THE PROBLEM OF MIDONS REVISITED
The use of the term midons to designate the beloved is a well-known feature of troubadour poetry. It raises several questions, grammatical as well as semantic, which have been the subject of recent inquiries by scholars such as Mary Hackett, Glynnis Cropp, Ruth Harvey, William Paden, and Aurelio Roncaglia. While each of these studies makes positive contributions to our understanding of the phenomenon, their analyses diverge on several points, and so the problem remains intact. This essay will undertake to summarize the present state of the question and to make some suggestions for resolving the controversy.
The project has been aided by a major recent development in troubadour studies, the publication of the Concordance of Medieval Occitan l . Thanks to this important new tool, it has been possible to retrieve all of the 648 attestations of midons and related terms found in the troubadour lyric, thus nearly doubling the corpus examined by the most extensive previous study 2. With the recent appearance of COM2, it has also been possible to
1. Peter T. Rickets et al, Concordance of Medieval Occitan I Concordance de l'occitan medieval (COM), Turnhout, 2001 (COMÍ), 2005 (COM2). Written before the publication of COM 2, this article is based primarily on data obtained from COM 1, which was confined to the troubadour lyric. Devoted to the non-lyric poetry, COM2 yields another 142 attestations of midons and another 146 of sidons, which will be analyzed briefly, especially in the notes. The COM team plans a third "tranche" devoted to prose texts, with a possible forth phase to include the variants (COM 1, "Preface", p. 4-5 ; COM2, "Preface", p. 27-28). The COM is based on existing editions, which limits to some extent the reliability of the statistics that can be generated from it for questions such as word separation (mi dons vs. midons) and final -s (midon vs. midons), regarding which a certain amount of editorial normalization may have occurred (see below, n. 1 3). Completely accurate statistics for these matters would require recourse to the manuscripts, including the variants, a project far beyond the scope of the present study. Nevertheless, the overwhelming trends indicated by the COM-generated statistics can be regarded as reliable, even if the statistics themselves may be somewhat less so.
2. William D. Paden Jr., « The Etymology of Midons », in Studies in Honor of
