However, looking back from the vantage point of today, we can see clearly a range of so-
cial and ecological costs that were not adequately considered when these dams were built.
Forexample,CeliloFalls,asiteofNativeAmericanhabitation fromtheearliest daysofhu-
man occupation of North America (more than 12,000 years before the present) and a signi-
in the 1950s. By the time the dam was constructed, local Native Americans had already
been dispossessed of their resources and involuntarily moved onto reservations. The Grand
Coulee Dam, completed in 1942, effectively cut off 1,700 kilometers of salmon-spawning
habitat in the upper reaches of the Columbia basin. Many topics have been written that eu-
logize the Columbia, mourning its loss as a site of ecological treasure, cultural memory,
and even spiritual significance (e.g., Harden 1996). 15 And the story of damage to ecosys-
tems and displacement of native communities was repeated wherever dams like this went
up in the United States.
Yet policy decisions are underpinned by human values, which constantly evolve. Amer-
ica is a different country today than it was seventy years ago. In 2007, several colleagues
and I hosted an international conference on dams in Washington State, in a beautiful loca-
tion along the Columbia Gorge about forty miles east of Portland, Oregon. As part of the
conference, we visited the Bonneville Dam, the lowest of fourteen dams along the main
stem of the Columbia River. As we toured the facility and descended by elevator into the
powerhouse, where we could feel the great turbines shaking as they sent power out to the
electrical grid, one of the members of our party asked the guide about strategies for im-
proving fish passage. Fisheries issues, particularly for the five species of anadromous Pa-
cific salmon, whose life cycle requires a traverse from freshwater rivers to the ocean and
back again, have become paramount concerns in the Northwest. In response, the guide pro-
ceeded to describe several design characteristics, including the shape of the turbines and
the presence of a fish ladder, and several operation strategies that manage the flow in such
a way as to reduce fish mortality.
Impressed by this answer, another conference participant asked about the effects of these
strategies on reducing fish mortality. Did such measures work? The guide responded that
their efforts had reduced fish mortality to about 10 percent. I recall thinking briefly about
what a triumph of engineering it was that nine out of ten fish could safely pass through the
structure, until I remembered that there are more than ten dams on the Columbia. Upstream
on the tributaries in Washington, Idaho, and Canada, the best historical spawning grounds
in the Columbia watershed, Coho salmon have gone extinct, and other species are listed as
threatened or endangered.
But as societal values shift over time, so do the policy decisions that stem from them.
In a subsequent conversation, a senior scientist at the Bonneville Power Administration—a
federal agency under the U.S. Department of Energy, with responsibility for operating the
Columbia River dams and distributing the electricity—told me that his organization takes