'n Ondersoek na Afrikaanse vertaalkenmerke in 'n korpus koerantberigte

Simple item page
dc.contributor.advisorFeinauer, A. E.en_ZA
dc.contributor.authorRoos, Deirdreen_ZA
dc.contributor.otherStellenbosch University. Faculty of Arts and Social Sciences. Dept. of Afrikaans and Dutch.
dc.date.accessioned2009-02-10T13:56:39Zen_ZA
dc.date.accessioned2010-08-13T13:11:33Z
dc.date.available2009-02-10T13:56:39Zen_ZA
dc.date.available2010-08-13T13:11:33Z
dc.date.issued2009-03en_ZA
dc.descriptionThesis (MPhil (Afrikaans and Dutch))--Stellenbosch University, 2009.en_ZA
dc.description.abstractIn this translation corpus study a monolingual comparable corpus of translated and nontranslated Afrikaans newspaper articles from Die Burger are compared with the use of WordSmith Tools 4. WordSmith Tools generates statistics, word lists and concordances that can be sorted in a variety of ways. The data generated for the translated and nontranslated subcorpora are then compared. This study follows on a translation corpus study of Afrikaans rugby articles by RG Bam (2005), which found that translated language differs from nontranslated language and that it also differs from the results for English in a similar study. The difference between the findings for English and Afrikaans is attributed to the commonality of the rugby articles. For this study the domains are extended to include topical articles, arts and entertainment, business news, foreign news and sport (rugby, athletics, soccer, cricket, bicycling, hockey and gholf). With the extended domains, my results are similar to the previous Afrikaans study regarding type-token ratio, average word length and lexical density, but not with regard to average sentence length and convergence. My finding on sentence length agrees with the finding for English newspaper articles. However, it is clear that Afrikaans translated articles differ from Afrikaans nontranslated articles and that Afrikaans differ from the way in which English translated articles differ from English nontranslated articles. A further extension on Bam's study is the use of an automatic Afrikaans part-ofspeech tagger that was developed by CTeXT in 2005. The tagged data was applied with good results to the calculation of lexical density and in determining the number of pronouns in the distinct subcorpora. Because corpus translation studies is a relatively young field, the methodology suggested by Laviosa-Braithwaite (1995) for corpus studies in English is tested to see whether it is applicable to Afrikaans. The methodology is in the form of hypotheses. Certain aspects are investigated easily by means of WordSmith Tools, but other aspects, such as die occurrence of superordinates, is not so readily applicable to the corpus methodology.en_ZA
dc.format.extent130 p.
dc.identifier.urihttp://hdl.handle.net/10019.1/4032
dc.language.isoafen_ZA
dc.publisherStellenbosch : Stellenbosch University
dc.rights.holderStellenbosch University
dc.subjectDissertations -- Afrikaans languageen
dc.subjectTheses -- Afrikaans languageen
dc.subject.lcshTranslating and interpretingen_ZA
dc.subject.lcshCorpora (Linguistics)en_ZA
dc.title'n Ondersoek na Afrikaanse vertaalkenmerke in 'n korpus koerantberigteaf
dc.typeThesisen_ZA
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
roos_ondersoek_2009.pdf
Size:
1.41 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Masters Degrees (Afrikaans and Dutch)