This study aims at testing if Turkish L2 users of English process predictive conditionals different than Turkish monolingual speakers in accordance with Multi-competence Theory (Cook, 2003). For data collection grammaticality judgment tasks (GJTs) are used and unlike traditional GJTs, contexts are added to clarify the tasks. The participants consist of 15 monolingual and 15 bilingual Turkish university-graduates. The results are evaluated using both descriptive statistics such as mean, median and standard deviation and also SPSS to show that the differences found between the groups are statistically significant. The analysis of the results shows that there is a statistically significant difference between monolingual and bilingual Turkish speakers. This finding implies that second language (L2) users have a different language system than monolinguals and they should be evaluated and compared for both their first language (L1) and L2 on their own terms rather than monolingual native speaker norms.