The roots of the safety plug decision could actually be traced back a decade, before PB had even become involved in the project. In 1989 the Massachusetts firm Metcalf & Eddy produced55 a conceptual design for the Deer Island tunnel and diffuser risers. To prevent ocean water from infiltrating the sprinkler nozzles on those diffuser heads during tunnel construction, M&E designers called for sturdy covers to be installed on each nozzle. But if an anchor ripped open the diffuser head, nozzle covers would obviously offer zero protection. So the designers decided a “secondary” safety plug should be placed lower in the riser pipe. They laid out three options for this secondary seal. Option 1 would put a plug inside the diffuser head, just out of reach of any anchor but still close enough to the ocean floor that the plug, like those nozzle covers, could ultimately be removed by divers swimming down to the sea bottom. Option 2 would put a plug inside each of the pipes connecting the risers to the tunnel, meaning the plug would ultimately have to be removed by crews working from inside the tunnel. Finally, option 3 would skip the secondary plug in favor of hiring a guard boat to patrol the waters above the diffuser heads, keeping that particular stretch of ocean free from boats and their anchors for as long as the sandhogs were in the tunnel.
M&E pointed out that the second option was problematic because of “the difficulty of re-entering the completed tunnel (with no track, lighting, or ventilation) to remove the plugs.” The firm’s report concluded with this assessment: “At first sight, solution 2 appears to be cheaper than 1, but it is believed that when the difficulties noted are taken into account, plugging from the top, which is the safest, will also turn out to be the cheapest.”
But because M&E was responsible only for the conceptual design, its advice was not binding. The contract for the final design of the tunnel and diffusers went to Parsons Brinckerhoff, a bigger global infrastructure planning company whose letterhead boasted “Over a Century of Engineering Excellence.” PB settled on a different design for the diffuser head. This one was mushroom-shaped and resembled a domed Apollo 11 command module, a design that would improve its ability to withstand a hit from an errant anchor. Still, PB agreed that a secondary safety measure would be prudent. It settled on safety plugs—which it called bulkheads—that would sit in the elbows of the pipes connecting the tunnel to each of the fifty-five risers. PB decided that option 1, which M&E had identified as the safest route, would present too many problems when it came time to remove the plugs. So it went with an internal plug design that resembled option 2, the one M&E had characterized as appearing at first sight to be the cheapest route.
But no one—not tunnel owner MWRA, not tunnel designer PB, not construction manager Kaiser—spelled out how those plugs could be safely removed. Instead, they transferred the responsibility for figuring that out to the contractor, kicking the can down the road. But they did make one crucial stipulation. The contract mandated that the safety plugs could be removed only after the tunnel was completed, meaning after the sandhogs had cleared out, taking their extensive ventilation, transportation, and electrical systems with them.
The rationale behind this unusual stipulation appeared to involve worker safety. PB designers said the overriding goal should be to limit the amount of time when there would be lots of workers in the tunnel who were protected from the ocean by just one safety seal, namely the covers on the nozzle caps.
In 1990, after winning the bid56 to build the tunnel, Kiewit signed the contract that required it to come up with a plan for “providing lighting and ventilation (or breathing apparatus) for the personnel” who would remove the plugs. By suggesting equipment “such as battery powered vehicles [and] self-contained support services (i.e. scuba gear),” the contract clearly envisioned that the plugs would be removed only after the main ventilation system and other utilities had been dismantled. Although the contract was silent on exactly how such a complex operation could be safely accomplished, it clearly assigned Kiewit the responsibility for figuring it out. So when Kiewit would later complain about the grave dangers inherent in following these specs and sequence, the other project players could all respond, with some justification, by pointing to the contract and saying: You bid it, you own it.
In Kiewit’s defense, though, the contract to build the world’s longest single-entrance tunnel—one of the riskiest underground projects ever undertaken—sprawled over nearly a thousand pages, with a breathtaking inventory of challenges. Many of them seemed a whole lot more pressing than removing a few dozen backup safety plugs at the end of the job. To the extent that Kiewit executives57 thought about removing the plugs, they assumed they’d be able to yank them out at the same time they were pulling out the bag line, electricity, and rail tracks. In truth, during the early years of the project, they barely thought about the plugs at all.
