Cultural Policy and the Tunisian Ma'lūf: Redefining a Tradition

Author(s): Ruth Davis

Source: Ethnomusicology, Vol. 41, No. 1 (Winter, 1997), pp. 1-21
Published by: University of Illinois Press on behalf of Society for Ethnomusicology
Stable URL: http://www.jstor.org/stable/852576
Accessed: 30-05-2016 15:28 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact [email protected].

University of Illinois Press, Society for Ethnomusicology are collaborating with JSTOR to
digitize, preserve and extend access to Ethnomusicology

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
VOL. 41, No. 1 ETHNOMUSICOLOGY WINTER 1997

Cultural Policy and the Tunisian Ma'lqft

Redefining a Tradition

RUTH DAVIS UNIVERSITY OF CAMBRIDGE

CORPUS CHRISTI COLLEGE

In 1961, five years after Tunisian independence, the government created

a separate Ministry of Cultural Affairs.1 Its mission, fixed by presidential
decree was, "to promote and harmonize cultural activities through the
elaboration and execution of a program of development and diffusion of
culture throughout the nation." In particular, the new ministry was required
to (a) cultivate the national cultural heritage; (b) provide a basic, popular
education in all its aspects; and (c) forge cultural links with foreign coun-
tries, especially with international organizations concerned in any way with
culture (Kacem 1973:30).
In order to effect its policies, the ministry established a nationwide
network of cultural and recreational institutions called dur al-thaqifa
("houses of culture"; s. dir al-thaqifa) whose activities were coordinated
by a parallel network of national, regional and local cultural committees.
The ministry itself was divided into various specialized departments, includ-
ing the Department of Music and Popular Arts which was directed for the
first eighteen years by the Tunisian composer, performer, and musicologist
Salah El-Mahdi. Its specific responsibilities were (a) to conserve and pro-
mote the national heritage; (b) to promote research in the domain of mu-
sic and popular arts; (c) to organize the artistic profession; (d) to organize
local, regional and international festivals; and (e) to maintain and stock the
national sound archive (ibid.: 36).
Underlying the government's cultural policy was a certain paradox. On
the one hand it promoted the ideology of a national cultural identity, based
on the concept of a cultural heritage; on the other, it required receptivity
to external influences, thus aspiring by implication to modernization and
change. The very concept of a national identity was, like the political ide-

? 1997 by the Board of Trustees of the University of Illinois

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
2 Ethnomusicology, Winter 1997

ology of nationalism, a Western import, hardly reflecting the reality of

Tunisia's ethnically, socially, and culturally diversified population. The ideo-
logical duality inherent in the government's cultural policy is crystallized
in a statement by the former Minister of Cultural Affairs, Mahmoud Messadi:
"Cultural development must be regarded both as a factor of national iden-
tity-or cultural identity-and as an instrument of transformation or change
of a society ... development and modernization must be pursued under the
triple banner of (i) fidelity to oneself, (ii) the profound will to renew, and
(iii) the wise and rational selection of borrowings and influences to inte-
grate into the modernization process" (ibid.: 40).
Between 1982 and 1985, I spent some two years in Tunisia research-
ing the urban Arab-Andalusian tradition, al-ma'lUf It soon became appar-
ent to me that the tradition-in the form in which it was cultivated in state-

sponsored institutions, documented in government publications where it

was designated "The Tunisian Musical Heritage," and represented by the
prestigious national ensemble, the Rashidiyya (named after the aristocratic
patron and amateur of the ma'l f, Muhammad al-Rashid Bey [d. 1759])-
was a relatively recent phenomenon: its origins apparently dated back less
than fifty years to the founding of the Rashidiyya Institute in Tunis in 1934,
the same year that Habib Bourguiba founded the Neo-Destour Party whose
policy of active resistance to the French Protectorate eventually resulted
in independence. Many Tunisians I spoke to considered this coincidence
to be significant, since it provided a symbolic link between the goals of the
Rashidiyya-to conserve and promote traditional Tunisian music-and the
broader objectives of Tunisian nationalism. These goals foreshadowed those
of the Department of Music and Popular Arts twenty-seven years later; Salah
El-Mahdi himself had been leader of the Rashidiyya ensemble since 1949,
and remained so throughout the period of his responsibility for the
government's musical policies.
In my conversations with Tunisians, archival research, and travels to
ma'luf centers outside Tunis, I found evidence of allegedly older ma'lfif
traditions which had largely been superceded since the government took
control of the nation's cultural activities in the 1960s. In certain respects,
the "official" ma'lff, originally promoted by the Rashidiyya Institute, then
taken up by the Ministry of Cultural Affairs, amounted to an "invented tra-
dition";2 what had once been the popular entertainment music of certain
urban communities within a limited geographical area, whose performance
practice had reflected local and even personal identity, had been trans-
formed, in the quest for a "national culture," into a symbol of national iden-
tity. The process had not only involved changes in its social function and
contexts and in the social status of its performers; it had also resulted in
radical changes in performance practice reaching to the very substance of

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
 Cultural Policy and the Tunisian Ma'lof 3

the melodies themselves. In the following article I will trace the process
of this transformation and illustrate its effects on the ma'lof tradition of one
particular community, that of Testour.3

