Semantic communication (SemCom) has emerged as a 6G enabler while promising to minimize power usage, bandwidth consumption, and transmission delay by minimizing irrelevant information transmission. However, the benefits of such a semantic-centric design can be limited by radio frequency interference (RFI) that causes substantial semantic noise. The impact of semantic noise due to interference can be alleviated using an interference-resistant and robust (IR$^2$) SemCom design. Nevertheless, no such design exists yet. To shed light on this knowledge gap and stimulate fundamental research on IR$^2$ SemCom, the performance limits of a text SemCom system named DeepSC is studied in the presence of single- and multi-interferer RFI. By introducing a principled probabilistic framework for SemCom, we show that DeepSC produces semantically irrelevant sentences as the power of single- and multi-interferer RFI gets very large. Corroborated by Monte Carlo simulations, these performance limits offer design insights -- regarding IR$^2$ SemCom -- contrary to the theoretically unsubstantiated sentiment that SemCom techniques (such as DeepSC) work well in very low signal-to-noise ratio regimes. The performance limits also reveal the vulnerability of DeepSC and SemCom to a wireless attack using RFI. Furthermore, our introduced probabilistic framework inspires the performance analysis of many text SemCom techniques, as they are chiefly inspired by DeepSC.