As head of the agency that owned the tunnel, Doug MacDonald knew that as much as the thing was a one-of-a-kind technical marvel, for the contractor it was shaping up to be a financial disaster. Kiewit was already in the red58 tens of millions of dollars, since the project had become so much costlier and more complicated than it had anticipated in its original $202 million bid.
On top of that, Kiewit was incurring late-fee penalties of $30,000 every day. Although these so-called liquidated damages are often negotiated downward at the end of a job, on paper at least the contractor had, by the spring of 1997, incurred a total penalty of more than $17 million—and counting. Complicating matters was the surprising way in which contractors were paid for big jobs structured like this one. Even though MacDonald’s agency was considered the tunnel owner, Kiewit had to front pretty much all the construction costs and would get paid only when it “sold” completed sections of the tunnel to the MWRA, piece by piece. MacDonald was grateful at least that Kiewit was such a big operation, since a less capitalized company would likely have already gone into bankruptcy. What worried his team, though, was how aggressively Kiewit might use its deep pockets to try to crawl its way back into the black. To win the legal claims war, Kiewit would need to demonstrate that the MWRA or Kaiser or PB had made errors and should therefore be forced to pay for them. In other words, Kiewit would try to transfer huge chunks of that red ink to the others.
In fact, the MWRA had already settled several of Kiewit’s claims, agreeing to pay the contractor an additional $32 million for unforeseen problems. Following Kiewit’s corporate approach of matching power to power, those settlement negotiations had taken place directly between MacDonald and Ken Stinson, the chairman and CEO of Kiewit’s construction group. MacDonald had to admit that he found himself a bit intimidated by Stinson, a tall, bald man in his mid-fifties who spoke in a crisp manner and exuded the aura of a natural-born leader. In his interactions with him, MacDonald felt as though Stinson were thinking, I’ll talk to you. But you don’t know what you’re doing.
Stinson had a long, successful track record. During the Vietnam War, he’d been a company commander for the Seabees, the navy’s construction arm. In 1969, as a graduate student at Stanford, he’d taken a summer internship with Kiewit, which unexpectedly had turned into a lifelong career. He lived and breathed59 the Omaha company’s corporate culture. It had been shaped by the late Peter Kiewit, who had turned his grandfather’s bricklaying outfit into a construction powerhouse. With his motto “Know your costs,” Peter Kiewit had insisted that all his managers keep close track of project expenses so there would be no nasty surprises at the end of a job. Because the company remained both privately held and intensely private, it had been able to maintain that corporate culture long after Kiewit’s death in 1979. For Stinson, the Deer Island project presented a serious threat, given how far its costs had diverged from original estimates.
55 (Chapter 3, note 9)
Metcalf & Eddy produced: Metcalf & Eddy, “Conceptual Design: Design Package 6, Effluent Outfall Tunnel and Diffusers,” final report, May 31, 1989.
56 (Chapter 3, note 10)
In 1990, after winning the bid: “Boston Harbor Project—Effluent Outfall Tunnel, CP-282,” MWRA Contract No. 5637, official notice to proceed given by MWRA on August 9, 1990. The specifications concerning “scuba gear” appear in Part 3, Section 3.01, “Equipment.”
57 (Chapter 3, note 11)
To the extent that Kiewit executives: Ken Stinson and Kirk Samuelson, interview by author, September 2011.
58 (Chapter 3, note 12)
Kiewit was already in the red: “Boston Harbor Project, CP-282 Effluent Outfall Tunnel, Status Report for the Office of the State Auditor,” May 25, 2000; also, project players representing the contractor, designer, owner, and construction manager, interviews by author.
59 (Chapter 3, note 13)
He lived and breathed: Stinson and Samuelson interview; Dick Robbins interview.