On a misty morning in April, a group of MPs gathered in Ottawa to hear Charles Sirois speak. The members of the Commons foreign affairs committee were expecting a briefing from the telecommunications visionary on an initiative to extend the internet to developing nations. But Sirois, Canada's star private-sector representative on the Digital Opportunity Task Force, never showed up. He had, committee chair Jean Augustine announced, been "caught in the fog."

Indeed. Sirois has spent the last couple of years struggling to steer his telecom empire through dense weather. Virtually no component of his vast Telesystem Ltd. holdings (ranked 983 in this year's Top 1000) has escaped bankruptcy protection or debt restructuring, a process that has proved painful for creditors and shareholders alike. Not to mention career-ending for a few of his friends.

Back in 2000, when the telecom expansion seemed unstoppable, Sirois declared, "More than ever, we must encourage dreams and the people who realize their dreams. [We must] not banish failure. Stop getting out the bulldozer each time someone makes a mistake. I wish we'd develop a culture that allows for failure, thus, initiative."

Was he trying to warn us?

The Sirois mystique is rooted in the heroic—for Quebeckers—story of a backwoods boy who rose to wealth and fame with Gatsbyesque flair.

In 1979, fresh from completing a master's in finance from Laval University in Quebec City, the then-unilingual Sirois returned to his isolated hometown of Chicoutimi, and took over his father's tiny paging business. Sirois was cocky and intrepid, and, at 25, already had a young family. But he was no techie. Sirois's forte was finance. In paging, he saw a fragmented industry of tiny operators at the mercy of equipment makers, their only source of long-term credit.

" Their suppliers had them by the balls," Sirois recalled in a 1999 interview. (These days, he's not talking to the press.) "I didn't have any particular interest in running that kind of business, but I was fascinated by the structure of the industry. As I often say, I always study the lake, not the canoe, because if the level of the water goes up, you can have the ugliest canoe in the world and it will go up like all the others."

By persuading banks to lend him up to four times a paging company's cash flow, Sirois began snapping up competitors across Quebec and Ontario, gaining half the Canadian market within a few years. He was leveraged to the hilt; but he was hooked. "I was coming from the bush; no one knew what to make of me," Sirois recalled. "My debt-to-equity ratio was 60-to-1. I was stretching it to the limit. Boy, was it fun."

In 1984, Sirois created National Telesystem (now simply Telesystem) to hold his paging interests, and sold 30% of the company to the Caisse de dépôt et placement du Québec for $2 million. The provincial pension-fund manager was committed to cultivating a francophone business class: An entrepreneur like Sirois was an ideal candidate. The deal was the beginning of an enduring relationship, one that saw the caisse back almost every one of Sirois's ventures. But the caisse was not the only mentor to figure in Sirois's rise.

By the time he met Sirois in the mid-1980s, Jean Monty was already an executive vice-president at Montreal-based BCE, even though he was still shy of his 40th birthday. His even younger new protégé, Charles Sirois, became a proxy for Monty. Sirois could show off or take the sort of gutsy risks that were off-limits in the phone company. Besides, the two hit it off. "A very special man," Sirois said of Monty. The University of Chicago MBA might be suave where Sirois was rustic, but they both craved action. "Sometimes, [Monty] didn't decide your way, but he made a decision on the spot," Sirois said. "For me, that's very important." In Sirois, Monty saw a kindred spirit. In Monty, Sirois found a guardian angel.

With wireless in its nascent stages, Sirois knew enough to hitch his fortunes to BCE, which, in addition to owning a competing paging business, had also just been awarded one of the country's first cellular telephone licences. Sirois persuaded Monty, then the BCE executive in charge of the wireless assets, to combine their respective paging operations. In 1987, Sirois, then only 33, swapped his stake in the merged business for 11% of BCE Mobile, which brought together all of BCE's wireless holdings. Sirois assumed the CEO's mantle at the new unit in 1988—just as the cellphone phenomenon was taking off. His reputation bloomed—he was the guy with a sixth sense for identifying future tech trends.

