A method for retrospectively assessing the average radon concentration in a dwelling, for a period as long as 40 years or more, is described. It is based on the free penetration of radon gas into bulky spongy materials called volume traps, and subsequent trapping of radon decay products inside them. This leads to a build-up of 210Pb, a radon decay product with a half life of 22.3 years, which reaches an equilibrium with the alpha emitter 210Po, indicative of the average radon concentration over the exposure period. The stuffing of mattresses, chairs, cushions, etc. can be considered a good volume trap. A chemical separation procedure is described, transforming the volume trap into a 210Po alpha source, whose activity is determined through alpha spectroscopy. The technique has been applied in over 300 samples, retrieved from dwellings in Serbia, Germany, Norway and Sweden. The technique has proven to be very robust and reproducible. In cases where radon data were available during the exposure period of the volume trap, a remarkable agreement was found. The best volume traps appeared to be polyester sponges found in many stuffings. The major drawback is the destructive nature of the method, its major advantage is the direct relationship between the radon gas and the 210Po signal. The technique can be useful in epidemiological studies of the lung cancer risk from indoor radon, for assessing the long term radon exposure.