some justice in Cadovra suggesting Feebles find himself a new placement -- assuming, that is, that the high-level IS managers at Cadovra felt that Feebles hadn't dealt quickly or severely enough with Willis, and the eruption of Willis's personality problems was something Feebles could have forestalled by dealing more proactively with him.

I was naive. I'll conclude this series Thursday with some of the aftermath and my reflections, but I have a special post for St. Patrick's day coming up for tomorrow.

Lead Narcissist -- V

I had no illusions about what Willis was doing -- he was regularly going to Fred Feebles and Feebles's boss, Vijay Patel, and telling them I was preparing some major disaster. Meanwhile, he was making my life miserable with his half-baked data base project and his bizarre behavior. He'd still come around to my cube with strange details of his personal life -- one day he confided that he beat the family cats when they got up on the furniture, but since they would now bite and scratch him, he was afraid of them. Another day he came around and wanted to know the meaning of the word "etagere".

"Oh," he said. "A knick-knack shelf. Now I know what my wife's lawyer meant. They're working on how to divide the property." Cadovra, by the way, issued everyone a dictionary. He may have been too lazy to look it up at first, but he likely went back to his office to check afterward, and was probably angry when he found out I knew what the word meant. I'd also begun to notice that, in conversations, he'd suddenly, without warning, change the subject to some long-running fight with his wife that he'd keep playing through in his mind. He'd make some angry remark completely unrelated to anything we'd been talking about, and I'd have to navigate my way through the situation.

He was my friend, except when he wasn't. He was my boss, except when he wasn't. The puzzling thing was that, as far as I could tell, he wasn't what they would call in Human Resources a "good performer", but his whims were being catered to by both Feebles and Patel. And it wasn't as though nobody could be fired from Cadovra. A couple of people I knew, liked, and respected were in fact fired during this period. Once I overheard a couple of contractors talking about it. It was very sotto voce in the hall, and I suspect they knew I overheard it, but they probably trusted me.

"What I don't understand," said the one to the other, apparently in the midst of a conversation about the most recent case, "is the ones they get rid of are the most capable and knowledgeable. It happens over and over." I've worked as a contractor myself, and I understand the contractor's perspective -- a contractor normally has to rely on technical ability and people skills to get the job done (and therefore get paid), and sees acutely the absence of those skills among regular employees.

I hesitated to go to Feebles over Bob Willis, since technically I'd be going around Willis, but it was plain I finally had to have a talk about what was going on. The constant suggestions I was doing it all wrong, the incoherent data base project, and the increasingly odd behavior were reaching the point where I had to elevate the problem.

"I don't understand this reporting relationship," I told him. "I've never seen anything like it before. Basically, he isn't responsible for what happens, but he gets to back seat drive."

Feebles didn't have much to say. His position was mostly "I'll tell Bob to sit down with you, and you can work it out with him." That, of course, wound up being nothing more than a continuation of his insistence that we had to eliminate all possible risk from issuing passwords.

So I went back to Feebles again and said nothing had changed. The result was that Bob Willis magically decided that he'd finished all the important work he'd needed to do, and he could now return to his previous position as manager and be completely in charge. At the same time, he backed down from insisting we do his data base boondoggle.

But then Vijay Patel reorganized the area and took Feebles out from in between Willis and himself. This cut Feebles's area of management in half, a bad sign for him. In hindsight, Patel probably didn't like it that Feebles apparently told Willis to fish or cut bait over being manager. Willis likely had some kind of a side channel to Patel in any case.

I probably never knew more than a small fraction of what Willis was telling Patel. I found out about one instance due to an error by Patel's secretary. Patel, by the way, used his secretary the old-fashioned way: he had a PC on his desk, but he never turned it on. Instead, his secretary would print out his e-mail and bring it in to him on paper. Patel would read the messages on paper, and when he was ready, he called his secretary back in and dictated his replies, which she dutifully typed into Patel's e-mail account and sent.

One day I got a call from a guy in a Federal regulatory agency that supervised our utility. I was neither in a department nor at a level where I normally got calls like that, but the guy, either going beyond his own pay grade or due to some mistake, called me and told me to come to a meeting in San Francisco. On one hand, this was something that somebody else in the organization needed to know about and deal with, but it also involved travel, which I couldn't approve for myself anyhow. Since Willis was my boss, the correct thing to do was to tell him about it and give it to him to handle, which I did. "Well," he said, "for now, just fill out a travel authorization request and send it up to Vijay to see if he wants to approve it. I'll talk to him about it."

Well and good. I filled out the request and sent it up. A day or so later, I got by return office mail a copy of a memo Willis had written to Vijay, explaining how I was angling for an unnecessary trip to San Francisco, another example of my bad attitude. Vijay had written a comment on the bottom that he completely agreed with Willis, and my travel request was completely out of line. The travel form was attached, with DISAPPROVED written in big letters in Vijay's handwriting. Apparently Vijay's secretary had addressed the envelope to me instead of Willis by mistake.

