My paper builds on my research and publications on gender and class in the study of Islam in Africa”. In the paper I am proposing for MESA, I engage in a comparative analysis across time – 1877 to 1994, and across space – from North Africa to East and West Africa, based on three case studies including my own of a Moroccan concubine. My analysis roots an understanding of the experience of Islam in Africa within the household where leading male figures of so-called ‘public’ Islamic society lived their religion next to and in interaction with mothers, wives, sisters and slaves, in their own families. It argues that those who are seen in traditional studies of Islam in Africa as marginal, namely women and slaves, were in fact central to the process of defining the experience of “being Muslim” over time and space. In their households, usually hidden from external eyes, they shaped how Islam was lived around them. This paper attempts to address the question of what is Islamic about slavery in Muslim societies, which was originally raised by Fred Cooper in his 1981 article "Islam and Cultural Hegemony: the Ideology of Slaveowners on the East African coast.” In this paper, I engage in a comparative study based on three case studies, each well known among Africanists: ‘Baba of Karo, a free Hausa woman of Northern Nigeria, 1877-1951); Bi Kaje, a free Swahili Woman of Mombasa, Kenya,1890-1981/2; and Fatma Barka, a slave then freed woman of Goulimine-Southern Morocco, (1900/1994). The first is a narrative collected and constructed in 1948-50 by the anthropologist Mary Smith; the second is a biography based on interviews by the historian Margaret Strobel in the 1970s, and the last is my own series of interviews with Fatma, a Malian slave and concubine to a Moroccan merchant, and her family in 1993-19944. From this research, spreading over the second half of the 20th C, I will attempt to derive insights into what the experience of being female-mistress and/or female-slave can tell us about twentieth-century ‘Islamic Slavery’ in West, East and North West Africa, respectively. In concluding, I will return to Cooper’s seminal question and suggest how looking at class and gender in the context of the household contributes to answering it.