The Development of the "Official" Ma'lfif

The term ma'lof ("that which is familiar; custom") designates the collec-
tion of songs and instrumental pieces allegedly imported by Andalusian refu-
gees-Moslems and Jews fleeing the Christian reconquest of Spain from the
twelfth to the seventeenth centuries. Its melodies are conceived according
to the principles of maqdm (pitch organization or mode; pl. maqdmat) and
iqd' (rhythmic-metric organization; pl. iqd'dt) which generally characterize
urban Arab music. The core repertory is classified by maqdm into thirteen
song cycles called n7ibdt (s. nuiba); in performance, a variable selection of
pieces from the same noba are organized in a fixed sequence of iqd'dt.4
Until the 1960s, performances of the ma'lif were generally confined
to towns and cities of the northern and coastal regions: the repertory was
virtually unknown among the rural and nomadic populations of the inte-
rior and the south (Abdul-Wahab 1918:117). In certain other respects, how-
ever, the ma'lof was relatively well-qualified to fulfill its newly designated
role as national musical heritage. As an urban repertory it was at least com-
mon to several different communities, unlike the rural and desert traditions
which are regionally considerably more diversified (Abdul-Wahab 1918:116).
Moreover, within its urban and regional confines it was a genuinely popu-
lar tradition, crossing class and religious boundaries: its melodies were
cultivated by aristocratic and bourgeois patrons and amateurs in their pri-
vate homes, by Sufi musicians in their lodges, and by Sufis and professional
musicians-typically Jews, barbers, and other members of the lowest so-
cial classes-in cafes and in celebrations for weddings, circumcisions, reli-
gious festivals and pilgrimmages. The Sufi firaq (s. firqa: musical group) used
to give ma'lof concerts after their religious ceremonies, which were at-
tended by the community at large. The Sufis also adapted the melodies of
the ma'lof to sacred texts for use in their ceremonies, while the Jews
adapted the melodies to Hebrew texts for use in synagogue worship.
As an Arab literary tradition with a basis in Arab music theory, the ma'lof
enjoyed a certain intellectual status at this time; it had a legend which
passed in both popular and scholarly imaginations as a history,5 and it car-
ried the prestige associated, among urban communities at least, with
Andalusian culture generally. Moreover, uniquely among Tunisia's diverse
range of indigenous musical traditions, the ma'lof had already been system-
atically classified, studied, documented, and provided with modem perform-
ing and teaching models by the Rashidiyya Institute.

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
4 Ethnomusicology, Winter 1997

When the Rashidiyya Institute was founded, the very concept of a

public secular institution devoted to the performance of indigenous music
was revolutionary. Traditionally, public music making outside the Sufi
lodges was relegated to non-Moslems and members of the lowest social
classes; amateur musicians of higher social standing were obliged to confine
their performing activities to the privacy of their homes. A principal mo-
tive of the Rashidiyya's founders, who included leading members of Tunis's
artistic, intellectual and political establishment (the founding president was
Mustafa Sfar, mayor of Tunis), was to provide amateur musicians, regard-
less of religion or social class, with a respectable performing environment,
comparable in status to a Western music conservatory or concert hall.6 Their
immediate model was the French music conservatory established in Tunis
in 1896.
From the outset, the Rashidiyya Institute introduced radical innovations
in teaching methods and performance practice, inspired largely by West-
ern models. Traditionally there were two types of performance practice for
the ma'lof. Ma'liffkham ("raw" or "unrefined" ma'lof) was associated par-
ticularly with the Sufi firaq; it comprised a male chorus singing in uni-
son, accompanying themselves only by hand-clapping and/or percussion
instruments such as the darbukka (vase-shaped drum), naqqdrat (pair of
small kettle-drums), .tar (tambourine), or bandir (frame-drum with snares
on the underside). Outside the lodges, the ma'lof was also performed by
small solo instrumental ensembles, traditionally comprising a rabab (two-
stringed fiddle without a separate neck) or violin,'ad 'arbi (fretless short-
necked Maghrebian lute with four strings), naqqirit and tar (El-Mahdi
1981:49); during the twentieth century various Middle Eastern instruments
such as the 'iad sharqi (six-stringed lute), qdnfn (trapezoidal plucked
zither), and ndy (bamboo flute without mouthpiece) were sometimes
added, as were European instruments of fixed pitch such as the harmonium,
mandolin, and piano (ibid.; d'Erlanger 1949:341). The instrumentalists
doubled as chorus and sometimes also as solo vocalists; the melody instru-
ments followed the vocal line, embellishing it spontaneously in a simple
heterophonic texture. Both types of ensemble were directed by a shaykbh-
usually an elderly member chosen for his superior knowledge of the rep-
ertory-who was responsible for teaching the pieces to the other musicians
by the laborious process of repetition and memorization. In performance
the shaykh sang and played within the ensemble, selecting the pieces,
controlling the tempo, and leading the transitions from one piece to the
next.

