Written Language & Literacy - Volume 10, Issue 2, 2007

Between the 15th and 17th centuries, German spelling underwent essential changes. Although most writers and typesetters were well acquainted with the high degree of formalization of Latin grammar and orthography, the changes were not brought about by principles planned ex ante or by application of spelling rules. Instead, these innovations spread through the lexicon, affecting the paradigms one by one in a continuous process of diffusion. The spelling changes occurred, as it were, behind the backs of the writers, who were apparently unwilling or unable to draw general rules from individual cases. But even without an existing norm-consciousness, innovations may in the end lead to rules and norms.

In the literature on punctuation we find a broad typological and historical distinction between prosodically and grammatically determined punctuation. The mainstream historical assumption is that the prosodic system changed into to a grammatical system in some languages. We will show that this view is confronted with serious empirical and conceptual problems. Our assumption is that the typological and historical variation at issue is motivated syntactically in all punctuation systems. The different punctuation systems are mainly distinguished by the comma, which is, therefore, the main topic of the present paper. The major use of the comma will be explained by four constraints, whose interaction may be congenially formulated in optimality-theoretic terms. The close relationship of the comma to prosody arises indirectly from the fact that syntactic structures are marked prosodically in many instances. The stylistic freedom of the comma, that is traditionally assumed for some languages and contexts of use, is a reflex of syntactic optionality.

This paper is concerned with the relation between our capacity for alphabetic reading and writing the sound forms of languages on the one hand, and the structure of speech and language on the other. It starts from two questions: (1) What structural properties of human languages enable us to read and write their sound forms with a handful of alphabetic symbols? (2) Why is learning this skill so difficult? Ad (1) it is argued here that the basis for reading and writing in an alphabet is the inherent segmentability of speech, stemming from the synchronization of articulatory gestures during speech production. This synchronization arises from inherent properties of both speech production and speech perception. Ad (2) it is suggested here that learning to read and write with alphabetic letters is so difficult, because in the mental structure of sound forms there are no pre-existing discrete phoneme-sized segments, at least not of a kind that language users are easily aware of. This makes analysis of sound forms into such phoneme-sized segments difficult, although such analysis is a prerequisite skill for alphabetic reading and writing. For easy learning, the relation between letters and speech segments should preferably be systematic and transparent.

The spelling process of Dutch second-grade students using the word processor was studied in three different feedback conditions. In the no-feedback condition, they merely had to type words without being told whether the spelling was correct or not. In one of the feedback conditions they were only told whether the word was spelled correctly, whereas in the other feedback condition they were given suggestions when words were spelled incorrectly. Students learned about the spelling of words without receiving feedback. Providing feedback, however, improved spelling performance substantially more than no-feedback. The nature of the feedback affected spelling performance of native and non-native Dutch words differently. Native Dutch words benefited more from suggestions than non-native words, and both word types benefited equally when no suggestions were provided. The quality of the suggestions provided by the spelling checker was substantially better in case of native Dutch words. Thus, Dutch second graders are capable of making efficient use of the spelling checker only in case of native Dutch words.

Dutch compounds with ‘e’ or ‘en’ as linking element between modifier and head were presented to mother tongue speakers of Afrikaans in an experimental setting that explored the possibility that these different spelling formats would suggest a singular or plural meaning of the modifier. The participants appeared to interpret ‘en’ in the linking element as an indication for singular, and ‘e’ as signifying plural. This outcome supported the findings in comparable studies on Dutch, which also revealed a tendency to understand the spelling of the linking schwa in relation to conventions for the spelling of the plural suffix. In Afrikaans the spelling of the plural forms is ‘e’, whereas in Dutch the spelling of plural forms is ‘en’. This explains why the results of the Dutch and Afrikaans experiments, while using the same materials, are each other’s mirror image.

The representation of word boundaries constitutes a central issue of many writing systems and orthographies, especially when two different components of a segmental phonographic and a word-based morpho-syntactical principle interact. This interaction will be studied in the context of literacy transfer from a L2 learnt at school to a vernacular and non-written L1. The central question of this paper will be how three different types of literate and metalinguistic knowledge intervene in this process. As a case study, the spontaneously written narratives of Moroccan school children who speak Tarifit Berber as their first language in Germany (with German as L2) and Morocco (with Modern Standard Arabic as L2) will be analysed. We assume that the writing of Berber in these two settings does not only differ on the level of the two different scripts, but also on the level of very divergent orthographic principles at the morpho-syntactical level. This paper presents material from two research projects on literacy acquisition of Moroccan students in Germany and Morocco, in which students from different grades were asked to produce oral narratives in Berber (and their second language, i.e. Arabic or German, respectively) and then to transcribe them according to their preferred script system. The similarities in the treatment of word boundaries cut across the overt differences in the choice of the script system: pupils in Germany, even if they use the Arabic script, transfer a word concept from German, while in Morocco, even in the case of Latin letters, a different model is dominant. As will be shown with respect to the problem of prepositional phrases in detail, the different models function like “matrix scripts”. They indicate at the same time, albeit in different degrees, a cultural accommodation to the dominant school or official language.

The distinction between deep and shallow orthographies is a central issue in studies of alphabetic writing. This paper aims to contribute to the debate on the relative merits of these by investigating how the coronal plosives [d] and [t] map onto the corresponding letters d and t. It turns out that an interesting asymmetry exists in word medial position in Dutch: both experienced users and children learning to write Dutch prefer ds over ts. Several mutually not always exclusive alternative explanations are provided. The fact that the d-preference is sensitive to phonological and morphological distinctions and may be influenced by speech rate, suggests that a deep orthographic representation might be the better option.