Problems with the lexical evidence used to posit Hadza as an Afroasiatic language are discussed. The failure to identify certain loanwords and the misanalysis of certain Hadza morphemes are problems rectified by having greater familiarity with Hadza and its linguistic contact history. Other problems are more general to the specific methodology employed. The overly wide semantic range of meanings often allowed in establishing form-meaning sets increases the likelihood of chance resemblances. The use of certain words that are likely onomatopoeic also reduces the impact of the proposed cognate sets. Ultimately, it is the lack of regular, repeated sound correspondences between Hadza and Afroasiatic that makes the proposal of their familial relationship unconvincing.​​​​​​​