The Dam That Was Not Built
Latvia dammed its main river three times and drowned the landscape its folk songs are made of, and the one time the public objected out loud, in 1958, it lost. When a fourth dam was proposed in the mid-1980s the objection worked — and the campaign that stopped it became the movement that took the country out of the Soviet Union.
The castle at Koknese was blown up by retreating Saxon troops in 1701 and has been a ruin ever since, which is not the odd thing about it. The odd thing is where the water is. The ruin stands on what was a high bank above the Daugava, twenty-odd metres up, and today the river laps the foot of the masonry and the whole promontory is a narrow spit with picnic tables on it. The castle did not move. In 1966 the reservoir behind a new dam fifteen kilometres downstream came up to meet it, and it has been standing in the water ever since.
There are three dams on the Latvian Daugava and there were nearly four. The three that exist supply a substantial share of the country's electricity — a great deal more in a wet year than in a dry one, which is the character of hydro everywhere. The fourth was designed, sited and approved, and then it was stopped, and the stopping of it is the reason a visitor to Latvia can still stand in a river valley that ought to be under forty metres of water.
The cascade that got built
Ķegums came first, begun in 1936 under the Ulmanis government, built largely by a Swedish contractor and opened in 1939 — the largest thing independent Latvia had built, and independence outlasted it by less than a year. Pļaviņas followed between 1961 and 1966, at Aizkraukle, and it is still the largest hydroelectric plant in the Baltic states: spillway and powerhouse combined in a single structure because the dolomite bed would not carry them separately. Riga HPP, the last and lowest, went in during 1974.
Pļaviņas took the most. Behind it went the rapids, the Pērse waterfall at Koknese, and Staburags — a cliff of tufa laid down over centuries by a calcium-rich spring, some eighteen metres of it, and the single most-sung piece of rock in Latvian folk poetry. There was a public argument about this in 1958, conducted in the newspapers, which is a remarkable thing to have happened in the Soviet Union at all. Writers and scientists said the valley was worth more than the megawatts. They were denounced for it, some of them lost their positions, and the dam was built on schedule.
October 1986
Twenty-eight years later the ministry in Moscow proposed a fourth station, at Daugavpils. The reservoir would have flooded the stretch of valley above the city where the river turns back on itself in a series of deep meanders, together with a set of villages in it. In October 1986 the weekly Literatūra un Māksla published an article about the project by a young journalist, Dainis Īvāns, and an engineer, Artūrs Snips. It was not a manifesto. It was a piece of technical and economic scepticism about whether the plant was needed, what it would cost and what it would take, and it was carefully argued precisely because the people who lost the 1958 argument had not been.
It was permissible to be angry about a river when it was not yet permissible to be angry about anything else, and everybody involved understood exactly what they were doing.
Why a dam and not a flag
The response was tens of thousands of letters, which under glasnost was a quantity nobody in the apparatus knew how to dismiss. In November 1987 the Soviet Council of Ministers cancelled the project. It is worth being clear about why an environmental campaign was the one that worked: national grievance was still unsayable in 1986 and a badly costed hydro scheme was not, so the argument about who decided what happened to Latvia was conducted, for two years, entirely in the language of engineering. A proposal for a Riga metro was killed the same way over the two years that followed, on the grounds of what tunnelling would do to the Art Nouveau quarter and to the city's soft ground. Both campaigns were about infrastructure and neither was only about infrastructure.
Helsinki-86, founded in Liepāja in July 1986, had been saying the unsayable directly and its members were arrested, deported and harassed for it. The river campaign said something narrower and could not be arrested for it, and it reached a much larger number of people. In October 1988 the Popular Front of Latvia held its founding congress and elected Dainis Īvāns — the journalist from the Daugava article — as its first chairman. Ten months after that, on 23 August 1989, roughly two million people across the three Baltic republics stood in a line from Tallinn to Vilnius holding hands. Latvia declared the restoration of its independence in May 1990, and the Soviet Union recognised it in September 1991.
The valley is still there, and so are the dams
Daugavas loki, the loops, is now a nature park and has spent years on Latvia's tentative list for World Heritage. You reach it from Daugavpils in half an hour and there is a wooden tower at Vasargelišķi that looks straight down the best of the meanders. Below it is Slutišķi, a village of squared-timber houses belonging to Old Believers whose ancestors left Russia after the church schism of the 1660s and who kept the old rite in a corner of Latgale where nobody troubled them about it. All of it would have been reservoir.
Nobody in Latvia proposes taking the three existing dams out, and that is the honest end of the story. The Daugavpils project has been raised again once or twice and has never got anywhere, but the three existing plants are not controversial in Latvia in the way that Staburags still is. They are the largest domestic generation the country has that does not burn imported gas, and after the Baltic states cut themselves out of the old Soviet BRELL ring in February 2025 and synchronised with the continental European grid the following day, flexible capacity that can be turned up in minutes is worth more than it was. What the dams do not have is a working way past them for fish: the salmon that once ran to the headwaters stop at the first wall, and the run is now kept going by a hatchery. Standing at Koknese with the water at the castle wall, the useful thought is not that a mistake was made. It is that the same country made the same decision three times and then, on the fourth, was asked and said no — and that being asked at all turned out to be the thing worth having.