The Nahuan languages are group of closely related languages spoken in Mexico and El Salvador, which form a well-defined sub-branch within the southern branch of the Uto-Aztecan language family. They are the only Uto-Aztecan languages that form part of the Mesoamerican linguistic area. Proto-Nahuatl displays assimilation to the languages of Mesoamerica in all aspects of linguistic structure including phonology, morphology, and syntax. Surprisingly, the lexicon does not appear to have been affected to the same degree, as most of the core vocabulary of Proto-Nahuatl can be traced back to Proto-Uto-Aztecan. Identification of borrowings between Mesoamerican languages has played an important role in studying prehistoric processes of the area. For example the word cacao, proposed by Campbell and Kaufman (1976) as borrowings from a Mixe-Zoquean language that was widely diffused within the region has been seen as significant argument for the identification of the Olmec culture as Mixe-Zoque speaking. However, Dakin and Wichmann (2001) later argued that the word ‘cacao’ might have been of Uto-Aztecan origins and suggested that Nahua speakers had an early presence and a dominant role in trade networks in Mesoamerica (Dakin 2003). This argument was rejected by Terrence Kaufman and John Justeson (2007, 2009) who maintained that prior to the rise to dominance of Nahuan speaking peoples in the Post-Classic period, Proto-Nahuatl was primarily a recipient of borrowings from other Mesoamerican languages. They proposed a number of additional borrowings from Mesoamerican languages into proto-Nahuatl, from Mayan, Tepehua-Totonacan, and Mixe-Zoquean languages. These proposals of borrowings into Proto-Nahuatl have been used to locate the place of origin of Nahuan languages in the North-Eastern periphery of Mesoamerica during the Classic Period rather than in North Western Mexico closer to the other Uto-Aztecan languages (e.g. Beekman & Christensen 2003), or whether the protoNahua community was already located within central Mexico as argued by Dakin (2003). This challenges us to find out whether Proto-Nahuatl was indeed mainly a recipient language in Mesoamerica, or perhaps also a donor. Ongoing reconstruction work on proto-Nahuatl and the intermediary stages of Uto-Aztecan shows that many loans identified by Kaufman and Justeson can equally well be seen as inherited from proto-Uto-Aztecan, suggesting the opposite direction of borrowing. However, Proto-Tepehua-Totonacan and Proto-Mixe-Zoquean lexicons have demonstrated cases where there are viable reconstructions in both language families, making it a hard to determine the direction of borrowing. Therefore, there is a pressing need to develop methodologies to assess and evaluate the overall probability of the different borrowing scenarios involving the Mesoamerican languages. The paper describes the challenges involved and suggests some avenues for developing an approach to this challenge.