And since Vijay had moved Feebles away from our area, Willis decided this meant any agreement he'd had with Feebles to back off on the data base project was void. He started pestering me about it again, this time as my full supervisor, not as some back seat driver. As his divorce proceedings continued, he seemed more and more often to leave work-based conversations behind and suddenly change the subject, arguing with me on divorce-related issues as though I were his wife's attorney.

Lead Narcissist -- IV

What I'd been doing in solving the password problem, of course, was making Willis look bad. He'd been shunted into his position of managing the security wallahs from a more responsible position, systems programming manager. Nobody had quite put a finger on what he was or wasn't doing right, but eventually someone decided he wasn't working out there, and they moved him aside. They worked out some kind of a deal with Fred Feebles, who was willing to trade his current password manager for Willis.

But then Feebles discovered that Willis wasn't working out on passwords, either, any more than he'd worked out on systems programming. He'd turned a situation that was middling-OK into a months-long backlog, and the supervisors who couldn't get passwords for their new hires were starting to complain. At that point, Feebles put in to hire someone else to "help" Willis with the passwords, and that turned out to be me. All I'd done was the job I'd been hired to do, but the world is a complicated place. Feebles, of course, was happy. Willis wasn't.

Willis began to complain to Feebles that he had too much work. Exactly what that work was, I could never quite figure out -- I was doing most of what Willis was supposed to have been doing before I came. But he convinced Feebles that he had projects so urgent that he had to take a temporary break from managing the passwords. Not that his title would change, of course, or even that his position would change on the org chart. But Willis worked it so I would sorta-kinda be in charge of the area, but he'd still be my supervisor. What it boiled down to was if anything went wrong, it was my fault, and that would prove Willis had been doing things right all along. Not only did Feebles approve this arrangement, but his higher-ups did, too.

At this point, Willis got The Shadow. He began to worry. "We've got the password backlog solved," he said. "But they're doing everything so fast, what if they make a mistake? How do we know they're doing things right? What if they lose a request?" For starters, of course, the problem had been solved to the point where dozens of unsatisfied customers -- the supervisors who hadn't been able to get passwords for their new hires -- were now satisfied customers. In an imperfect world, an occasional error might lead to another occasional unsatisfied customer, but not the dozens they'd had earlier.

That, of course, wasn't enough of an answer, and I knew it. I began to suggest other approaches. The password requests were all logged. We already double-checked to be sure all the requests were matched by a new password. We sent out evaluation forms to a percentage of the supervisors to see how well we'd done. I suggested we follow up and do a full QA check on a percentage of the passwords issued. All this was what an MBA would likely recommend. Willis was still in a tizzy. "We can't have any errors," he said. "What if the system breaks down, so we not only lose a request, but we forget to log it? Then we wouldn't know we'd lost it." I answered that at worst, if that happened, the supervisor would call to follow up a day or two later. If it happened once in a blue moon, it was still far better than how things had been a few months earlier, when nobody was getting their passwords.

"That's unacceptable," said Willis. He began to propose a solution where the security-wallahs would make an entry in a PC data base package for every password request they got. He'd already started designing it. He put in at least a dozen fields, all the information you could think of, name, title, supervisor name, job function, department, employee ID, date requested, date filled, so on and so forth. The problem was, of course, that it would take much longer just to enter the information in the data base than it did to issue the password -- and we already had the original password request on file with all that information, along with the existing log, in any case. This was adding a lot more work just to get information we already had and used.

We'd be back to square one, taking twice as long to issue a password and building up a backlog again. This, I suspected, was what Willis had in mind -- if he hadn't planned it out that way, it was certainly going to be the practical effect. Willis and I went back and forth on this for weeks. I could see his knuckles whitening in fear as he seized the arms of his chair when I'd make my suggestions. There was no way I was going to win this with common sense.

And the data base project then became the important task he could fill his time with, since I clearly wasn't doing my job properly. Like his doodling over whether he needed different kinds of tires on the left and right sides of his car, it was a long, complicated boondoggle that stubbornly resisted light-of-day solutions.

Data base programming, in fact, was beyond Bob. He would get very confused while working on the project. He'd call one of my co-workers in hoping to get help, but my co-worker told me that what would happen would be that Bob would start to ask a question -- get out the first few words -- and then freeze. He wouldn't say anything more, but he'd sit staring into space, and my co-worker wouldn't know whether to stay or leave. I began to think Bob actually got himself high when he became that confused, since he did it so often.

Lead Narcissist -- III

Cadovra has always been interested in its own corporate culture -- in fact, as much as a corporation can be so, I think you can characterize it as self-absorbed. For instance, it hires sociologists to study itself. And the company is self-aware enough to identify problem areas in its corporate culture -- though like many people who visit therapists, they're aware of the problems, in fact talk them to death, but aren't motivated enough to do anything about them. Every once in a while, somebody from Human Resources refers to a problem area. For instance, she might say, "We need to identify our top performers. However, our current system of appraisals doesn't tell us who our top performers are."

