'n Akoestiese en artikulatoriese analise van die /r/ in Malmesbury-Afrikaans

Abstract

In hierdie studie word daar spesifiek gefokus op die /r/-klank in Afrikaans van die Malmesbury-streek. Daar word tussen slegs twee variasies onderskei, naamlik ʼn alveolêre /r/ en ʼn bry-/r/. ʼn Bry-/r/ sluit enige variasie van /r/ in wat nie ʼn alveolêre tril is nie. Die verskil tussen die tweede en derde formante (F2 en F3) word wyd gerapporteer in studies van die /r/-foneem. Omdat die tong na die agterkant van die mondholte toe skuif en die agterkant van die tong oplig wanneer ʼn bry-/r/ gevorm word, behoort ʼn bry-/r/ ʼn laer F2 en hoër F3 waarde te hê as wat ʼn alveolêre /r/ sal hê. Uit die metings wat in hierdie studie gedoen is, blyk dit egter dat die berekening van die verskil tussen F2 en F3 nie die beste metode vir hierdie studie is nie. Vervolgens is die duur van die /r/-foneme gemeet. Hierdie analises toon groot prakties beduidende verskille tussen die lengte van die alveolêre /r/ en bry-/r/ en kan die meting van die duur word as ʼn goeie metode vir analises van /r/-klanke. Die laaste akoestiese analise wat in hierdie studie gedoen is, is die meting van die vier spektrale momente nl. gravitasie-middelpunt (“centre of gravity”), standaardafwyking (“standard deviation”), kurtosis en skeefheid (“skewness”). Hierdie spektrale momente word gewoonlik gebruik in die analise van frikatiewe en is tot dusver nog nie regtig in studies van /r/ gebruik nie, maar aangesien /r/-klanke sommige belangrike akoestiese eienskappe met frikatiewe deel, was dit moontlik dat dit ook hier van toepassing kan wees. Uit die analises het dit geblyk die spektrale momente ʼn besonder vrugbare analiseermetode is. Om die analise van die /r/-foneem verder artikulatories te verryk word ʼn gevallestudie ten slotte op ʼn enkele spreker gedoen met behulp van ʼn ultraklankbeeldaftaster om die posisie van die tong tydens die artikulasie van ʼn bry-/r/ aan te toon en te bespreek. The /r/ sound of Afrikaans in the Malmesbury region is the main focus of this study. A distinction is made between only two variations of /r/, namely an alveolar /r/ and a grout /r/. A grout /r/ includes any variation of /r/ that is not an alveolar trill. The difference between the second and third formants (F2 and F3) are widely reported in studies of the /r/ phoneme. Because the tongue moves to the top and back of the oral cavity when a grout /r/ is formed, a grout /r/ should have a lower F2 and higher F3 value as an alveolar /r/ have. From the measurements taken in this study, it appears that the difference between F2 and F3 is not the best indicator of the alleged difference between the two sounds. Subsequently the duration of the /r/ phoneme was measured. These analyses show great practical significant differences between the length of the alveolar /r/ and grout /r/ and can be regarded as a good method for the analysis of /r/ sounds. The last acoustic analysis that was done in this study, was the measurement of the four spectral moments i.e. centre of gravity, standard deviation, kurtosis and skewness. These spectral moments are usually used in the analysis of fricatives and thus far not really in studies of /r/. Because /r/ sounds share some important acoustic properties with fricatives, it was possible for it to be applicable in this study. From the analysis, it appeared that the spectral moments are a particularly good method of analysis. To further enrich the analysis of the /r/ phoneme acoustically a case study is done on a single speaker using a ultrasound scanner to determine and discuss the position of the tongue during the articulation of a grout /r/.