The BCE Mobile deal was Sirois's coming out on Bay Street. Toronto finally took note of the jeune loup with the thick accent, steel-wool hair and gummy smile. How could it not, when he called analysts "imbeciles" for shortsightedly criticizing BCE Mobile's losses? When he flashed his Porsches and Jaguars, his ski-and-scuba lifestyle, his diamond-studded Fabergé watch? When he turned around and pooh-poohed the wealth? "You can't eat 20 filet mignons a day," Sirois told journalist Matthew Fraser. "If you do, you will be sick. I don't care about money. I want to build something."

Sirois is, by his own admission, an entrepreneur, not a manager. He left BCE Mobile (now Bell Mobility) in 1991. Soon afterward, he turned his attention to Teleglobe, then Canada's monopoly overseas telephone carrier. Privatized in 1987, it was performing well below its potential. Sirois, backed by fellow minority shareholders BCE and Rogers Communications Inc., launched a hostile takeover, waging one of Canada's first proxy fights to install a new board. By 1992, Sirois, with a 20% stake, was installed as Teleglobe's CEO. His first move was to fire every management employee—right down, he said proudly, to the secretaries. "I just wanted a clean slate," he said.

As Sirois saw it, Teleglobe's monopoly on overseas calls from Canada wasn't worth being shut out of other markets. He agreed to surrender the federally awarded domestic monopoly as of 1998—a U.S. government condition for gaining access to stateside traffic. "Bell Canada's market is limited at 30 million [Canadians]," Sirois said. "My limit is six billion people."

It seemed to make sense. Foreseeing that the internet would create limitless demand for voice and data transmission, Sirois embarked on a $5-billion (U.S.) plan to connect 160 cities around the world with a fibre-optic network. "I tried to explain that PCs are like cars. They were getting faster and faster, but they were running on sand roads," he said. "I saw all these Ferraris going around the world and no highway to cross the ocean. To me it was obvious."

Curious, then, to see what Sirois was predicting only a few years earlier. In 1995, he co-authored a thin volume on the Information Highway. "It is relatively easy to point to the absence of demand for many of the anticipated products and services that will be carried on the I-Way," Sirois and Claude Forget wrote in The Medium and the Muse. What's more, "the transformation of telecom carrying capacity into a mundane 'commodity' bought and sold on the basis of lower and lower prices is virtually certain to occur."

Prophetic words. But Sirois, like so many others, gave in to the giddiness of the late 1990s: Exponential growth in traffic would pay for Teleglobe's investment. It was the same logic that led him to add retail long- distance to Teleglobe's business in 1998, with the purchase of Dallas-based Excel Communications Inc. The deal raised eyebrows. Three-and-a-half-billion U.S. for a company that relied on part-time sales people to sign up friends who signed up friends? Investors bid up Teleglobe's stock anyway. "I don't need to spend a penny on advertising…I just need to have thousands and thousands of people who decide to sell my product around the globe," Sirois explained. "It's just like building an atomic bomb. You need to have enough plutonium to start the chain reaction and make the explosion."

As always, he did have a unique way with a metaphor. The Amway of telecom did in fact bomb with consumers, while its complicated commissions and billing systems were an administrative nightmare. Meanwhile, Teleglobe was faced with an increasing number of competitors, including Global Crossing and 360networks, all building their own overseas fibre-optic networks. Rate-slashing razed margins, fulfilling the "commodity" prophecy Sirois had made in 1995. By mid-1999, Teleglobe was forced to issue an unprecedented profit warning, slashing its forecasts.

A full-time CEO might have been able to work through the problem. But Sirois divided his time between Teleglobe and Telesystem, his private holding company, and its ever-expanding stable of investments. By the late 1990s, Telesystem had major stakes in, and operating control of, Microcell Telecommunications Inc., Canada's fourth-biggest wireless phone provider, via the Fido brand; Telesystem International Wireless (TIW) Inc., which was staking its claim to wireless phone and paging franchises in such far-flung markets as Asia, Brazil, Eastern Europe and the U.K.; and Look Communications Inc., which was taking on BCE and the cable giants with its wireless internet and digital TV services.

