In contexts where English is considered a foreign language, the classroom provides the only exposure to language for most learners. As such, it becomes an important site to investigate the cause(s) of low proficiency and learning problems that have been documented in a number of studies carried out in Saudi Arabia. What motivates this study is a hypothesis that poor student performance is due to the lack of opportunities to use language inside the classroom. To investigate this, I conducted a semester-long study of one EFL classroom. Through close transcription of talk, the study identifies patterns of language use, norms of participation, typical communicative events, and amount of participation in these events. It attempts to show how students are socialized to use English in the process and find out what counts as language learning and its relation to students opportunities to use English. A quantitative analysis at the turn and word level was carried out on a sample taken from all speech events. It shows that student talk remains scarce and in most cases does not reach sentence level. The turn is shown to be a crude measure of the distribution of talk since analysis at the word level provides a more fine-grained measure that reflects the true nature of classroom discourse. The next phase of the analysis utilized a qualitative approach to capture the complexity involved in analyzing spoken discourse. Within speech events, exchanges were coded through inductive cyclic analysis of data and open coding. The patterns of communication identified were divided into two major groups. The first group highlights interactional details, while the second reveals the instructors assumptions about language and their relation to what counts as knowledge and language learning. Findings suggest that classroom talk engenders grammatical accuracy and knowledge of discrete vocabulary items. This is shown to have a direct negative effect on fluency and opportunities for language use. The overall nature of classroom talk is characterized as a one-way transmission of knowledge that does not allow for student interaction. The study offers suggestions for further research, implications for classroom practices, and venues for professional development.