ABSTRACT
There are two main reasons for the long and virtually unchallenged reign of the normative approach in most thinking about translation since the Romans. The first is the use of translation in the context of language learning, as evoked above. The second and more important reason is the historical coincidence that the foundation text of the main religion of the West, the Bible, proved accessible to the great majority of the faithful, clergy and laity alike, only as a translation. This accident of history has tied translation and power together, in a union sometimes unholy and perceived as indissoluble for centuries. The progressive secularization of Europe did little to reverse the normative approach which came to rest on the institutionalized power of critics, philologists, and linguists. Only in the last three decades has the reign of the normative begun to crumble rapidly. It is now generally accepted that no general norms can be given for all translational activity, a realization linked to the realization that translation is an activity sui generis after all. As long as translation was considered mainly a secondary activity to be left to robot-like figures, there was little point in admitting that instructions might sometimes have to be diversified. After all, why confuse the machines?
