Linguists give definitions of code-mixing in a similar way. Ho (2007) defines code-mixing as the change of a language to another inside the border of sentence at the levels of phonology, lexicon, grammar and orthography in the same utterance. Code-mixing is an alternation of two languages. It can be in the form of words, phrases or larger units, generally occurring in the same sentence or paragraph. It is considered
a natural second/foreign language phenomenon and is not an indication of a language disorder (Scotton, 2004).
Language mixing is a phenomenon occurring when a bilingual talks to another bilingual in the same language and changes from one language to another automatically in the course of communication (Crystal, 1997). Various elements of language such as phonology, morphology, grammatical structure or lexical items are involved in code-mixing (Richard, Platt and Platt, 1992).
Appel and Muysken (1990) state that people may include a single word from another language in their communication because they may forget the exact word in the 7 first language or not yet learn the word in the base language of the sentence. Hoffman (1991) adds that code-mixing is used for more exact and clear-cut utterances in communication and is thought to facilitate understanding.
Wardhaugh (1992) says that a code-mixing phenomenon is mainly observed in both bilingual and multilingual situations. There is no rule in mixing a second language/foreign code in communication. It varies and depends on each situation. A person may use code x in one situation, code y in another, and mixed code in a third situation.