In their efforts to conserve and promote the ma'lof the founders of the
Rashidiyya first had to assemble its sources. To this end, they invited the
most outstanding shaykhs in Tunis to form an ensemble. The initial response

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
 Cultural Policy and the Tunisian Ma'ltf 5

produced an unusually large and arbitrary collection of instumentalists and

vocalists, including doublings of the same type of instrument, whose first
attempts at rehearsal were chaotic: many of the shaykhs were unfamiliar
with each others' versions of the melodies, while the clarity of texture
characterizing the solo instrumental ensembles was compromised by the
instrumental doublings.' In an attempt to rescue the situation, the president
of the Rashidiyya and mayor of Tunis, Mustafa Sfar, invited the young Tu-
nisian violinist Muhammad Triki both to lead the ensemble and to standard-
ize the shaykhs' diverse interpretations through the use of musical notation
(see Davis 1992:90ff).
Triki remained leader of the ensemble until 1949, when he was re-
placed by his younger colleague, Salah El-Mahdi. During this period Triki
created a more coherent format, which he acknowledged was deliberately
modeled both on the Western symphony orchestra and the contemporary
Egyptian ensemble (see El-Shawan 1984:271ff.). Although the exact lineup
has always been variable, by the mid-1940s the Rashidiyya ensemble had
settled into a standard format characterized by a mixed male and female
chorus and a separate instrumental section comprising a core body of vio-
lins, one or more cellos and double basses, naqqirit, tir, and darbukka, and
an arbitrary smattering of traditional urban Arab melody instruments such
as 'td 'arbi, 'td sharqi, rabab, qintan, and nay.
During the same period, Muhammad Triki supervised the transcription
of the entire repertory of the ma'lif known to the shaykhs of Tunis from
oral tradition into Western notation, modified to accommodate the variable
intervals of the Tunisian maqimit. On the basis of this work he established
separate patterns of rehearsal and performance for the instrumental and
vocal sections which have remained unchanged until the present. The in-
strumentalists were coached by Triki and aided by notation on music stands;
the vocalists, in contrast, learned the repertory from a chorus master by the
traditional method of repetition and memorization. Finally, both sections
would reunite to perform under Triki: the chorus from memory, the instru-
mentalists still following notation. Triki himself conducted the ensemble
from the front with a baton, like the conductor of a Western symphony
orchestra.8

In 1944 a new recruit, the young nay player Salah El-Mahdi, established
a separate music school in the Rashidiyya Institute, functioning indepen-
dently of the ensemble. In addition to solmization and music notation, its
syllabus included theory of the maqamat and iqi'At, Arab music history, and
instrumental technique. After Tunisian independence, the Rashidiyya's
curriculum was to become the cornerstone of the music syllabus offered
by the National Conservatory of Music, Dance and Popular Arts. The Na-
tional Conservatory's syllabus, which combined studies in Tunisian, Middle

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
6 Ethnomusicology, Winter 1997

Eastern and Western music, was formalized by presidential decree on 1

January 1958. It replaced the French music conservatory, and became the
model for music education institutions throughout the country.
Before the creation of the Ministry of Cultural Affairs, the work of the
Rashidiyya had no affect on the ma'lff traditions of the Sufi brotherhoods,
or on ma'lfif practices generally outside the capital. After independence,
rather than foster the traditional types of performance practice and social
contexts, the government created the mechanism whereby the innovations
of the Rashidiyya would be extended throughout the nation. It created a
network of amateur ensembles and music schools along the lines of the
Rashidiyya and its school in doir al-thaqdfa throughout the country; it sup-
plied these institutions with instruments and sent out graduates of the
Rashidiyya school and its successor, the National Conservatory, to organize
and teach the new ensembles using the Rashidiyya's transcriptions as
sources. Thus the ma'lof was introduced to areas where it was previously
unknown, and new types of performance practice and institutional contexts
were introduced to communities where the repertory had traditionally been
cultivated.

In order to both encourage and monitor the state-sponsored ensembles,

the Department of Music and Popular Arts established a program of inten-
sive, residential music courses held in the school vacations, and an annual
cycle of competitions and festivals, culminating in the national competition
and festival held each summer in the traditional ma'lof center, Testour. In
1969 the government instituted a policy requiring professional musicians
to hold qualifying cards for their particular types of repertory; since then,
the regional ma'lof festivals have doubled as examination centers for the
licensing system.
The impact of the government's initiatives was reinforced by the dra-
matic demise of the traditional centers of communal music making, the Sufi
lodges, or zawaiya (s. zdwiya). The new government regarded the Sufi
movement as both politically suspect and socially regressive: certain
shaykhs were accused of cooperating with the French against the resistance
movement, while the Sufis' esoteric beliefs and activities, particularly the
extreme physical practices they performed in trance, were seen as shame-
ful relics of an oppressed and backward society, counteractive to modern-
ization and progress. The government reacted by confiscating the Sufis'
lands and possessions, suppressing their extreme physical practices, and
generally denigrating the Sufi movement as regressive and corrupt. As their
activities were restricted and their social prestige plummeted, so member-
ship of the brotherhoods declined and lodges throughout the country
emptied and ceased to function. In some communities, former Sufi musi-
cians continued their traditional musical activities outside the lodges; how-

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
 Cultural Policy and the Tunisian Ma'lfif 7

