Written Language & Literacy - Volume 8, Issue 1, 2005

Although verb spelling is rule-based, it is a relatively difficult aspect of Dutch orthography. Assink (1985) was the first to show that Dutch verb spelling is affected by two factors, namely, frequency of occurrence of the verb and the context in which the verb appears. This study is a developmental follow-up on his work, in which students of two age groups and three different levels of education, that is, students from primary school, secondary school, and university students, participated. It appeared that frequency and context already affect spelling behavior in the early stages of learning to spell. These aspects are still prominent determinants in the group of most experienced and good spellers, albeit the relative contribution of each of the determining factors changed over time. The interpretation of these findings was that verb-spelling performance in Dutch is not just the result of its inherent difficulty, it is also affected by the fact that it requires continuous active monitoring of the spelling process, while being involved in a main literacy activity.

This article demonstrates that many of the same concepts and tools developed for the sociolinguistic study of how language communities choose spoken languages can also be profitably applied to the study of how they choose scripts. These similarities include the choice of a language or script to either identify with or create distance from another group, borrowing elements from other languages and scripts, the death of languages and scripts, contact induced change in languages and scripts, and the identification of languages and scripts with gender.

In previous studies we have found that the writing of words (typing) is made up of a highly structured time course. Words are written using a course of accelerations and decelerations at certain points within the words. These points correlate highly with the syllabic word structure and also with the morphological structure. At the beginning of these subword units we find significantly higher latencies than within units. We therefore assume that written word production starts with a frame of the whole word, certain subword frames (according to the syllabic and morphological structure) and subsequent filling of the frames with segmental information. So the segmental information is not completely available at the beginning of a word or syllable but is delivered during motor execution. In the present study we try to find out if there are also subsyllabic units that are produced in a frame and content mode. As a candidate for such a unit we choose polygraphemes (e.g. in engl.in “think”, in germanin “schon”). If they are produced in this way we expect an increased delay at their onset and an accelaration within the unit. The results suggest that because of their grapho-phonological structure some types of complex graphemes are produced in a frame and content mode.