We do not draw anything until we have slept near the site. On Hvaler that meant three days in a borrowed boathouse, a pair of boots each and no laptop.
A survey tells you where the rock is. It does not tell you where the wind comes from at four in the afternoon in October, or which hollow stays dry when the rain moves in from the Skagerrak. Those things only arrive slowly, and only if you stay long enough to stop looking for them. So before the first sketch for Havhuset, Amara, Tobias and I took the ferry out with a tent, a thermos and a roll of tracing paper we promised not to use.
The first day is always disappointing. You walk the boundary, you find the survey pegs, you stand where the living room ought to go and wait for the place to speak. It rarely does. What you notice instead are small, stubborn facts: a seam of quartz in the granite, a juniper bent flat by the north wind, the exact point where the path from the jetty gives up and becomes a scramble.
By the second day the facts begin to arrange themselves. We sat on the rocks at dawn and at dusk and timed the light across the bay. We followed the water after a squall to see where it pooled. We found a sheltered shelf on the lee side, a few metres below the obvious summit, where you could hear the sea without being in it. Nobody said that was where the house should be. We just kept going back there to eat.
The third day is for the neighbours. An old fisherman two coves over told us which winters the ice came into the sound, and where the boathouse roof had lifted in the storm of 1992. His daughter showed us the path the children used to take to school by boat. None of it appears on a drawing, but all of it shaped one.
When we finally unrolled the tracing paper back in Oslo, the plan came quickly. It was not invention so much as recall. We still do this on every project, even the small ones, and we write the days into the fee. Clients sometimes ask whether it is necessary. We have never once regretted the time.
Written by Eirik Fjellheim, founding partner.
