TY - JOUR
T1 - When knowing grammar depends on knowing vocabulary
T2 - Native-speaker grammaticality judgements of sentences with real and unreal words
AU - Barcroft, Joe
PY - 2007/3
Y1 - 2007/3
N2 - This study examined how the presence of real versus unreal words in sentences affected the ability of native English speakers to make accurate grammaticality judgements and forced-choice decisions for sentences with violations in the use of dative alternation and comparatives. Sentences with dative alternation violations contained polysyllabic verbs (John explained Mary the plan) that were real (e.g., explained), similar (e.g., explunned), and dissimilar (e.g., tidnopped) to real verbs. Sentences with comparative violations contained polysyllabic adjectives (*Robert is demandinger than Allen) that were real (e.g., demanding), similar (e.g., demunding), and dissimilar (e.g., natormunt) to real adjectives. Accuracy of grammaticality judgements was much lower for sentences with unreal words than real words. For sentences with comparatives, accuracy also was higher in sentences with similar words than with dissimilar words, demonstrating a graded effect for partial access. These findings provide support for theoretical accounts that associate knowledge of these structures with knowledge of real words and for instruction oriented toward the development of vocabulary knowledge.
AB - This study examined how the presence of real versus unreal words in sentences affected the ability of native English speakers to make accurate grammaticality judgements and forced-choice decisions for sentences with violations in the use of dative alternation and comparatives. Sentences with dative alternation violations contained polysyllabic verbs (John explained Mary the plan) that were real (e.g., explained), similar (e.g., explunned), and dissimilar (e.g., tidnopped) to real verbs. Sentences with comparative violations contained polysyllabic adjectives (*Robert is demandinger than Allen) that were real (e.g., demanding), similar (e.g., demunding), and dissimilar (e.g., natormunt) to real adjectives. Accuracy of grammaticality judgements was much lower for sentences with unreal words than real words. For sentences with comparatives, accuracy also was higher in sentences with similar words than with dissimilar words, demonstrating a graded effect for partial access. These findings provide support for theoretical accounts that associate knowledge of these structures with knowledge of real words and for instruction oriented toward the development of vocabulary knowledge.
UR - https://www.scopus.com/pages/publications/34247587240
U2 - 10.3138/R601-H212-5582-0737
DO - 10.3138/R601-H212-5582-0737
M3 - Article
AN - SCOPUS:34247587240
SN - 0008-4506
VL - 63
SP - 313
EP - 343
JO - Canadian Modern Language Review
JF - Canadian Modern Language Review
IS - 3
ER -