that this ṭaraba was the root of the European word troubadour.
I was surprised both by the facility of the pronouncement and by its apparent status of established fact in the world of oriental studies, since in Romance philology nothing could be more remote than such certainty about the origins of the word troubadour. I knew even then that among Romance scholars it was a cause célèbre, its origins unknown and disputed, a textbook etymological riddle still assigned in Romance philology courses. I also knew that any intimations or suspicions that it was Arabic in derivation must be largely subterranean, not part of what was presented in the standard course. I thus demurred, saying only that it did not seem to me that this was the accepted etymon. The professor then suggested I might look it up and deliver confirmation or denial. I did indeed look it up, and I subsequently spent years sorting out the pieces of the puzzle, uncovering the tracks of an almost obsessive etymological search that had engaged countless linguistic and literary historians before me.
Few other etymologies, in fact, have provoked and sustained the interest of Romance scholars as that of trobar and troubadour, both because it has seemed to be introuvable (as Spitzer would pun when he added his name to the list of those succumbing to the lure of the mystery) and because it is emblematic and symbolic of an even greater and more obsessive mystery of origins, that of the troubadours’ poetry itself. The sign may be arbitrary, but the ties it has to its object are formidable. Medievalists of all stripes have never succeeded in dismantling or avoiding the maze of disputes over the origins of troubadour poetry and “courtly love,” and philologists have been hardly more successful in their efforts to establish the roots of trobar, or even its meaning. Before the divorce of historical linguistics from literary studies, the same individuals would address the origins of the two phenomena, the word and the poetry, in the same breath; after it, the phenomena might be addresed separately but hardly less vigorously.
Thus, as a relative innocent I found myself in the middle of what are among the most hotly debated topics in Romance studies, and I faced a trajectory of scholarship I could not explain in rational terms, for in looking into the etymology of trobar I had discovered what other philologists knew so well, that this bête noire and touchstone of Romance philology had produced staggering amounts of scholarship and had been debated from virtually every possible viewpoint. Indeed, from Diez’s and Schuchardt’s somewhat fanciful (but no less vigorously defended) notion that it came from tŭrbāre, as in tŭrbāre aquam, “to disturb the water,” to the Neogrammarian plea for a reconstructed (and unattested) tropāre, few stones had been left unturned, few arguments left unargued. But “virtually” and “few” turn out to be key qualifiers. One argument—that it came from an Arabic word—was not only not favorably received, but worse, it was not even deemed worthy of heated and acrimonious discussion. The Arabic etymon was apparently destined never to figure as one of the OED’s list of possible solutions of an unresolved etymological mystery.
I did ultimately find the source for the Arabic etymon, the view accepted as patent truth by many Arabists. It turned out that in 1928, when this etymology was very much a riveting feature of the intellectual preoccupations of Romance scholars and when there seemed to be an unresolvable impasse among the different proposals that had been circulating since Diez’s time, the eminent Spanish Arabist Julián Ribera had suggested that trobar may have come from the Arabic ṭaraba. It was a simple proposal, based on the hypothesis that the interaction of Romance and Arabic cultures in northern Spain and Provence had been substantial, especially in the musical and musical-poetic spheres. The evidence for his view existed in numerous historical sources and, more suggestively, in other related etyma, such as those pertaining to musical instruments. Aware of the linguistic difficulties of the solutions debated among Romance scholars, Ribera also pointed out that the proposed Arabic derivation was, relatively speaking, almost completely problem-free. It presented neither the semantic nor the phonetic difficulties of tŭrbāre and tropāre. Ṭaraba meant “to sing” and sing poetry; ṭarab meant “song,” and in the spoken Arabic of the Iberian peninsula it would have come to be pronounced trob; the formation of the Romance verb through addition of the -ar suffix would have been standard. He was right on all counts, and his case was effortlessly documentable.2
After reading through the morass of disputes, lacerating contradictions, and fanciful speculations that characterized Romanists’ arguments over the word’s origins from halfway into the nineteenth century until past the quarter mark of the twentieth, and then reading Ribera’s proposal, it occurred to me that the latter might well have been received as a welcome newcomer in the sweepstakes, a fruitful new path to explore, something new to fight about. As a novice and a naive student of the vicissitudes of Romance scholarship, it even occurred to me that philologists might have been enthusiastic about finding an unexpected solution to such a tenacious problem. But what I discovered in reading post-Ribera notes and articles on the subject, was that nothing could have been further from the truth. In fact, if today’s philology student does not encounter this proposal under circumstances rather unusual for a Romanist (such as being in an Arabic class) it is still quite possible to remain completely ignorant of the fact that the Arabic etymon was ever proposed.
Ribera’s etymon was never considered by Romanists. There is not a single etymological dictionary of English or a Romance language that gives the Arabic etymon as even a possibility, although the question is usually noted as “unresolved.” While Romance scholars were aware of this new proposal, their attitude and the seriousness with which they dealt with it were neatly summarized by Jeanroy in the single sentence I have used above as an epigraph: It is an argument not worth arguing, because it could not convince anyone. Jeanroy’s position is far from the exception; it is the norm.
Why had a plausible, straightforward, and not overtly irrational etymon for a highly problematic word in Romance been utterly rejected, cast into the oblivion of half-forgotten curiosities? Was it merely because it was Arabic? And why, when I turned to the parallel discussion of the origins of the poetry of trobar, did I encounter an analogous situation, one in which belief in the “Arabist theory” (a nebulous term that covers any number of different and often conflicting theories) was considered at best idiosyncratic?3
Whatever attitudes lay behind such wholesale rejection suddenly seemed to me considerably more intriguing than the original, clearly illusory, possibility of finding or establishing the truth about the origins of the word itself, let alone the poetry. From my perspective, the greater mystery lay in the configuration of a discipline within which the possibility that the answer to the riddle was an Arabic one produced either highly negative reactions or no reactions at all. The very possibility of an Arabic solution was shunned as taboo. For me, the question became not whether this word or that image came from the language, poetry, or philosophy of al-Andalus, as the Arabic-speaking Europeans called their Iberian homeland, but why discussions of such possibilities had such a different cast from others that concerned the medieval period and its cultural milieu—assuming that such discussions took place at all. This book is the result of my exploration of that question.
The number of books and articles on some aspect or other of the Arabist question is much greater than most Romance scholars might assume. It is a subject that, perhaps paradoxically, has engendered many studies over many years. The earliest were written in the eighteenth century. From that vantage point, in fact, the earliest theory of the origins of Romance vernacular love poetry is Arabist. But as I began to read in this extensive body of scholarship, I noticed that most works have several things in common. The first is that almost all begin or end with the observation that acceptance of the apparent facts or tendered theories they present is difficult, that Westerners—Europeans—have great difficulty in considering the possibility that they are in some way seriously indebted to the Arab world, or that the Arabs were central to the making of medieval Europe. A second shared feature of such studies is that they do not explore such observations further and go forward with the explicit or implicit assumption that whatever the nature of those blinders or the cause of the neurosis, the weight of the truth, reasonable argument, or unarguable
