Social media is international: users from different cultures and language backgrounds are generating and sharing content. But language barriers emerge in the communication landscape online. In the quest for language diversity and universal access, the vision of a cosmopolitan Internet has stumbled over the language frontier.

Expatriates, minorities, diasporic communities, and language learners play an important role in forming transnational networks, creating social ties across borders. Many users of social media are multicultural and multilingual; they are mediating between language communities. In the microblogging site Twitter, information spreads across languages and countries. How are multilingual users of Twitter connecting language groups? What are the factors influencing their language choices? This research advances a step towards understanding the network structures and communication strategies that enable intercultural dialog, cross-language sharing of information, and awareness of global problems.

This dissertation research aims at: (1) exploring the ways in which multilingual users of Twitter are connecting different language groups in their social network; (2) modeling how the network influences their language choices; (3) and exploring what the textual features of their posts can elicit about language choices and mediation between groups.

This dissertation goes beyond survey information about multilingualism and provides a deeper understanding about the structural relations between language communities in Twitter. This research work is one of the few that apply social network analysis to the study of sociolinguistic questions on the Internet. Focusing on the social networks of multilingual users, this dissertation contributes an original classification of network types based on the patterns of connections between language groups. Also, it applies the novel idea of modeling the influence of network factors in the language choices of the user. Finally, this dissertation tests the hypothesis that the type of exchange influences language choice, and explores with a theme analysis how other textual features might elicit cross-cultural awareness. These results can inform the design of social media platforms.