Linguistic relativity is the principle that the structure of the language that one speaks influences their cognition. This dissertation extends this principle into sociological territory, arguing that the structure of the language that a group speaks influences how its members interact. In the first chapter, I outline the new arsenal of computational tools that have been developed to work with the vast new stores of digital text data that have become available, and I highlight how these tools and data have been used by social scientists to glean insights about the social world. In the second chapter I put these tools to work on the development of a new language structure attribute, information density, which is the average amount of information contained within the words of a language. I show that there is significant variation in the rate of information density across languages, and that this rate is systematically related to rates of conceptual density and speech information transfer across languages. In the third chapter, I investigate whether the language information density rate is associated with the performance of expeditions to the Himalayas. I find that language information density is associated with greater expedition performance and with faster performance, especially within low-hierarchy teams where the benefits of communication would be most likely to be observed. This dissertation aims to advance a sociology of language wherein variation in language structure is the analytical input that shapes processes of social interaction, collective cognition, and group performance. It points to the importance of considering language structure as a significant force shaping social life.