Sirois had also jumped on the convergence bandwagon, snapping up control of Coscient Inc., a Quebec TV and film production and distribution concern that Sirois renamed Motion International. Finally, there were the dozens of tiny start-ups in which Telesystem had invested. Providing ongoing funding for the lot had proved relatively easy, so long as the Sirois aura of omniscience seduced banks and investors.

That aura attracted big names, such as Brian Mulroney, to Telesystem's board of directors, and compelled two former Quebec deputy finance ministers, Alain Rhéaume and Claude Séguin, to join Sirois's executive suites. It allowed Sirois to share the bill with Bill Gates, Rupert Murdoch and Warren Buffett at CEO gabfests in California.

And above all, the aura inspired loyalty from Jean Monty, even while Sirois openly sought to steal market share from BCE subsidiaries. Look took on Bell ExpressVu, Microcell invaded Bell Mobility's market, and TIW competed against Bell Canada International outside Canada. "Jean talks to me about files where we are partners but not about ones where we are competitors," Sirois said. "Microcell even lodged a formal complaint before the CRTC against Bell Mobility the very evening I had dinner at Jean Monty's house. He found out about it the next day, just like everyone else."

By late 1999, when Teleglobe issued another profit warning, Sirois's aura was dissolving—for everyone, it seems, but Monty. In early 2000, with Monty now installed as CEO, BCE announced plans to buy the 77% of Teleglobe that it did not already own for $9.6 billion in stock, and to commit an additional $1 billion in funding to the increasingly troubled unit. "We will have our work cut out for us," Monty conceded. Investors balked. Shares in both BCE and Teleglobe tanked. In June, BCE buckled and renegotiated $400 million off the purchase price, which, with the reduction in BCE's share price, ended up at $6.4 billion. Still, Monty was making a fatal error, one that could only be explained by Sirois's Rasputin-like hold on his mentor. "Jean is loyal to a fault," says Lawrence Surtees, director of telecom research at IDC Canada Inc. "Yet, he is also a guy that is so steeped in the telecom business, that if there's one situation where he ought to have sat Sirois down and said 'Charles, what are you doing?', it was at Teleglobe. But he didn't."

Sirois walked away with stock in BCE that was then worth more than $800 million. Around the same time, he sold Motion to TVA Group Inc. and the caisse for $150 million, ending his foray into content. (TVA subsequently wrote off $120 million against its production unit, most of it related to the Motion acquisition.) Liberated from his responsibilities at Teleglobe and Motion, Sirois went sailing, grew a beard and chucked the tie.

The vacation was short. Within months, his entire empire was teetering. Look's aspirations were crumbling under its $250-million debt load. TIW was fending off creditors around the globe. And Microcell was struggling to justify its existence in Canada's suddenly overcrowded cellphone business.

Back at Teleglobe, Monty's problems were mountains higher. An insider recalls, "Jean was fuming mad" when he realized the extent of the mess he had inherited from Sirois. His protégé's projections of voice and data traffic proved to be wild exaggerations; overcapacity plagued the industry. Revenue per minute of voice traffic was in a free fall. BCE halved the fibre-optic construction project Sirois had started, and cancelled his $100-million sponsorship of Jacques Villeneuve's Formula One racing team. Excel, acquired for $3.5 billion (U.S.), was sold back to its original owner for about $225 million (U.S.) in 2001, with Teleglobe left holding its $1.3 billion (U.S.) in debt.

It was all too little, too late. Teleglobe still couldn't generate enough cash to service its debt, and BCE was forced to funnel funds its way. Shareholders revolted. BCE stock, long trusted by widows and orphans, lost a third of its value in the first quarter of this year.

In April, BCE finally cut Teleglobe loose, announcing it would no longer fund the unit. Monty resigned, cutting short a brilliant career at 54, as BCE announced it would write off the $8.5 billion it had invested in Teleglobe.

