The aim of this article is to redefine the phenomenon of borrowing on the basis of the Corpus of Spoken Lithuanian: to analyse the frequency and use of foreign words, to determine the proportion of new borrowings among foreign words, and to examine whether some of the foreign words may be treated as the instances of code switching or code mixing. The article attempts to make a distinction between new borrowings and other foreign words and discusses the phenomenon of code mixing in spoken language for the first time in Lithuanian; it also explains the terms of code switching and code mixing, which have not been discussed in Lithuanian linguistics. The results of the quantitative research have revealed that foreign words comprise only 0.33% of all words in spoken language. The largest proportion of these words is English words and phrases (93%), while lexemes or phrases from Russian, Latin, or Italian comprise only 7% of the words. The research has determined that new borrowings constitute 44% of all foreign words in the conversations under the present investigation, and code mixing has been employed in 56% of all instances. The research has demonstrated that code mixing is the most typical of spoken private speech, while the use of new borrowings is more frequent in more formal registers of spoken language, i.e. academic and media language.