ever, separated from their source they operated in a social and spiritual
vacuum that was hardly alleviated by the new secular contexts.
During the 1960s and 1970s the Ministry of Cultural Affairs published
the Rashidiyya's transcriptions in a series of nine volumes entitled Al-turath
al-misiqi al-tunisi (The Tunisian Musical Heritage), which were distrib-
uted to the new ensembles and conservatories throughout Tunisia. In his
introduction to the third volume and again in his official history of the
Rashidiyya Institute, Salah El-Mahdi refers to three national congresses or-
ganized by the Ministry of Cultural Affairs in the early 1960s, to which
shaykhs of the ma'lof from all the regions were invited (the congresses were
held at Tozeur in 1963, and in Tabarka and Carthage in 1964). According
to El-Mahdi, all the ntibat known to all the shaykhs were recorded, subse-
quently to be transcribed and systematically compared by a "Committee for
Musical Comparisons"; apparently, the results revealed 'no essential differ-
ences' between the versions (n.d.(a):12; 1981:81-82). Thus the transcrip-
tions in Al-turathb al-misiqi al-tfnisi are officially presented as the sole
national tradition of the ma'lof.
During my research in Tunisia I discovered considerable controversy
over El-Mahdi's claim, and over the alleged homogeneity of the Tunisian
ma'lhif. Many of my interviewees, including musicians employed in state
music institutions, maintained that there were in fact distinctive regional
traditions. Muhammad Triki, who had directed the "Committee for Musi-
cal Comparisons," confirmed that some shaykhs had recorded some of their
repertory at the congresses, but he denied that the recordings had been
comprehensive, or that any systematic transcriptions or comparisons had
been made. I could find no one who admitted to having seen the alleged
transcriptions, and El-Mahdi himself claimed that they were lost. Evidently
the purpose of the three congresses was to establish the authority of the
notations published by the Ministry of Cultural Affairs as a pan-Tunisian
tradition, thus justifying their use as prescriptive sources for ensembles
throughout Tunisia. In our conversations, El-Mahdi described the published
notations as the sole "correct" versions of the melodies, containing all their
essential details.

In the archive of the R.T.T. (Radiodiffusion Tdlivision Tunisienne)

in Tunis I came across recordings of several nubat of the ma'luf performed
by the firqa of the 'Isdwiyya brotherhood of Testour; the recordings were
made by government officials in 1960, the year before the creation of the
Ministry of Cultural Affairs, and the musicians were directed by Shaykh
Muhammad ben Ismail, who was later to represent Testour in El-Mahdi's
three national congresses. On listening, it seemed to me that there were
indeed differences between the Testour versions of some of the songs and
the notations in Al-turath al-musiqs al-tainisi.

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
8 Ethnomusicology, Winter 19.97

I continued my research in Testour, to discover whether the same

versions of the melodies were still being performed there, more than twenty
years on. I was curious to know whether the people of Testour themselves
accepted the "official" view of a single pan-Tunisian tradition, or whether,
on the contrary, they considered their own tradition to be distinctive. More
generally, I was interested to assess the relationship of the Testour tradi-
tion with the one currently promoted by the Ministry of Cultural Affairs.

The Ma'lhif in Testour

Testour lies in the Medjerda river valley, some seventy-seven kilome-

ters northwest of Tunis. Its inhabitants claim to be descended from
Andalusian refugees who founded the town in the early seventeenth cen-
tury, and they maintain a strong sense of their Andalusian identity. In rec-
ognition of this identity and as part of a general policy of decentralization,
the government has held the annual international festival of the ma'lof there
each summer since 1968.
I first visited Testour in June 1982 to attend the ma'lof festival as guest
of the Ministry of Cultural Affairs. The ensemble representing Testour com-

prised three violins, two cellos, 'Od sharqi, naqqairat, tar, darbukka, and six
young male and female vocalists. The instrumentalists were all young men,
and they were led by the 'Td player, Mongi Garouachi. They performed from
memory, but Mongi confirmed that their sources were the notations in Al-
turath al-mfisiqi al-ttinisi. During my visit, I discovered that there was a
second, entirely separate ma'lof ensemble of Testour comprising former
members of the firqa of the 'Isawiyya brotherhood. Apparently, these older
musicians still performed ma'lof kham at communal celebrations, but they
had dissociated themselves from the dcr al-thaqafa and were not involved in
the festival. Over the following year I revisited Testour several times, accept-
ing the generous hospitality of Mongi Garouachi's family and attending re-
hearsals and performances of both ensembles. Gradually, I pieced together
the following account of ma'lof activities there since independence.
In the wake of the government's hostilities, the firqa of the 'Isawiyya
withdrew from the zawiya, but the musicians continued their traditional
ma'lof performances at weddings and circumcisions. When the dar al-
thaqafa was built in the early 1960s, the firqa at first gravitated there to
resume its rehearsals; however, the musicians felt uncomfortable in the
strange environment which lacked the atmosphere and traditions of the
zawiya; gradually, their rehearsals decreased, their standards dropped and
their repertory declined, and teaching became an ad hoc affair, mostly
conducted in caf6s and private homes. In the past, performances at wed-
dings and circumcisions had been given freely, as an honor to the host fam-

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
 Cultural Policy and the Tunisian Ma'luf 9

ily and the community. Now the musicians began to demand payment for
their services, and while they continued to accept younger recruits in their
rehearsals, they excluded them from their paid performances so that there
would be more money to go around. Previously, membership of the Sufi
firqa had spanned all social classes; by the late 1960s, it had dissolved into
ad hoc assemblings of poor people seeking opportunities to earn money.
In 1969 the government provided the dir al-thaqafa with musical in-
struments and a music teacher from Tunis. The latter was a graduate of the
National Conservatory and a violinist in the Radio ensemble; however the
firqa resisted instruction in what they regarded as their own tradition by
an outsider, and a young man. Defeated, the violinist returned to Tunis and
the government suspended its funds for a music teacher.
Meanwhile, one of the younger members of the firqa, Mongi Garouachi,
had moved to Sfax, a large coastal town southeast of Tunis, where he had
found work as an engineer. The music school of the dar-al-thaqdfa of Sfax
had developed into a conservatory with the same syllabus as the National
Conservatory in Tunis. Mongi became a part-time student there, and was
thus initiated in the published versions of the ma'lif. When Mongi returned
to Testour in 1973 he agreed to take charge of the musical activities of the
d~r al-thaqafa unpaid. His aim was to create a modem-style ma'lof ensemble
which would nevertheless preserve the individuality of the Testour tradi-
tion, by integrating the former Sufi musicians with new, younger recruits:
the experienced musicians would teach the traditional melodies to the
younger ones, who would learn the instruments provided by the govern-
ment. Using his newly acquired notational skills, Mongi began to transcribe
the traditional melodies from the older men as a basis for teaching the in-
strumentalists.

