154 H.G.M. Williamson undervalued. For most purposes, therefore, the precise nature of the Hebrew text is of little interest to the vast majority of Bible readers. They are completely unaware of the problems with which scholars of the text struggle daily, many of which in fact get lost from view in the imprecise science of translation, and they would probably be initially quite surprised if they were told that in fact no translation is free from some form of necessary textual emendation or conjecture (3). 1. The Present Position During the first Christian millennium, the Jewish community, as would be expected, maintained its adherence to the Hebrew text and continued to develop it in some respects, not, of course, deliberately changing it by this time, but adding vocalization and other scribal marks and conforming variations between textual traditions to a single ideal type. The results of their careful work are known to us now from a large number of medieval manuscripts. The best complete one that still survives may celebrate its 1000th anniversary this very year (4), (3) E. TOV, “The Textual Basis of Modern Translations of the Hebrew Bible: The Argument Against Eclecticism”, Textus 20 (2000) 193-211, has argued that, because of the complexities of this history of the Biblical text, “the public deserves a diplomatic translation of a single text, be it MT, the LXX, or the Vulgate” (210), and he cites NJPST as an example to follow. However, this is not entirely practical, in that MT sometimes cannot make sense as it stands, so that effectively some form of indirect emendation has to be invoked in order to arrive at a readable translation, a procedure which invites some critical comments from Tov himself (n. 45). In my view, a well-grounded emendation, whether textually supported or conjectural, is often a more satisfactory, as well as more honest, way of treating such difficulties. As currently happens, footnotes can alert the reader to the fact that this has been done. Interestingly, with regard to Isa 7,11 discussed below, NJPST in fact adopts the emended vocalization without even a note. It is possible the translators would justify this by appeal to one of the devices proposed by a few previous scholars to explain the Masoretic pointing as being in conformity with the emended version, but this is desperate and a concession of the essential point; for details, see D. BARTHÉLEMY, Critique textuelle de l’Ancien Testament. Isaïe, Jérémie, Lamentations (OBO 50/2; Fribourg – Göttingen 1986) II, 46-47. Isa 14,4 is an even more striking example of the translation explicitly following a variant reading from MT. For additional comments in response to Tov, see B. ALBREKTSON, “Masoretic or Mixed: On Choosing a Textual Basis for a Translation of the Hebrew Bible”, Textus 23 (2007) 33-49. (4) The first colophon dates the manuscript according to five different chronologies, giving rise to four possible dates, namely 1008, 1009, 1010 and 1013. Based on the considerations that the mistakes are more likely due to the use of chronologies that were not commonly used at the time of the copying together with an agreement between two of those that survive this screening process, BEIT-