The relationship between local recurrence and survival has been examined using the example of soft tissue sarcomas of the extremities. Studies that appear to indicate that local recurrence does not jeopardize survival are shown to have been inappropriately analysed, to have inadequate patient numbers, or both. A survival deficit due to local tumour recurrence cannot be excluded on present data, and clinical decisions regarding conservative versus radical surgery will therefore continue to involve an element of value judgement.