Drawing from research on racial formation processes in discourse concerning immigrants to Italy, this article argues that when racializing discourses are introduced as unmarked in a conversation, the coparticipant is put in a position of having to readily agree or openly disagree—with the second option endangering face. Racializing discourses introduced as unmarked thus tend to obtain acquiescence from the coparticipant. At the same time, they work at constructing shared identities around racist and racializing stances, where the unmarking becomes indexical of such shared identities. Instead of seeing racism as a stable belief in the mind of the individual, my analysis suggests that we should consider it as something the person does in interaction. The requirements of conversational rules of engagement, the needs of face, the relationship among the interactants, as well as their moral view, influence the participants' response to racializing discourses in interaction, and whether they might coconstruct or oppose them. [racialization, markedness theory, facework, racism, immigration, Italy]