This study takes a critical look at ‘English-only’ medium of instruction in Ghana. Like many African countries, Ghana has multiple languages. Davis and Agbenyega (2012) estimate that Ghana has approximately 49 spoken languages and dialects. Local languages are used for everyday life in local communities. English is learned in school and spoken mainly in official places. English was used by British administration as the language of government, law, education and the national newspaper (Sey 1973) and remains the sole official language of Ghana today. This study draws on a historical overview of Ghanaian language-in-education policy (from 1961 Education Act to present). The comparative case study examined the effects of English-only instruction on students in two classrooms, one in rural Ghana and another in urban Ghana. The research questions are: 1) How do students whose mother tongue is not English respond to English-only instruction? and 2) What are the distinguishing effects of the policy on rural and urban students education and lives? The two elementary schools are located in an Akan language (Twi) region. Data sources include classroom observations (level of participation, instructional strategies, and classroom discourses of emerging bilinguals) and one-on-one interviews with teachers. Teachers were asked about the respective roles of English and Twi in their daily school activities. In light of differences in rural and urban community social, economic, and infrastructural circumstances, this study provides answers to the question of how one national language-in education-policy affects students differently. Findings from the study show that the English-only medium of instruction gets negotiated and resisted by teachers. Teachers engage in code-switching, code-mixing and translation, and sometimes Twi dominated lessons in the classroom. I conclude that English-only language policy in Ghana does not work equally for the students in both schools. Although students in the Urban School have challenges with the policy, students in the Rural School are comparatively more disadvantaged. I recommend bilingual education of English and Indigenous languages in Ghana to address linguistic imbalances in its education system. Policymakers should not lose sight of the fact that each community has unique socio-economic and linguistic circumstances and that a general language policy will not work for all. Lastly, I recommend that Ghana should redefine literacy learning and the education of the nation. Education and literacy should not be defined as only people who can read, write and communicate fluently in English. People who can exhibit the same skills and qualities in their mother tongue must be accorded the same recognition and should be allowed to play elite vital role in formal settings.