Universal Grammar

Universal Grammar (UG) is the theory, most closely associated with Noam Chomsky (1965 onward), that humans are born with an innate system of linguistic knowledge that constrains the possible forms human languages can take. UG is the content of the Language Acquisition Device — the abstract set of principles and parameters that every child brings to the task of language acquisition.

Core Components

Principles

Principles are universal constraints that hold across all human languages. For example:

• Structure dependence — grammatical operations depend on structural relations, not linear order (no human language forms questions by simply reversing word order)
• Binding principles — constraints on how pronouns and reflexives relate to their antecedents
• Subjacency — constraints on movement operations (e.g., what can be extracted from which syntactic positions)

Parameters

Parameters are points of variation that are set through exposure to a particular language. The classic example is the head-direction parameter: languages are either head-initial (English: verb before object) or head-final (Japanese: object before verb). The child does not learn this from scratch — the parameter is innate, and exposure to input triggers the correct setting.

The Poverty of the Stimulus Argument

The central argument for UG is the poverty of the stimulus (also known as Plato's problem): children acquire complex linguistic knowledge that goes far beyond what the input alone could provide. The input is:

• Finite — children hear a limited set of sentences
• Degenerate — input contains errors, false starts, and incomplete sentences
• Lacking negative evidence — children are rarely told which sentences are ungrammatical

Yet all children converge on the same grammar, including knowledge of structures they have never encountered. Chomsky argued that this gap between input and attainment can only be explained by innate linguistic knowledge.

UG and Second Language Acquisition

The central debate in generative SLA is whether adult L2 learners have access to UG. Three positions dominate:

Full Access

Adult learners retain full access to UG, including the ability to reset parameters to L2 values. L2 grammars are constrained by the same principles as L1 grammars. Difficulties arise from performance limitations, not competence deficits. This position aligns with the Identity Hypothesis.

Key proponents: Schwartz & Sprouse (1996) — the Full Transfer/Full Access model proposes that the L2 initial state is the L1 grammar (full transfer) but that UG remains available to restructure this grammar when L1 settings fail (full access).

Partial Access

Some UG principles remain active, but parameter resetting is constrained — particularly for parameters that have no reflex in the L1. This position accommodates evidence that adult learners can acquire some UG-governed properties but systematically fail on others.

Key proponents: Hawkins & Chan (1997) — the Failed Functional Features Hypothesis proposes that uninterpretable features not instantiated in the L1 become inaccessible after the critical period.

No Access

UG is unavailable to adult L2 learners. Whatever linguistic knowledge adults acquire in L2 is built through general cognitive mechanisms (analogy, memorisation, explicit rule learning) and L1 transfer. This is the position of the Fundamental Difference Hypothesis (Bley-Vroman, 1989).

Criticisms of UG

• Usage-Based Theory and Emergentism argue that linguistic knowledge can be explained without innate grammar — frequency, distributional learning, and general cognition suffice
• Connectionism has demonstrated that neural networks can acquire complex linguistic behaviour from input alone, undermining the poverty of the stimulus argument
• The theory has become increasingly abstract and difficult to test empirically
• UG has undergone major revisions (Standard Theory → Extended Standard Theory → Government and Binding → Minimalist Program), raising questions about what exactly is innate

Significance

Whatever one's theoretical position, UG has shaped the entire landscape of SLA research. Even researchers who reject UG define their theories partly in opposition to it. The question of whether linguistic knowledge is innate or emergent remains the deepest divide in the field.