Multilingualism — the use of two or more languages by an individual or community — is the global norm rather than the exception. Bilinguals are not two monolinguals in one person; their languages interact in complex ways, with different levels of proficiency across domains. Code-switching, the alternation between languages within a conversation or even a single sentence, follows grammatical constraints and serves social functions such as marking identity, signaling solidarity, or managing topic shifts. Language attrition describes the gradual erosion of a language due to reduced use, demonstrating that linguistic competence is dynamically maintained rather than permanently fixed.
Record or observe bilingual conversations and identify points of code-switching, then analyze what social or grammatical factors motivate each switch. Study Myers-Scotton's Matrix Language Frame model to understand the structural constraints on intrasentential switching. Compare balanced vs. dominant bilinguals to see how frequency of use shapes proficiency across the two systems.
Your introduction to sociolinguistics established that language use is socially situated — speakers vary their language systematically according to who they're talking to, what register the situation calls for, and what social identity they wish to project. Multilingualism extends this insight into a new dimension: when a speaker controls two or more languages, each language becomes a resource in its own right, and the choice of *which language* to use carries social meaning just as dialect choice does within a single language. The global reality is that the monolingual speaker is the demographic exception, not the norm. Most of the world's communicators navigate multiple languages daily.
Code-switching — alternating between languages within a conversation or even within a single sentence — is the most visible manifestation of multilingual competence, and it is the feature most misunderstood by outsiders. The critical insight from your linguistics training is that code-switching is not random or sloppy. It obeys grammatical constraints: switches tend to occur at major constituent boundaries and preserve the structural integrity of both languages involved. Myers-Scotton's Matrix Language Frame model formalizes this: one language provides the grammatical skeleton (the matrix language) while the other contributes embedded lexical items. The matrix language's morphology and phrase structure rules take precedence. A switch that would violate either language's grammar tends not to occur naturally among fluent bilinguals.
Beyond grammar, code-switching serves social and pragmatic functions. Speakers switch to signal in-group identity, to soften or sharpen a point, to quote someone in their original language, or to handle a domain — technical vocabulary, emotional topics, humor — that one language handles better than the other in their personal history. These are communicative resources, not failures of control. From your study of second language acquisition, you know that high proficiency in a second language requires extensive practice and exposure; the ability to switch fluently and appropriately requires high competence in *both*, making bilingual code-switching one of the most sophisticated linguistic behaviors humans perform.
Language attrition reveals something important about the dynamic nature of linguistic competence: it is maintained through use and can degrade when a language is no longer regularly engaged. L1 attrition in immigrants and L2 attrition after formal study ends follow similar patterns — phonology and low-frequency vocabulary are most vulnerable, while core grammar is more resistant. Crucially, attrited features often re-emerge with renewed exposure, suggesting that the underlying representations persist in a weakened or suppressed state rather than being deleted. This connects back to what you know about the procedural/declarative distinction in SLA: explicit knowledge may fade faster than implicit procedural competence. The practical implication is that "use it or lose it" describes language maintenance accurately, but "reactivation" describes recovery — attrition is gradient and reversible, not a one-way door.