Soru

Passage A
A cornerstone of modern generative linguistics is the poverty of the stimulus argument, which posits that the linguistic input accessible to young children is fundamentally insufficient to account for the rich, uniform grammatical competence they ultimately attain. In natural language acquisition, children rapidly and effortlessly master intricate structural constraints—such as the hierarchical rules governing auxiliary fronting in complex interrogative sentences—despite rarely encountering explicit negative evidence or corrective feedback from caregivers. Simple linear models of language, which attempt to predict sentence structure purely through sequential word adjacency, inevitably fail to explain why young learners consistently avoid formulating erroneous questions by displacing the first linear auxiliary in a sentence containing an embedded relative clause. Because primary linguistic data is inherently fragmented, noisy, and rife with syntactic ambiguities, generative theorists argue that syntax cannot be acquired through unconstrained inductive inference or general-purpose cognitive learning. Instead, human infants must be biologically equipped with domain-specific cognitive constraints—a universal innate template—that pre-structures possible syntactic forms and severely restricts the hypothesis space that a child must navigate during language development. Without such innate scaffolding, children would be mathematically incapable of selecting the correct target grammar from countless structurally possible alternatives.

Passage B
Recent advances in computational linguistics and corpus analysis have mounted a robust empirical challenge to the enduring assumption that ambient child-directed speech is too impoverished to support grammatical acquisition. When large-scale naturalistic language corpora are evaluated using high-dimensional probabilistic modeling, ambient speech exhibits dense networks of distributional cues, lexical co-occurrences, and acoustic boundaries that earlier theoretical analyses largely overlooked. Behavioral experiments demonstrate that infants possess exceptionally acute statistical learning mechanisms capable of tracking conditional probabilities between phonemes, identifying word boundaries, and abstracting syntactic frames from recurrent communicative episodes. Rather than requiring specialized, pre-wired linguistic modules dedicated exclusively to syntax, the mastery of hierarchical grammatical structures can be plausibly understood as an emergent outcome of general cognitive abilities—such as pattern recognition, category abstraction, and communicative intention reading—operating over a structured informational environment. Consequently, while children unquestionably acquire sophisticated language capabilities with astonishing efficiency and speed, this remarkable developmental milestone reflects the latent depth of linguistic input and the computational power of domain-general learning mechanisms, rendering the hypothesis of an innate syntactic blueprint superfluous.

Which of the following statements identifies a point of agreement between the author of Passage A and the author of Passage B?