On its own, Teleglobe stood no chance of survival. Three weeks after Monty's departure, Teleglobe filed for bankruptcy protection, citing debts of $6 billion. Creditors were told recoveries would be "de minimus." In the hopes of attracting a buyer, Teleglobe essentially went back to being what it was before Sirois took over—a wholesaler of voice traffic capacity for overseas calls from Canada. An entire decade of development was snuffed out.

On a blustery, grey morning in May, Sirois strolled into Microcell's annual meeting. The beard was gone. The tie was back. As reporters swarmed him, a red-faced Sirois smiled sheepishly. His press handler intervened. "I've told you," she snapped, "Mr. Sirois is giving no interviews."

It's a big switch. In his heyday, Sirois spread his wireless word with evangelical zeal. Indeed, he could be counted on to proffer his ideas to reporters on almost any topic, from Quebec separation (he recruited candidates for the provincial Liberals in 1998) to how to run a company (his 2000 book Organic Management proclaimed: "The thinking or sentient being receives and attempts to understand information as soon as he or she is made aware of it.")

Could Sirois's vow of silence reflect contrition? Microcell, TIW and Look have hit penny-stock status. At their peak in early 2000, the combined market caps of Sirois's main publicly traded holdings approached $10 billion; today they're worth $100 million on a good day. In September, 2001, Look filed for bankruptcy protection, re-emerging as an even more marginal player than when it entered.

TIW has fared no better. Last July, its British division, Dolphin Telecom PLC, sought bankruptcy protection. Last fall, TIW itself was forced to seek concessions on $700 million (U.S.) in debt under a restructuring that saw Sirois's voting stake fall to 28% from 40%. With the Brazilian, Mexican and Chinese markets abandoned, the once globally ambitious company is now active in only the Czech Republic, Romania and India.

The jewel in Sirois's tarnished crown remains Telesystem's 25% equity stake in Microcell and its Fido cellular brand. But even the jewel is chipped, since the structure of the industry and its own hampered finances dictate that Microcell can neither easily grow nor sell itself. So it, too, may be forced to restructure its debt. In May, RBC Capital Markets analyst Richard Talbot issued an unprecedented 52-week outlook for Microcell's stock: nil.

Sirois has lost the Forbes billionaire status he enjoyed during the tech boom. The bondholders, shareholders, banks, BCE and the caisse, who backed him then, are unlikely to do so ever again. "It's safe to say BCE is not going to be a supporter of his in the future," says one analyst. "And I bet even he would agree that the challenges faced by all the companies he's been associated with will make it more difficult for him to raise capital [from other sources]."

" This guy did have a vision that, if not ahead of its time, was perfectly timed," adds Surtees. "But he was also driven by ambition and ego. He allowed himself to think he could be involved in so many things at once. And he dropped the ball."

Early last year, Sirois said that he would focus for the next five years on strategic alliances and small investments in companies that develop wireless and broadband applications. "It will all be small moves. But one day, people will say, 'Ah, that's what Sirois has been up to.' " Adds Garner Bornstein, CEO of Airborne Entertainment Inc., a Montreal-based wireless content start-up in which Telesystem has invested: "It's been difficult for Charles. But this is not a Charles Sirois phenomenon. Everyone in the telecom space is suffering. Charles is a tremendous entrepreneur with tremendous vision. I guarantee he will be back."

As for Sirois's "small moves," there's his stakes in more than 70 start-ups, of which none has produced a payoff. Most, like Airborne, which distributes entertainment content for wireless phones, remain in their development phase—read unprofitable. Sirois-sponsored ESP Media, which offered reduced-rate airtime to cellphone users willing to put up with ads, went bankrupt last year. And in April, Microcell closed down its i5 spinoff after it failed to sell any of its wireless applications to other carriers. Not even Sirois's 27-year-old son, François-Charles, i5's CEO, was spared.

As Dad would say, that's what happens when the lake falls.