Authorship attribution is an important problem in text classification, with many applications and a substantial body of research activity. Among the research findings are that many different methods will work, including a number of methods that are superficially language‐independent (such as an analysis of the most common “words” or “character n‐grams” in a document). Since all languages have words (and all written languages have characters), this method could (in theory) work on any language. However, it is not clear that the methods that work best on, for example English, would also work best on other languages. It is not even clear that the same level of performance is achievable in different languages, even under identical conditions. Unfortunately, it is very difficult to achieve “identical conditions” in practice. A new corpus, developed by George Mikros, provides very tight controls not only for author but also for topic, thus enabling a direct comparison of performance levels between the two languages Greek and English. We compare a number of different methods head‐to‐head on this corpus, and show that, overall, performance on English is higher than performance on Greek, often highly significantly so.

The present paper is a preliminary attempt to codify assimilation and deletion phenomena in Modern Greek involving the word-final /n/ of certain high-frequency function words and the word-initial voiceless stop (/p/, /t/ or /k/) of the following word. In natural speech, every relevant phoneme combination affected by these post-lexical phonological processes will generate one phonetic realisation from a set of legitimate variants. The variants’ distribution is not random but subject to both linguistic and extralinguistic conditions. This work also explores possible ways to exploit the findings of recent sociolinguistic research on Modern Greek with a view to accommodating the observed variation and the rules codifying it within the framework of text-tospeech (TTS) synthesis. Such a development is expected to improve synthetic speech output, both in terms of naturalness and intelligibility.