This study investigates how inaccurate phoneme processing affects recognition of partially onset-overlapping pairs like "DAFFOdil-DEFIcit" and of minimal pairs like "flash-flesh" in second-language listening. Two cross-modal priming experiments examined differences between native (L1) and second-language (L2) listeners at two stages of lexical processing: first, the activation of intended and mismatching lexical representations and second, the competition between those lexical representations. Experiment 1 shows that truncated primes like "daffo-" and "defi-" activated lexical representations of mismatching words (either "deficit" or "daffodil") more for L2 listeners than for L1 listeners. Experiment 2 shows that for minimal pairs, matching primes (prime: "flash," target: "FLASH") facilitated recognition of visual targets for L1 and L2 listeners alike, whereas mismatching primes ("flesh," "FLASH") inhibited recognition consistently for L1 listeners but only in a minority of cases for L2 listeners; in most cases, for them, primes facilitated recognition of both words equally strongly. Thus, L1 and L2 listeners' results differed both at the stages of lexical activation and competition. First, perceptually difficult phonemes activated mismatching words more for L2 listeners than for L1 listeners, and second, lexical competition led to efficient inhibition of mismatching competitors for L1 listeners but in most cases not for L2 listeners. (Contains 4 tables, 4 figures, and 3 footnotes.)