The discussion on whether some conversational implicatures (CIs) are more ‘default’ than the other has taken place for a long time. While neo-Griceans (NG) insist on the distinction between generalized and particularized CIs, which are said to differ along numerous dimensions, so far, most studies focusing on computational speed showed any enrichment is more costly than the literal understanding, and therefore challenge the distinction between PCIs and GCIs. In this study, a novel approach – deceptive language with false implicatures – was used to test speakers of German and Mandarin Chinese. The main findings show that (i) false GCIs resemble verbal utterances and thus correspond to lies, while PCIs are congruent with non-linguistic deceptions based on actions. We argue that this observation, in opposition to most previous experiments investigating the GCI-PCI complex, supports the theoretical distinction made by NG. (ii) The response behavior of German and Chinese participants seems to be very similar when sociocultural factors are controlled for, suggesting that this pattern is mandated linguistically. Furthermore, two control experiments reveal that the patterns observed are not due to moral judgments but that they were caused by a genuine linguistic distinction.