The study of the emergence of Islamic doctrine and practice, indeed of early Islamic history in general, is confronted by the problem of the sources. More than a century ago, Ignac Goldziher initiated this debate regarding the (in)authenticity of the sources, especially the Hadith literature. Goldziher’s critical remarks were point of departure for many researchers after him. Departing from the tacit assumption that the Hadith traditions are nonetheless potentially valuable historical sources even when inauthentic, several competing paradigms have, since Goldziher’s time, emerged concerning the dating of traditions; in their essence all stem from different evaluations of the distinction between oral and written history and the transition from the first to the latter. A case study on the number of daily obligatory Salat-prayers in Islam (a question not explicitly dealt with in the Qur'an) is the main topic of this presentation which will also afford new insights about the methodology of dating of Hadith. Starting off by isolating the relevant traditions, submitting them to the different existing dating methods - mostly based on analyses of the transmission chains (sing. isnad) – then critically comparing their outcomes and finally testing these results against data retrieved from other sources (historiography, exegesis, awa'il literature etc., as well as data on Jewish/Christian influences on Islamic practice), the analyses presented here on this particular cluster of traditions will provide us with an alternative dating. Moreover, the presentation provides insights on the merits of the various competing paradigms concerning the authenticity of the sources and it will offer a number of possible generalizations based on the empirical data used. In addition, a contribution will be made to our understanding of the early history and development of a cornerstone of Islamic liturgical practice, the Salat prayer.