This paper explores the link between sign language research and sociolinguistic theory. It demonstrates how sign language research benefits from a sociolinguistic approach and provides validation for sociolinguistic theory. Previous research on the sign language of the deaf is reviewed, and a distinction is made between "structure-oriented" and "community-oriented" approaches. While the former sees language as an abstract, autonomous system, the latter views it as part of the communication process among human beings. It is argued that the community-oriented approach, which is sociolinguistic in nature, is particularly well suited to sign language research. This is because it avoids pitfalls of the structure-oriented approach, in particular, an orientation toward spoken language and the assumption that sign is homogeneous. It is shown that the usefulness of the community-oriented approach to sign language provides validation for the concepts, methods and scope of sociolinguistics, and thus supports those who argue for a radical change in linguistics along sociolinguistic lines. (Author)