Well and good, one might say -- an insightful analysis. But do we do anything to change our current system of appraisals? Of course not. We might hire some consultants, form a committee, and a couple of years down the line decide to change the appraisal forms so people will get a "Needs Improvement" instead of a "Meets Some Expectations" -- but bottom line, the same bad people seem to get good appraisals, and the same good people are sent out the door. Willis, who was shunted laterally to get rid of him throughout his career, probably never got a bad appraisal.

Or the Human Resources types might say, "We know we suffer from a risk-averse corporate culture. We need to find ways to let our managers tolerate more risk in their decision making." A worthwhile, progressive-sounding statement. But let's look more closely at what they mean by a risk-averse culture. "Risk-averse" carries with it a certain positive connotation of prudence, like the person who might choose to drive a Volvo instead of a BMW or invest in bonds rather than stocks, a matter of reasonable personal preference. A risk-averse person might be a fuddy-duddy, but not a wacko.

But in practice at Cadovra, to be "risk averse" is to have a hysterical fear of making tiny mistakes. This fear is so dominant and unreasonable that I can only compare it to obsessive-compulsive disorder -- for example, the person who leaves the house and can never be completely certain that he's locked the door, so he feels the need to keep going back to double-check. Somehow most people find a way to accept and manage the risk that they've forgotten to lock the door and get on with their lives. They might develop a going-out routine that reliably incorporates a twist on the knob as they shut the door, or some other way they can assure themselves with nearly complete certainty that they've left things safe.

Then there's the percentage who can't handle that small risk, and they feel the need to return to check the door once, then check it again in case they forgot to double-check it the first time, then check it a third time in case the first two times were just something they remembered doing yesterday, and so forth. Naturally, in the worst cases, people with this problem will never get where they want to go, because they'll never be able to feel comfortable enough that they've locked the door. Let's call this feeling that you're never quite sure if you' ve done something The Shadow. This Shadow creeps into Cadovra activities at unpredictable times.

For instance, I was hired at Cadovra, among other things, to find a way to eliminate the backlog of password requests. If it was taking months for people to get their passwords, this was a major problem, and it had to be fixed. So a little bit at a time I had the security wallahs start logging the requests. Then, Mary Poppins style, I started pointing out how many requests you could actually do in a short time, and raised the level of awareness and expectation. Pretty soon the backlog went away, and people were getting their passwords in no time at all.

Somewhere toward the end of this process, Bob Willis got The Shadow.

Lead Narcissist -- II

WIllis had, in fact, been bounced around Cadovra for two decades before I showed up there. He started out in Marketing. It's worth pointing out that Cadovra is a utility, and marketing jobs in a utility are notoriously cushy, since utilities are monopolies. There's no need to seek out customers -- customers come to you, because you're the only place they can get electricity or natural gas or phone service. Somehow, Willis screwed up in Marketing, and Marketing was able to fob him off into Information Systems, which was the bottom of the barrel. If you screwed up anyplace else, you got sent to IS.

On the other hand, it's not as though anyone in Marketing had duties important enough to make errors or omissions matters of any consequence. All they typically did was take important customers to lunch, but since those customers had no choice but to deal with Cadovra, the lunches were pro forma. My conjecture is that Marketing simply found Willis creepy, as nearly everyone else did.

During one of his divorces, he went to a party at the home of a single female programmer. He was just another guest, with no reasonable expectation of making a score with her or anyone else. Nevertheless, he brought a small gym bag with him to the party that apparently contained a change of underwear and a bottle of mouthwash. At some point during the party, he left the gym bag in the woman's bathroom, and he put the bottle of mouthwash prominently on the counter by the sink. As the other guests said their good-byes, the woman conspicuously returned the gym bag and bottle of mouthwash to Willis and ushered him out the door.

One of the smaller problems for me in working for him, in fact, was his assumption that, since I worked for him, I was his friend. Bosses who want their subordinates to be drinking buddies or golfing companions do it because they aren't going to get drinking buddies or golfing companions any other way but, in effect, to hire guys to do it. This was the case with Willis. The worst part of going to conferences and similar travel was that I'd frequently have to go with him, and sometimes my wife and I would have to socialize with him and his wife. One year he had the office Christmas party at his home, naturally a compulsory event. The guy was so tactless that at one point during the party he asked, "Where's John?" and went and opened the door on me in the bathroom to be sure that's where I was.

Another time we drove together in his car to another company facility. He decided to stop on the way at a tire store, where he talked earnestly for an hour with a salesman. It seemed he'd carefully figured out the number of uphill left turns versus downhill right turns he made on his daily commute, or something like that, and they didn't balance out. So he was concerned that maybe he needed to have different tires on the two sides of his car, and he wanted the salesman's advice. Remembering this episode, it's occurred to me that if he drove home by the same route he went to work, all the uphill-downhill right and left turns would cancel themselves out. But this was actually typical of the way Willis would get bogged down in details and never see a simple solution. (I doubt if he actually bought two new tires, by the way; he was too cheap, and his second divorce wasn't far off.)