Through an experiment investigating the processes used by several speakers of Asian languages to decode passages by speakers of Australian English, this paper seeks to establish and categorize the types of problems encountered by multilingual speakers when decoding the speech of monolingual speakers in one of their (the multilinguals') second languages. The paper first sketches the differences between problems of encoding and decoding by a multilingual when communicating with a monolingual in one of his (the multilingual's) second languages. The authors then present a preliminary typology of decoding problems of multilingual speakers. Three main types of errors are isolated: omissions, intrusions, and replacements. Subtypes under these headings are considered and the relative roles of the phonological, semantic, and syntactic mechanisms are discussed. The final remarks pose questions for further research with a view to consideration of the differences between monolingual and multilingual decoding. (VM)