A strong argument in favour of the existence of rightward, string-adjacent head movement comes from Han, Lidz & Musolino (2007). They argue that Korean language-internal variation with respect to the scopal order between negation and universal quantifier objects shows that in the variety where negation takes wide scope, the negative marker must have moved along with the verb/auxiliary to the head of IP. This would then constitute evidence for rightward string-adjacent head movement. In this paper, I argue that this analysis actually makes different predictions than have been attested in Korean. Moreover, I argue that, following a well-known stand with respect to the nature of polarity-sensitivity (Chierchia 2013), these facts follow naturally once it is assumed that in one variety but not the other, universal quantifiers are Positive Polarity Items (PPIs). This makes the attested language-internal variation in Korean less exceptional since language-internal variation with respect to polarity-sensitivity is something that can be widely attested.