Email: alex @ phonetics.ucl.ac.uk

We report on methods of improving multilingual text alignments that have been produced in a simple dynamic-programming scheme, by automated detec-tion of possible misalignments. Details of methods involving cognates, specially-identified words, and propositional contents of sentences are given, together with notable features of their performance on parallel corpora in a number of different types of European languages. 1. Background In a previous paper (Campbell and Fang 1995) we described the background to our studies of automated alignment in multilingual corpora, and summarised our con-tribution to the European-Union-sponsored research project TRANSLEARN, as well as indicating some directions for future research to continue after the end of that pro-ject. The present paper reports our work since then. The European Union (EU) has 11 official languages, with the possibility that more may be added as countries with relatively advanced economies in easter