I did fieldwork with Roma people in Romania. I first worked on the vocabulary of "cunning" (şmecherie) and "slyness" (ciorănie) used by Roma professional musicians (lăutari). These musicians use such words to denote general attitudes but also analytically, to point to specific musical operations. They treat tunes in a modular way, allowing for many recombinations, one of the incentives to recombine being to add more "cunning tricks" – if you’re a virtuoso — or remove some — if you can’t play them and need to build up something simpler. For this research I stayed extensively in the village Zece Prăjini, near the city of Iaşi, North-East of the country, generously hosted by the Panţiru family core of the Shukar Brass Band. Here is one of the tunes they play.

This is typical of the moldavian brass band style, which is quite different from the serbian/macedonian one : heavy harmonic section, heterophonic melodic section (no parallel thirds here, no solo either), and a relaxed meter (one of the bars is 6 beats long instead of 4). This kind of tune is made of melodic motives which are also found recombined in other tunes. Observe also that the second part here is built on the motive from the end of the first part, in a domino-like succession

Later I worked with Roma singers to make a Romani songbook for French scholars. This project was sponsored by the Cité de la Musique in Paris. It involved an approach which was new to me. First because of the pedagogical aim, second because it gave me the opportunity to dwell into Roma "community" singing. The lăutari, with whom I had worked earlier perform mainly for strangers — Romanians, Hungarians, other Roma, Westerners… They are professional musicians, whose resources depend on finding a substantial number of customers for their work. Napoleon Constantin on the other hand — main character of the songbook — sings mainly for his relatives and neighbors. His vocal style, his lyrics and his percussive accompaniment are signatures of a way of singing which is only found nowadays in some "traditionalist" Roma communities.

Technicity is here less important than fantasy, "daring" (tupeu) and improvisation on the spot. The above is a "classic" Roma dance tune, of the funny kind.

Well actually, this genre has been deemed "new" at least since 1989. Some say it dates back to the Ottoman times. For sure there is some tension between "modern" and "archaic" features, both in the music and in the lyrics. The tension also builds in exotic references to "the West" or "the East", in a pure orientalist style. Some Romanians seem to play their new freedoms in manele performances. Others despise them vehemently (and actually comment that the former Romanians are all "Gypsies").

A manea song sung by Cristi Nucă during a wedding in Iaşi. He is accompanied by a saxophone, an accordion and a keyboard. You can hear the "dedication" interactions between the singer and the audience. The manea is typically a live genre, best appreciated with real musicians playing for you. Should you use recordings, make sure you make them blast through your car speakers, your neigbour’s wall, etc., and that you have some friends around to enjoy collectively.

In 2009-10 I had the chance to do one full year of fieldwork in Bucharest on the manea, thanks to a grant from the New Europe College.

Lately I grew interested into one of the lăutari’s assumptions about music (lăutari is the word for Roma professional musicians): music, they say, is not so much about expressing one’s own feelings; it is rather about manipulating someone else’s. In other words, an interactional technique to "enchant" the listener. The term was proposed by Alfred Gell in quite a different context, but the lăutari would probably agree with it. I’m currently trying to take this idea one step further and see if it could ground a more encompassing paradigm.

For that, I build on ethnographies from different parts of the world, on experimental data from the cognitive sciences, and on various theoretical proposals by other anthropologists. In particular, I try to highlight the ontological assumptions underlying musical experience and to describe the distributions of agency between humans and sonic structures. The aim is to go beyond the basic idea that music is a way to "communicate": while this may seem an evidence in some societies, it is much too narrow to account for the broad range of things people do with music throughout the world. I'd rather view music as a social technique which enables people to interact in various ways with each other and with non-human entities. For this to happen, those people need to attribute music particular ontological properties which stand in contrast with those of other types of sounds. In such properties are rooted specific kinds of social agency, by which people achieve things musically which would otherwise be impossible. The definition of these properties are the core of my current research.

I suppose this sounds pretty abstract for now. Hopefully some empirical flesh will wrap it up soon…