Cross-linguistic influence (CLI) has been claimed to occur under the conditions of structural overlap, interfacing, syntactic complexity and language dominance. This study tested adjective placement in the Italian of 19 adult German–Italian simultaneous bilinguals, comparing naturalistic and experimental data. The results show no CLI from German, although the conditions for CLI are given. Instead, bilingual adults tend to overuse a structure that is unique to Italian, unlike bilingual children in previous studies. However, they do so only in the experimental data. In order to account for this, I introduce the concept of cross-linguistic overcorrection in contrast to cross-linguistic influence.