Mongi's experiment was short-lived: the new standards and methods

he had acquired in Sfax conflicted with those of the older musicians. He
found their teaching methods, which involved remoreslessly repeating
passages that were too long for their pupils to remember, inefficient; and
while he tried to correct the old mens' pronunciation and grammatical
mistakes, after countless repetitions of Mongi's versions they reverted to
their former habits. The ensemble performed in the Testour festivals of 1975
and 1976, and then dissolved. The older musicians withdrew from the dir
al-thaqifa to confine their performances to the traditional, communal con-
texts, while the younger musicians formed a new ensemble under Mongi
in the dar al-thaqafa. Mongi continued to prepare his own transcriptions
to use as sources; however, liberated from the older musicians, he no longer
felt bound to replicate their versions exactly. He began to adapt the melo-
dies according to his personal taste, which he admitted was influenced by
performances of the Rashidiyya ensemble, recorded from the radio, and by

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
10 Ethnomusicology, Winter 1997

the versions in Al-turath al-musiqi al-tunisi. As the only person in Testour

with a fluent knowledge of musical notation, Mongi taught the ensemble
by the traditional method of repetition and memorization; he sang and
played the melodies, phrase by phrase, on the 'ud, and the ensemble re-
peated them until they were memorized.
The activities of the youth ensemble focused exclusively on the Testour
festival: the musicians began to rehearse their program a few weeks before-
hand; afterwards they split up until the following summer. Mongi explained
that it was difficult to assemble the musicans because of their incompat-
ible schedules: some were schoolchildren who could not rehearse late at
night, while those who studied or, like Mongi, worked outside Testour,
could not rehearse at any other time. Apparently some thirty people had
wanted to perform in the 1983 Testour festival, but only fifteen were
sufficiently prepared to do so.
Many people thought that the inhabitants of Testour were gradually
loosing interest in the ma'lff. The younger ones tended to be more enthu-
siastic about the new, popular songs on the radio and television than the
ma'lff of the Rashidiyya, the dir al-thaqifa, and the old men at weddings:
there were no glamorous models for the ma'lff to attract them. Over the
past decade, people from surrounding rural communities had begun to
settle in Testour in search of work; the newcomers had brought their own
musical traditions, including the mizwid (single-reed bagpipes with paired
chanters), a traditional instrument which had been popularized by the mass
media. Increasingly, the indigenous families were inviting the mizwid play-
ers to perform at their weddings, after a token appearance by the old men.
Mongi frequently complained about the poor technique of his players:
their intonation was unreliable and their phrasing undisciplined, but he had
no time to remedy these faults. Many of the instruments needed repairing
or replacing, but he had neither the time nor the money to do either. Even-
tually, Mongi no longer found time to prepare his own transcriptions: for
the 1983 festival he used the published notations as his source for a per-
formance of niabat al-asbabdn. And instead of the sequence of pieces tra-
ditionally sung in Testour, Mongi chose the standard repertory of the
Rashidiyya ensemble which he had learned at the residential music course
run by the Department of Music and Popular Arts the previous year.9
The people of Testour believed that their town had a distinctive tradi-
tion of the ma'luf which was still maintained by the older musicians. How-
ever, many, including Mongi, thought that this tradition was in danger of
being entirely obliterated by the version "imposed" from Tunis. Mongi
maintained that the only way to secure both the survival and official rec-
ognition of the Testour tradition was to provide it with a complete source
of musical notations, comparable to the ministry's publications. With a full-
time job outside Testour, Mongi lacked the time to carry out such a project

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
 Cultural Policy and the Tunisian Ma'lof 11

himself: he insisted that it would take the resources of a full-time music

teacher to transcribe all the traditional melodies known to the older musi-
cians, and to teach the ensemble musical notation so that they could per-
form them accurately. Moreover, his proposed project had suffered a blow
the previous year with the unexpected death of Shaykh Muhammad Ben
Ismail, unanimously acknowledged to be the last great shaykh of the ma'lof
in Testour, without a successor. Many traditional melodies were believed
to have died with him.

There was a general feeling among Testour's inhabitants that the gov-
ernment had let them down: it had recorded their musical traditions, but
it had not provided copies of the recordings, nor had it provided the dar
al-thaqafa with recording equipment of its own. It had sent them an unsat-
isfactory music teacher withour providing the funds for a replacement. It
had supplied them with musical instruments without providing the funds
to maintain them. Each year, the government used Testour to promote
ma'lof ensembles from all over the Maghreb, yet it had done little, they felt,
to promote the town's own tradition.

Musical Evidence

In example 4 (Appendix) I compare two sources from Testour (T60

and T83) of the opening piece of nafbat al-asbahan, the btaybi "ft jannati'l
firdawsi" with the notated version in Al-turath al-mrisiql al-tfnisi (TMH).
Source T60 represents my transcription of the 1960 recording in the R.T.T.
archive; T83 represents my transcription of a performance by surviving
members of the firqa of the 'Isawiyya brotherhood which I recorded in May
1983.

