The Universal Grammar hypothesis proposes that humans possess innate biological endowments for language—a system of principles and parameters constraining possible grammars. The poverty of stimulus (children acquire complex syntax from limited data) supports innateness. UG explains why all languages, despite surface diversity, share deep structural similarities and why acquisition is remarkably uniform across human populations.
From your study of language acquisition, you know that children acquire the grammar of their native language rapidly, uniformly, and without explicit instruction — often before age 5, without formal teaching, from input that is imperfect and incomplete. From your study of language universals, you know that despite the enormous surface diversity of the world's languages, they share deep structural properties: all have nouns and predicates, all have ways of marking questions, all respect constraints on movement and reference. The Universal Grammar (UG) hypothesis, associated primarily with Noam Chomsky, connects these two observations. Both phenomena — rapid acquisition and cross-linguistic universals — are explained by the same cause: humans are biologically endowed with an innate language faculty that constrains the space of possible human grammars.
The central argument for UG is the poverty of stimulus (POS), sometimes called Plato's Problem. Children acquire grammatical knowledge that goes beyond what their input data could logically support. The classic example involves auxiliary fronting in English questions. Children correctly produce "Is the man who is tall happy?" rather than "*Is the man who tall is happy?" — selecting the right auxiliary to move. This complex structure-dependent rule is applied correctly and spontaneously, despite children rarely hearing such sentences in the input and never being taught the rule. The POS argument concludes that the relevant knowledge must be innate — part of the child's initial endowment, not extracted from the input. UG provides the structure that makes acquisition possible; the input triggers specific settings within it.
UG is not itself a grammar — it is a set of principles (universal constraints on possible grammatical operations, true of all languages) and parameters (binary or small-valued switches that vary across languages). For example, the head-directionality parameter determines whether a language is head-initial (English: verb before object — *eat sushi*) or head-final (Japanese: verb after object — *sushi eat*). When a child hears enough input to identify this setting, a large number of related grammatical properties fall into place simultaneously — a phenomenon called the parameter cascade. This explains why acquisition is so fast: the child is not inferring grammar from scratch but adjusting a small number of pre-specified switches, each of which sets dozens of related properties at once.
UG is supported by several converging lines of evidence. The critical period for full language acquisition (exposure before puberty is necessary for native-like grammar) suggests a biological substrate with maturational timing, analogous to other biological endowments. Specific language impairment (SLI), where children show targeted grammatical deficits despite normal general cognition, suggests grammar is at least partially dissociable from general intelligence. Studies of creolization show that when children acquire structurally impoverished pidgin languages, they spontaneously create creoles with richer, more UG-consistent grammar than the input — they appear to be filling in the grammar from innate resources. Against UG, usage-based and constructivist theories argue that general learning mechanisms — pattern recognition, statistical generalization — are sufficient to explain acquisition without innate grammatical structure. The debate remains active, and Chomsky's own minimalist program has progressively reduced the content of UG, asking how much of language's complexity is truly innate versus emergent from general computational properties shared with other cognitive systems.