Roberts, David, Keith Snider & Steven Walter (2016): Neither deep nor shallow: testing the optimal orthographic depth for the representation of tone in Kabiye (Togo). Language and Speech 59:1.113–138. (original) (raw)
Related papers
Some orthographies represent tone phonemically by means of diacritics; others favor zero marking. Neither solution is entirely satisfactory. The former leads to graphic overload; the latter to a profusion of homographs; both may reduce fluency. But there is a ‘third way’: to highlight the grammar rather than the tone system itself. To test this approach, we developed two experimental strategies for Kabiye: a grammar orthography and a tone orthography. Both are modifications of the standard orthography that does not mark tone. We tested these in a quantitative experiment involving literate L1 speakers that included dictation and spontaneous writing. Writers of the grammar orthography perform faster and more accurately than writers of the tone orthography, suggesting that they have an awareness of the morphological and syntactic structure of their language that may exceed their awareness of its phonology. We conclude that languages with grammatical tone might benefit from grammatical markers in the orthography.
Orthography and phonological depth.pdf
In Michael Cahill and Keren Rice (eds.). Developing Orthographies for Unwritten Languages, 27-48. Dallas: SIL International., 2014
Which level of phonological depth should be represented orthographically seems to be locked into phonological theories that predate the 1970’s. Typically, only two options receive serious consideration: the classical phoneme (shallow orthography) and the morphophoneme (deep orthography). Consistently representing either form is problematic, however, and the present work demonstrates why neither approach can be recommended as a general strategy. Stratal approaches to phonology, however, with claims that native speakers are more aware of the output of the lexical phonology than of any other phonological level, offer a worthy third alternative. Employing examples of morphophonemic alternations from a number of different languages, the present work demonstrates that regardless of whether the preferred orthographic representation is phonemic or morphophonemic, the level that works best from a practical viewpoint is consistently the output of the lexical phonology.
The standard orthography of Kabiye (Togo) does not mark tone. In such a context, how can a researcher adequately assess the degree of ambiguity in the written language and make a valid contribution to the debate about how tone might be incorporated in the second generation of language development? This article approaches that question, not from the perspective of phonological analysis which has tended to dominate the literature, but from the point of view of the linguistics of writing. Applying Catach's (1984) model of lexical ambiguity for Kabiye, it advocates the development of a homograph corpus in which words, roots and affixes are included or excluded on the basis of semantic, morphological and dialectal criteria. A homographic prefix with pronominal, negative and immediative interpretations illustrates how the corpus is then applied to a frequency and distribution analysis of ambiguity in natural written contexts, and an analysis of oral reading errors in the classroom. A dictation task reveals that subjects who were taught a segmental modification of the negative prefix write with greater accuracy than subjects who were taught to add tone diacritics.
Orthography and Tone: Tone system typology and its implications for orthography development
2011
It is a feature not found in French, English, Spanish, Portuguese – the present-day official and former colonial languages of many African countries, It is certainly not an easy matter to analyse the tone system of a language in preparation for developing a tone orthography, Most people have a natural supposition that an orthography which looks more difficult – i.e. with many diacritic signs – is automatically harder to read.
Decreasing dependence on orthography in phonological development
Language Acquisition and Language Disorders, 2016
Despite the general transparency of standard Turkish orthography, it fails to distinguish the (not fully predictable) contrast between coronal vs. dorsal laterals following back vowels in certain loanwords: the laterals in /ko / and /rol/ are both represented as . This contrast results in non-canonical vowel harmony, where the backness of a suffix vowel is determined by the lateral, rather than by the preceding vowel (e.g. /ko -a/ , but /rol-e/ ). While early English-Turkish learners performed at a significantly higher level of accuracy on selecting the target suffix vowel in these contexts with auditoryonly presentation of the stimulus than with auditory and written presentation, intermediate and advanced learners come to rely more on auditory stimuli and less on orthography.
Many studies have investigated the production and perception of second language learners. However, very few of them have discussed the effect of orthography. Although Mandarin and Cantonese share the same orthographic system, the phonological systems of the two languages are quite different. This preliminary study investigates the production and perception of Mandarin tones by Cantonese learners, and compares their error patterns in two conditions: when the subjects were presented with 1) stimuli written in Mandarin Pinyin (a transparent orthography), 2) stimuli written in Chinese characters (an opaque orthography). The result shows that orthography has different effects in tone production and perception: the Pinyin system facilitated tone production only, while the subjects performed significantly better with Chinese characters in perception. Possible accounts for the observed differences are discussed.
The shorter a word, the more likely it is to be lexically ambiguous. In the standard orthography of Kabiye, a tone language of Togo, numerous monosyllabic heterophonic homographs (tonal minimal pairs) and homophonic homographs occur in the imperative and six associated conjugations. This paper presents the complete catalogue of these verbs, and then examines them in natural contexts. It goes on to propose a morphographic spelling in which elided root-final labial consonants are written as superscript silent letters to help the reader identify the lexeme. This spelling is tested against a tonographic alternative in an oral reading experiment. The results show that those who learned the morphographic spelling gained more in reading accuracy from the addition of superscript silent letters than those who learned the tonographic spelling did from the addition of diacritics.
ABSTRACT: There is an ongoing debate about how tone should be represented in the emerging orthographies of African languages. One of the most significant strands in the debate is a small but growing body of literature describing formal experiments which test the different options. In this article, I present an overview of the existing repertoire which covers ten experiments and three decades. I adopt a comparative approach, examining all the experiments in parallel. I focus in turn on aims, design, sample profile, sample size, experience, training, test materials, tasks, scoring, results and interpretation. In conclusion, I offer some practical advice for future experimenters. I also attempt to identify whether any consensus is emerging about the profile of an optimal tone orthography. RESUMÉ : Le débat sur la représentation du ton dans les orthographes émergeantes des langues africaines a déjà fait couler beaucoup d'encre. L'un des plus importants volets dans ce débat est une littérature, modeste mais croissante, décrivant des expériences formelles entreprises pour tester les différentes options. Nous présentons un survol de ce répertoire qui couvre dix expériences et trois décennies. Dans cet article, nous adoptons une approche comparative, en examinant l'ensemble des expériences en parallèle. Nous nous focalisons sur les objectifs, la conception, le profil de l'échantillon, la taille de l'échantillon, le degré d'expérience, le niveau de formation, les matériaux expérimentaux, les tâches, le scoring, les résultats et l'interprétation. En conclusion, nous offrons quelques conseils pratiques pour de futurs expérimentateurs. Nous tentons également de dégager un consensus en ce qui concerne le profil d'une graphie tonale optimale.
In many tone languages, decision makers have opted for zero representation of tone. This generates homographic tonal minimal pairs that may trigger oral reading miscues. But it would be wrong to attribute the source of all miscues just to tonal minimal pairs; there may be other aspects of the orthography’s profile that inhibit word recognition. In the standard orthography of the Kabiye verb phrase, subject pronouns and modal morphemes are written attached to the root. The unforeseen secondary effect of this decision is that the identity of the root is often masked because the morpheme boundary is not explicit. A homograph analysis reveals that morphemic mismatches generate numerous tonal minimal pairs. But a miscue analysis reveals that the problem extends beyond these to any verb phrase that contains infrequent, alternating or multiple prefixes, whether or not they are homographs. It follows that to disambiguate just tonal minimal pairs would only solve half the problem. A modification that highlights the morpheme boundary would directly address the real source of readers’ difficulties. The results of a dictation task in a classroom experiment indicate that root initial capital letters would be a promising solution.