The btybih is the first of the five main genres of the noba. I was ad-
vised by musicians in Tunis and Testour that it would be the most suitable
for comparison: as the opening, most serious genre with a relatively long,
slow rhythmic-metric cycle, it offered the greatest scope for individual in-
terpretation. Example 1 shows the characteristic pattern of iqd' btcdybi ac-
cording to Salah El-Mahdi in his introduction to Volume 3 of Al-turath al-
muisiqi al-tisnisi (n.d.(a): 16).
The poetic text is in the form of a muwashshah; the transcription in
Al-turdth al-mtisiqi al-tfunisi comprises three strophes of two, three and
two lines respectively. The first and third strophes share the same rhyme
scheme and the same melodic setting (tdla) which is repeated for each
line; the central strophe has a different rhyme scheme and melodic setting
(bayt) which is likewise repeated for each line.
The notation in Al-turath al-mfisiqi al-tftnisi gives a single melodic round

for both t.la' and bayt, with signs indicating the number of repetitions. Ex-
ample 4 likewise gives a single melodic round for each section (tila': TMH

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
12 Ethnomusicology, Winter 1997

mm. 1-10, T60 and T83 mm. 1-9; bayt: TMH mm. 10-19, T60 and T83 mm.
11-18). In both Testour sources, the melody was repeated almost exactly
each time round, except for some slight discrepancies in the opening round
of the tala' in T60 which are indicated in the transcription; these probably
reflect the musicians' uncertainty at the start of the performance.
The Tunisian maqimat are built up of characteristic sequences of
smaller tonal units called 'uqid (s. 'iqd) comprising three, four, or five suc-
cessive pitch degrees. Since individual 'uqid may be transposed, typically
by intervals of a fourth or fifth, the definitive aspect of each 'iqd is its in-
terval structure rather than the actual degrees on which it is based. Example

2 shows
ing the El-Mahdi
to Salah scale of maqim al-asbah.n with
in his introduction its component
to Volume 'uqid,
Eight of accord-
Al-turath al-
miisiqi al-tunisi (n.d.(b): 33). Example 3 identifies the 'uqud actually used
in "fi jannati'l firdawsi" according to the sources in Example 4. The pro-
gression of 'uqid is indicated in Example 4 above the stave of TMH, with
each new 'iqd identified at its point of entry; the 'uqud for the Testour
sources are identical; they are identified above the stave of T60 only when
they differ from TMH.

Results of the Comparison

Sources T60 and T83 are virtually identical: during the twenty-three
years since the government had made its original recording, the firqa had
performed under the same shaykh, rejected the input of the music teacher
sent from Tunis, and dissociated themselves from the dar al-thaqifa and the
Testour festival: not surprisingly, perhaps, their traditional interpretation
had remained unchanged, showing no influence of the government's pub-
lished notations.

In the first section, or tila', all three sources share the same progres-
sion of 'uqud. However, in the Testour version, 'iqd bhij*z is less defined,
lacking the third degree (F#) or characteristic second interval in m.3. In
terms of melodic detail, TMH is generally more ornate (see for example m.
4 and the cadential passage mm. 7-9).
In both Testour versions and TMH, the second half of the bayt (mm.
14-18) replicates the second half of the tdla' (mm. 5-9). However, the

Example 1: iqi' btiyhIi

( = dum: strong beat. tak: weakbeat)

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
 Cultural Policy and the Tunisian Ma'lff 13

Example 2: maqim al-asbahan

HIJAZ NAHNAND

TA
lip- T

RAT RAST RAST NAHAWAN D RAST

progression of 'uqid in the first half of the bayt differs slightly in each ver-
sion. In T60 and T83, melodic cells x and y (mm. 11-12) are linked by an
ascent into 'iqd nabwand; in the corresponding section of TMH (mm. 10-
12) each melodic element is elaborated and repeated, thus extending the
melodic line. In T60 and T83, the following section (mm. 12-14) rests
within the trichord G-B of 'iqd nahawand; in contrast, the corresponding
section of TMH constitutes an elaborate descent from 'iqd nahawand
through hijaz and into rist, followed by an ascent through the same 'uqid.
In sum, TMH is generally more elaborate than T60 and T83 in terms of me-
lodic detail, and in the bayt it is also more dynamic in its contrasting of
'uqid and tessitura.

Contrasting Reactions
In general, I found two contrasting attitudes regarding the status of
regional ma'lof traditions such as that of Testour vis-at-vis the version pro-
moted by the Ministry of Cultural Affairs. The official view, represented by
Salah El-Mahdi and his succesor at the Department of Music and Popular
Arts, Fethi Zghonda, was that such alternative traditions simply did not exist:
any divergences from the official version were considered either "non-es-
sential" or "incorrect." Mohammad Triki, who produced the original tran-

Example 3: The 'uqiid of btiyhi 'fi jannati'l firdawsi' from niibat al-asbahin

HIJAZ
(TMH) NAHAWAND

A TI

/NAHAWAND

9)MH

SIKAH
(TMH)

