While literary representations of “çingene” by non-Roman re-embedded iconic caricatures through words, community-based control over images shifted with the advent and commercial success of a new musical genre, Roman dance music (Roman oyun havası). In fact, it is highly significant that the mass introduction of a neologistic cover term for these communities, “Roman,” seems to be directly associated with the appearance of this musical genre beginning in the mid-1960s. The musical category of “Roman oyun havası” as a 9/8 meter dance melody associated with Roman communities emerged through the marketing strategies of the 1960s recording industry agents. Fueled by growing markets for locally-produced 45 rpm recordings, newly-formed recording studios sought new local repertoire that they could record and then market. Roman dance music (Roman oyun havası) was

A second major shift in Roman musical representation began in the mid 1970s, when Roman wedding musicians from Edirne began develop a repertoire of locally-composed songs for mostly 9/8 meter melodies and dance songs for wedding events, and such songs were picked up for dissemination by record producers. The northwestern Turkish border town of Edirne has long been the home for a rich cosmopolitan cultural mix, as it is both situated at a crucial trade junction in the fertile Thracian region between Turkey, Bulgaria and Greece, and has a prestigious history as a former seat of the Ottoman dynasty. The town has some ten or more Roman neighborhoods, and a large population of Roman work as professional musicians serving the populations of the town and surrounding villages. [24] The media dissemination of a marked Roman repertoire was largely driven by the needs of small-scale record companies to find larger markets in which to distribute recordings, thus drove producers to build customer bases in areas outside of Istanbul. From evidence that I collected from interviews, there were three contributory practices among Edirne Roman that contributed to creating this newly-fashioned genre: 1) 9/8 meter instrumental tunes and songs which were part of the varied repertoire of Roman communities as discussed above; 2) adoption of kanto stage songs which reference “çingene” life in their song texts; 3) the personal innovations of a few wedding musicians from Edirne who developed topical songs for their Roman audiences during weddings. 

The confluence of these practices with individual local innovations led to local Roman embracing 9/8 as a distinctly Roman genre. [25] Drawing on distinctly local references to places, commentary on local events and people, use of Romanes and in-group slang, these songs created new outlets for communal expression, which were sought for exploitation in the studios to support a rapidly growing recording industry.

A short biography of a few of the early artists who spurred this process illustrate how in-community references became embedded in local songs that in turn came to signify greater Roman identity. Roman musicians in Edirne credit violinist Hacı Üründülcü (Kemâni Hacı Üründülcü, or “Acı”) as the first to create and perform songs with Roman oyun havası melodies. Kemâni Hacı Üründülcü’s musicianship was prized among Roman for his ability to make up topical song texts about his own experiences and that of his Roman audiences. While locally popular as a wedding musician, he did not have access to the recording studios, and passed away from tuberculosis before the expansion of the recording industry.

However, the repertoire of dance songs created by Kemâni Hacı became the staple for a well-recorded duo, nephew Kadir Üründülcü and clarinetist Deli (“Crazy”) Selim Kızılcıklar, who carried on the new tradition of setting new and re-fashioned instrumental melodies to topical song texts. One such example from Hacı Üründülcü was recorded by Kadir Üründülcü and clarinetist Deli (“Crazy”) Selim Kızılcıklar, and embeds distinctly local references to places, people, events, and personalities.

In this portrayal, the gambler is not “every woman’s husband” as in Ciguli’s “Binnaz”, but a specific person (the father of the singer) who has a gambling problem. Also note the references to particular neighborhoods and their characteristics, such as attractive girls of Edirne’s “Sögütsu” --a Roman neighborhood. Also interesting are references to enlarged circuits of contact, including places in ∫stanbul such as the neighborhood of Üsküdar, a district on the Asian side of İstanbul; Unkapanı, the Istanbul district that houses most of the recording companies; and Balat, a historical neighborhood near Unkapanı. These references may also point to distinct Roman experiences, as Üsküdar contains the Roman neighborhood of Selamsız, and Balat had been a predominantly Roman neighborhood since the 15^th century. In addition, there are many Edirne Roman families that have relatives that have settled in Selamsız and Balat, and/or originally came to Edirne from these areas.

The circulation of locally generated representations continued to be supported by a recording industry that had gained its impetus in the 1960s by serving local communities. In events observed during fieldwork in the 1990s and reconstructed from oral histories, once a wedding ensemble had been contracted to record, producers remained somewhat distant from repertoire selection.  As a result, this body of commercial recordings has retained a somewhat closer connection to material performed in local communities. [26] Thus the cultural expressions in such songs can be understood to mediate localized ideals of self-presentation into small-scale commodity exchange networks. Such mediations were also the result of “best guesses” on the part of producers and musicians as to potential commercial success. As a result, many songs also are imprinted with mixed references to specifically local people, places, events and concerns alongside repetitions of stereotyped images from the non-Roman “çingene” repertoire.

As such paths for new representations became enlarged through recording circuits, musicians in the latter 1990s saw this as a double-edged sword. On the one hand, ability to perform such repertoire guaranteed entry into the recording studios. On the other, association with the 9/8 meter wedding-style Roman dance music had the potential to consign artists to a musical ghetto, labeled as a “wedding musician,” (düğün çalgacı) associated with particular regional Roman communities, and a chronically marginalized and exploited, albeit guaranteed, market. Thus a musical term under the potentially liberating rubric of “Roman” began to be perceived as a condensation back into a narrow caricature, despite its recent history as a cover term signifying, and effecting, social empowerment. This time, “Roman” had become a caricature whose referents were condensed into particular sounds, text, instrumentation, and rhythmic patterns. Ironically, such features had in large part been produced within Roman communities, and shaped by ongoing mediations between community members and non-Roman producers.

In the struggle for control over representations, two artists have enacted differing strategies for negotiating the potential problems of symbolic condensation. Rather than posit these as exemplary “types” which in themselves only constitute iconic categories, I propose that these two examples be viewed as case studies for strategic negotiations of self- and communal expression of identity in tension with pre-established discursive categories. In this way, their actions and musical choices can be seen as discourses (in the sense of conversation) with and against already established discourses (in the Foucaultian sense). [27]