Loneliness has been consistently identified as one of the specific ‘social problems’ which accompanies old age and growing older: 90 per cent of the general population of Britain feel that loneliness is a problem associated with old age. There is a widespread presumption that loneliness and isolation have become more prevalent in Britain in the period since the Second World War as a result of the decline in multi-generation households and changes in family structure. This paper examines the accuracy of this stereotype and considers if current cohorts of older people are more likely to report experiencing loneliness than previous generations of elders, through a comparative analysis of historical and contemporary data. Historical data are provided by three ‘classic’ social surveys undertaken in England between 1945 and 1960. Contemporary data are from a postal survey of 245 people aged 65–74 living in South London in 1999. The questions used in all four surveys were comparable, in that respondents self-rated their degree of loneliness on scales ranging from never to always. The overall prevalence of reports of loneliness ranged from five to nine per cent and showed no increase. Loneliness rates for specific age or gender sub-groups were also stable. Reported loneliness amongst those living alone decreased from 32 per cent in 1945 to 14 per cent in 1999, while the percentages decreased for both those reporting that they were never lonely and that they were ‘sometimes’ lonely.