ISF Research Grant 285/13 to Dorit Ravid 2013-2017
Input-output patterns in the acquisition of Hebrew root usage: A corpus-based psycholinguistic study
Most researchers consider the Semitic root to be Hebrew’s main lexical prime, relating root-based morphological families in the three content-word classes. Psycholinguistic research testifies to the central role of the root in Hebrew acquisition and processing, from children’s early ability to extract roots from familiar words to the critical significance of roots in spelling and reading Hebrew words in literate adults. Understanding how what roots are and how they are learned is gaining insight into vocabulary learning in Hebrew. Nonetheless, there is no sufficient information regarding the actual amount, types and distributions of roots in Hebrew, rendering our understanding of how roots are learned partial and incomplete. We have no idea how many roots there are in Hebrew, how many of them are regular and how many are irregular, what are the actual distributions of roots (types and tokens) in speech and writing, and how these numbers change across acquisition. Even worse, we do not exactly know what a “root” is – a consonantal skeleton as in אמר – האמיר? A meaningful morpheme as in חשב – החשיב – חישב – התחשב? Something in between as in מכר – התמכר? Answers to these questions are critical at they determine how many roots there are in different communicative contexts and how roots are learned.
These are huge questions requiring the analysis of large databases with different content word classes. Our current project focuses on verbs (as the most typical habitat for roots) input to children and in children’s linguistic output (as the core lexicon of Hebrew). During the academic year that is now drawing to completion we have focused on analyses of verb roots in three new corpora of linguistic input to children: Child directed speech in two mothers of toddlers aged 1;10-2;3, 100 storybooks targeting preschoolersaged 1-6 years, and 40 schooltexts for 2nd graders. We are looking for empirical evidence regarding type and token frequencies of input verbs, the distributions of regular and irregular structural root classes, affixation patterns across binyamin and semantic relations between verbs sharing the same root.
Look out for new findings that will be published on this page! The revealed structure of the early root category questions established views of root-based morphological families, offering a novel model of root learning in Hebrew.
See our new chapter in print pdf