North Koreans in South Korea often try to adopt South Korean accents to avoid discrimination, with varying degrees of success. Language attitudes studies have tended to investigate reactions to distinct varieties rather than speakers’ attempts to approximate new varieties for their own benefit, and while a few have considered the effects of listeners’ categorization of speakers on evaluations of them, they have generally focused solely on macro-categories such as place of origin, especially in languages other than English. This study investigates South Koreans’ attitudes toward North and South Koreans’ native and adopted accents and how these attitudes may relate to listeners’ categorization of the speakers. Eighty-two South Korean listeners rated recordings of three Korean speakers, two from different cities in North Korea and one from Busan, South Korea, who each read the same text in two versions: their native variety and their adopted Seoul variety. Listeners only consistently identified the Pyongyang variety as North Korean, but rated all three speakers more positively in Seoul guises than in their native guises. Additionally, when North Korean speakers were identified as being from outside South Korea, ratings were less positive. However, the Pyongyang speaker was rated more highly than the South Korean speaker on positive qualities, and qualitative data suggests that other types of speaker categorization may be relevant to attitudes, such as ‘professional’ for the Pyongyang speaker in both guises and ‘youth’ for the South Korean speaker in her Seoul guise. Thus, attitudes studies may benefit from qualitative data on speaker identification beyond the usual macro-categories, addressing the multiple categories indexed by a speech sample.