The aim of the article is to compare two 17th century translations of the New Testament: the translation by Heinrich Stahl (1638) and the translation by Johannes Gutslaff (1641–1656). Relying on the model of description presented by Torop (1989) and Torop and Osimo (2010), the translations are described by three aspects (1) linguistic, (2) functional, (3) that of the translator`s position within three different time perspectives: achronic, synchronic and diachronic. Within the achronic perspective, the two translations seem to be quite different in all three aspects. In the case of Stahl′s translation, the original text was Luther′s German translation and the target language was the North-Estonian vernacular. Different parts of the Bible had probably been translated by different persons and Stahl collected them, publishing them in a volume of his church manual. Gutslaff, on the other hand, translated the text himself using the original Greek version. His target language was the South-Estonian vernacular but his work was never published. Within the diachronic perspective, the differences disappear and the two translations seem to be quite similar. From today`s point of view both tr anslations are linguistically archaic and have no functional value except as object s of scientific research. Synchronic description is the most complicated one. The article tries to put to the test the possibilities of describing old translations from the point of view of their translation method. Very few examples of the Estonian language of the 17th century have been preserved, most are translations from German written down by German pastors. In addition to that, researchers know almost nothing about the Estonian language spoken by native speaker s of the time. As a re sult of this, it is hard to evaluate the linguistic status of many constructions and idioms used by the translators. This in turn makes it difficult to evaluate the stylistic intentions of the translator and renders the whole synchronic description problematic.