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
14 Ethnomusicology, Winter 1997

scriptions for the Rashidiyya ensemble in order to standardize the diverse

versions of the individual shaykhs, recognized that his work represented a
compromise, reflecting his personal preferences; nevertheless, he regarded
these transcriptions as the ideal version, or yardstick against which all in-
terpretations should be measured. Triki described my recordings from
Testour as "corrupt," maintaining that the former Sufi musicians had dis-
torted the melodies through their ignorance and lack of artistry.
In the traditional ma'lof centers of Zaghouan and Sidi Bou Said, where
the former Sufi musicians had been more receptive than those of Testour
to the efforts of music teachers from Tunis, I discovered a historical dimen-
sion to the official ma'lof ideology. According to these communities, the
published notations represented the pure, unadulterated version of the
repertory originally imported by the Andalusian refugees, which had be-
come distorted and fragmented through centuries of oral transmission and
neglect under foreign rule. In recent years, the tradition had been restored
to its original identity by the Rashidiyya Institute, whose work had been
taken up, since national independence, by the Tunisian government itself.
By drawing on the myth of the ma'lof's Andalusian origins to justify the
authority of the published notations, this explanation supplied a link be-
tween the ideology of national identity and nostagia for a past golden age,
when the Arabs were rulers rather than slaves to Ottoman or French op-
pression.
The official view represented the ma'luf as a fixed canon with a fixed
melodic identity, defined by the notations in Al-turath al-mursiqi al-tunisi.
In contrast, certain establishment musicians, including members of the
Rashidiyya and radio ensembles and teachers in the National Conservatory,
argued that there were other, authentic ma'lof traditions which were not
represented by these notations. Muhammad Saada, leader of the Rashidiyya
ensemble during the period of my research, and Abdulhamid Ben Algia,
former leader of the Rashidiyya ensemble and Director of Music at the R.T.T.,
took this argument a step further: not only did they defend the legitimacy of
such alternative interpretations; they maintained that as an oral tradition the
ma'lof was inherently variable: its authority was vested in the personal inter-
pretations of individual shaykhs, whose particular renderings at any given
time determined the "traditions" of their respective communities.
In a previous article I showed how successive leaders of the Rashidiyya
ensemble had transferred the alleged freedom of interpretation enjoyed by
the shaykhs to the written tradition: whether overtly or covertly, certain
leaders had modified the notations of their predecessors according to their
personal tastes (1993). Thus El-Mahdi's notations in Al-turath al-m7siqi al-
tunisl were not necessarily note-for-note reproductions of Triki's original
transcriptions, nor were Ben Algia's notations for the Rashidiyya and radio

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
 Cultural Policy and the Tunisian Ma'luf 15

ensembles identical to el-Mahdi's. Similarly, when Mongi took over the lead-
ership of the youth ensemble of Testour, he admitted to altering the tradi-
tional versions of the melodies to suit his own taste. However, a process
that may have occurred naturally and spontaneously in oral transmission
was inevitably proving deliberate, cumbersome, and time-consuming when
applied to notated sources. Despite their ideological convictions to the
contrary, by the early 1980s pragmatic considerations had taken over, and
both Mongi Garouachi and Muhammad Saada were basing their perfor-
mances on the notations in Al-turath al-mCisiqi al-tfinisi.
With the demise of the principal patrons of the oral traditions, namely
the Sufi brotherhoods, and with new commercial repertories making live
performances of the ma'lf increasingly redundant in cafes and communal
celebrations, the repertory has been depending increasingly for its survival
on the modem institutional outlets established by the government since
independence. And just as Triki's transcriptions were originally introduced
into the enlarged ensemble of the Rashidiyya to serve a practical need, so
in the modern performance contexts of the 1980s their successors, the
notations in Al-turath al-mrsiqi al-tfnisi, were proving indispensible tools
of convenience. Thus while the ideology justifying the authority of the
published notations could well remain in dispute, in practice these nota-
tions seem destined to become the only sources left.

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
16 Ethnomusicology, Winter 1997

Appendix

Example
asbahlin 4: Three versions of bt.iyhi 'ff jannati'l firdawsi' from niibat al-

Snahawand rist
TMH

Fi- ja- n na FI ja-na- ti_- I Fir

1st time only.

T60 FI jan na ti__ 1 Fir

-n Fi jan na ti 1 Fir

T83

F - ja n Fi jan na ti-- 1 Fir

hijaz nahawand
TMH

daw si a - 'a ya la lan y la

T60

T83

daw s ya la la l ya la la

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
 Cultural Policy and the Tunisian Ma'lhif 17

TMH

la la la lan ra ay tal qu su

1st time only.

T60 y ta
T60All other times

la ra a y

T83

la ra a y

O rst nahawand rdst

TMH _ _ _

ra 1 - li ya ' a la

A1st time only.

T60 1 4

ta 1 qu al 'i li yah yah la

T83

ta_ _ __ 1 qu al 'a- a Ii yah yah Ia

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
18 Ethnomusicology, Winter 1997

FINE
lo 20 ' x

TMH

la- la la lan a la la la- n ya- hal

T60 1,2, 3, 4, 5, 7.

l_ ra ml
1,2,3,4,5.

