Phonemes, morphemes and literacy development : evidence from Greek
It has been proposed that literacy development follows a sequence from simple to complex rules: children acquire simple phonological rules before they learn more complex orthographic rules such as conditional rules or morphological rules. I hypothesise that Greek children start reading and spelling by using a simple phonological strategy and later develop more complex phonological and morphological strategies. The hypothesis that young children fail to use complex phonological and morphological rules, the processes involved in reading words with complex phonological rules, the predictors of children's use of morphological strategies in spelling and the relations between different instances of morphological spellings were investigated in six studies. In the first three studies the hypothesis that young children fail to use complex phonological strategies in reading and the processes involved in reading words which involve complex rules were examined. Children (6-8 years) were asked to read words and non-words (analogous and not-analogous to real words) either in isolation or in the context of a sentence, assigned to three categories in tenns of the rules involved in reading them. The children - especially the younger ones - performed better in words and non-words that involve constant relations between graphemes and phonemes than in words and non-words that involve variant relations between graphemes and phonemes. All the age groups performed better in the analogous nonwords that involve complex phonological rules than in the not-analogous non-words. Children and adults read words that involve variant but predictable spelling patterns either by establishing connections to whole words or segments of known words. Younger children benefited more from context than the older ones and the effect was bigger for more difficult words. In the fourth study the hypothesis that younger children fail to use morphological strategies in spelling was tested. Children (7-10 years) were given a task involving three instances of spelling of the final morpheme. Young children spelled the final morpheme using phonological strategies while older children used morphological ones. In the last two studies, children (7-10 years) were given oral measures of grammatical awareness, a standardised verbal ability test, measures of grammatical spelling knowledge and a measure of their ability to interpret novel words. Significant correlations between grammatical awareness, different instances of morphological spelling and children's ability to interpret novel words were found even after age and verbal ability were partialled out. I conclude that even in a language that is transparent (at least from spelling to phonology) a stage model of simple rules first, complex rules later still holds. In reading, complex phonological strategies must be acquired for the reading of words that involve conditional rules. Morphological spelling strategies are important for correct spelling in Greek (which is not transparent from phonology to spelling).