T83

la r- mi

nahawand

TMHX A

tu r- ya hal tu ra man ki n a
0

T60 nahdwand hijaz

6. xI _ _

la ya 'al Ii lik bil__ ba s 'ah__

T83

la ya 'al Ii lik bil ba s Ah

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
 Cultural Policy and the Tunisian Ma'lof 19

hijdz rdst hijaz nahawand

TMH..

a ya la la la la la la lIn min hu

T60 nahawand

yd 1a 1 la 1wa ya

T83 .

yI 1I 1I li wa yu hi

1rst nahawand

bi ha li hub bay

T60

hiz zu sum ril 'd

T83Lo t I V _ __ a

zu 'h sum ril d

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
20 Ethnomusicology, Winter 1997

17
rast
hhijz
" rst 1, 2.
TM H

ma li- h yd la la lan ya la la la la I yu har

T60 Inahdwand rdst

1 'a li ya h ra mi
T83

1 'a li ya h

nahawand
3.
TMH l a y a

la la la la yu a

Notes

1. The original designation Secretariat d'I!tat aux Affaires Culturelles et i~ Information

was changed to Ministdre des Affaires Culturelles in 1969. For the sake of simplicity, I use
the single designation Ministry of Cultural Affairs.
2. I use the term in the sense coined by Eric Hobsbawm to denote "a set of practices,
normally governed by overtly or tacitly accepted rules and of a ritual or symbolic nature, which
seek to inculcate certain values and norms of behaviour by repetition, which automatically
implies continuity with the past." (1983:1).
3. A shorter version of this paper was read at the Thirty-second World Conference of the
International Council for Traditional Music, held in Berlin in 1993. The research on which it is
based was partly funded by the Social Science Research Council and the National Endowment
for the Humanities, whose support I gratefully acknowledge. I wish to thank my Tunisian friends
and colleagues for their indispensible contributions and help. My spelling of Tunisian names
follows established convention rather than any consistent transliteration system.
4. The nufba is a large-scale formal principle common to Arab urban music traditions
throughout the Maghreb, combining unity of maqam with diversity of iqa'dt. For a detailed
examination of the structure of the Tunisian noba, including an analysis of a performance,
see Davis 1993.

5. The myth of the ma'lof's Andalusian origins is supported by Rodolphe d'Erlanger

(1949:334 ff.) and Salah El-Mahdi (1981:9-10). Tunisian scholars H. H. Abdul-Wahhab
(1918:115-17) and Mahmoud Guettat (1980:172-74), while acknowledging the ma'lof to be

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms
 Cultural Policy and the Tunisian Ma'luf 21

an imported repertory, maintain that its development occurred simultaneously in Islamic Spain
and the Maghreb, and that the Andalusian refugees merely reinforced a pre-existing tradition.
6. Tunisian musicians cite various other factors that contributed towards the founding
of the Rashidiyya Institute (see Davis 1993:89). An important catalyst was the First Interna-
tional Congress of Arab Music, held in Cairo in 1932, to which Tunisia had sent a delegation
of five leading musicians from Tunis under the auspices of the Baron Rodolphe d'Erlanger.
On their return, the musicians reported the Congress's recomendation that institutions be
established in all the Arab countries devoted to the conservation of the indigenous musical
traditions (El-Mahdi 1981:28).
7. The original ensemble comprised six violins, two rababit (s. rabab), five a'wid (s. 'Od),

three qawanin
& Marzuqi (s. qinfn), t.r, naqqirat, six male vocalists and one female vocalist (El-Mahdi
1981:49-50).
8. A similar distinction between vocal and instrumental sections was made by the Firqa
Al-Mfsiqd Al-'Arabiyya, established in Cairo some thirty years later. In both Tunisian and Egyp-
tian ensembles, the vocalists continue to rely on oral transmission while the instrumentalists
perform from notation.
9. This selection did not include the btayhi "fi jannati'l firdawsi," transcribed in exam-
ple 1. My recording of nobat al-asbahan performed by Mongi's ensemble in May 1983 corre-
sponded almost exactly to the notations in Al-turath al-mrisiqi al-tanisi.
10. Their performances corresponded almost exactly to the versions in Al-turdth al-
musiqi al-ta2nist.

References

Abdul-Wahab, Hassan Husni. 1918. "Le Developpement de la Musique Arabe en Orient,

Espagne et Tunisie." Revue Tunisienne, 25:106-17.
Davis, Ruth. 1992. "The Effects of Notation in the Performance Practice of Tunisian Art Mu-
sic." The World of Music 1:85-114.
. 1993. "Melodic and Rhythmic Genre in the Tunisian Noba." In Ethnomusicologica II,
edited by F. Giannattasio and G. Giuriati. Siena: Accademia Musicale Chigiana.
D'Erlanger, Rodolphe. 1949. La Musique Arabe, Vol. 5. Paris: Librairie Orientaliste Paul
Guethner.

Guettat, Mahmoud. 1980. La Musique Classique du Maghreb. Paris: Sinbad.

Hobsbawm, Eric and Terence Ranger, eds. 1983. The Invention of Tradition. Cambridge:
Cambridge University Press.
Kacem, Abdelaziz. 1973. "La Politique Culturelle Tunisienne." Annuaire de lAfrique du Nord,
29-44.

El-Mahdi, Salah, ed. n.d.a." Al-niba f'il-maghrib al-'arbi." In Al-turath al-mtisiqi al-tanisi.
Tunis: 3:3-16.
. n.d.b. "Niba al-asbahdn." In op. cit., 8:49-63.
El-Mahdi, Salah and Muhammad Marzuqi. 1981. Al-ma'had al-rashidi li'l mfisiqda al-tunis7ya.
Tunis: manshorat al-ma'had al-rashidi li'l-mosiqs al-tunisiya.
EI-Shawan, Salwa. 1984. "Traditional Arab Music Ensembles in Egypt since 1967: 'The Conti-
nuity of Tradition Within a Contemporary Framework?'" Ethnomusicology 28(2):271-
88.

This content downloaded from 132.174.254.159 on Mon, 30 May 2016 15:28:52 UTC
All use subject to http://about.jstor.org/terms