About Us
Our Activities
Programs in Israel
Publish Date

4 - Feb - 2004

  Midstream July/August 2003 For more information contact Midstream at: Midstream, Subscription Dept. 633 3rd Avenue - 21st Floor, New York, NY 10017
Related Links :
Help Israel!
  Join this urgent student campaign and help Israel with one of the toughest battles: the battle for public opinion on the Internet.
In The Web
Jewish Music

Sephardic Song

By: Judith Cohen

It's medieval -or is it? It's Spanish -or is it? It's Jewish -well, yes, it certainly is. But what makes it Jewish? What is Sephardic music anyway?

Sephardic music is really a complex of musics. Depending on how one interprets the term "Sephardic," it might refer to music of the Jews descended from those exiled from late 15th-century Spain and Portugal, or it may refer to the music of all Jews not categorized as Ashkenazic, even though such a large proportion of them would more accurately be designated "Mizrakhi" -or "Eastern," "Oriental."

For our purposes, the first, original sense is the one which will be used: Sephardim will be understood as those Jews descended by blood or by culture from those who left Spain and Portugal -Sefarad -before or at the time of the Expulsions (1492 from Leon and Castile, 1496-1497 from Portugal, 1498 from Navarra), or even afterwards, leaving as converted New Christians and resuming their Jewish identities where it was safer to do so. Nothing is simple, and, of course, many of them became assimilated into environments in which their Hispanic language eventually was lost, in Syria or in Amsterdam for example. But here we'll talk about those who at least until recently did maintain the language, including in their stories and songs.

Before we get to these songs, a brief look at the language is in order. Popular perception often sees it as some sort of "frozen medieval Spanish." In fact, it is neither frozen nor medieval, though it does preserve many archaic aspects of vocabulary and syntax that endear it to philologists. At the time of the Expulsion, there was not a single language known as "Spanish," so even if the language of the Sephardim were "frozen," it would have to have been so in many different forms. At the time of the Expulsions, Castilian was just emerging as the dominant language of what was becoming Spain. Catalan and Galician were the important literary languages, and the Jews who left spoke Castilian, Catalan, Galician, and/ or whatever varieties of Andalusian Romance, Asturo-Leonese, and so on were common where they were living. Those from Portugal spoke Portuguese, and, if they had fled to Portugal from Spain, whatever they had spoken there. It is unclear to what extent Jews mixed Hebrew into their everyday spoken language. All this coalesced during the centuries of the Diaspora into a primarily written language, Ladino, and into many varieties of the vernacular, which were known by different names in different places.

Ladino, strictly speaking, is the word-for-word translation from Hebrew, as if one were to render "ha-laila hazeh" into English as "the night the this," "la noche la esta." This was done primarily to avoid the possibility of any errors; in the spoken language, people said "esta noche," "tonight." Ask a middle-aged Sephardic person who says he speaks "Ladino" whether his mother used the term and you are likely to hear, "No, at home we called it spaniol," or "No, at home we called it dzhidio" (the exact equivalent of "Yiddish," the "Jewish" language). Or "dzhudezmo." Or, in Morocco, "khaketia." The habit of calling all of these "Ladino" arose relatively recently, in the past couple of decades, originally as an error, and now such a widespread one that it is becoming the norm. Scholars and many others prefer the umbrella term "Judeo-Spanish," which includes all of these, and this term will be used from here on: if "Ladino" is used, it refers specifically to the literal translation from Hebrew described above.

A popular misconception is that Judeo-Spanish song is "medieval Sephardic," or "medieval Spanish" music. Jewish and Muslim musicians and scribes did not notate their music in the Middle Ages. It is important to differentiate between the texts and the melodies of songs, and, in fact, between stories and sung texts: there is a tendency to confuse the origins of the words with the origins of the music. Very often, a song text, especially in the case of the romances (Hispanic narrative ballad) does in fact go back to the late Middle Ages, or the Renaissance, and/ or its story goes back to much earlier epic poems and prose chronicles. But their melodies are not medieval. As well, many, or even perhaps most, of the Sephardic songs heard on commercially available recordings are not from this early repertoire, but rather are songs from a later period of Sephardic life, either acquired in other ways long after the Expulsions from the Iberian Peninsula, or adapted from their Diaspora host cultures, or newly composed. The fact that the same text is found in both the Eastern and Western Mediterranean Sephardic Diasporas has often been taken as evidence of a song's pre-Expulsion origins; while this is sometimes the case, caution is needed when drawing conclusions, as there are other possible explanations. Jews continued to leave the Peninsula for many years after the Expulsions, as New Christians, often resuming or forging their Jewish identities in another country; some Jews also came to Spain and Portugal from other countries as traders and merchants. Even when performers do know that the "medieval connection" is tenuous at best, they and their audiences often cling to it, for reasons ranging from romantic historical associations to good marketing technique: the very term "medieval," in popular culture, tends to refer largely to fairs, restaurants, and street performances of an indeterminate epoch, bearing a largely wishful-thinking relationship to medieval European life.

While we have little idea of exactly what Jews sang in Iberia before the expulsions, there is no doubt that they were active in musical life. There is numerous mention, some by name, of male and female Jewish musicians in medieval Spain and Portugal; unfortunately, we have barely any record of the Jews' actual writing, and none of their music, or, indeed, of Muslim music of that time. We fare better with poetry: many Jewish poets ale documented in medieval Iberia. (See Brann; and, for women, Cohen 2001.) They wrote mostly in Hebrew, or sometimes Arabic; very little Jewish poetry in the vernacular has been preserved in manuscripts. Elsewhere, I have presented suggestions for reconstructions. (Cohen 2000)



Though we do not know how Sephardim sang in medieval Iberia, we do know how they sing today and have a good idea of how they sang up to about a century ago. There are recordings from early 20th-century Turkey and musical transcriptions beginning in the late 19th century, which, though far less reliable than recordings, at least give us some idea of Sephardic melodies of the time. The first recordings are mostly of male singers, with the exception of the legendary Victoria Rosa Hazan; these men were usually trained in the Ottoman tradition as well as in synagogue singing. Much of their singing has instrumental accompaniment; though they are traditional instruments, none of this is typical of women's singing, which traditionally took place mostly in domestic contexts, and a capella. There are a few documentary recordings from the 1950s; then from the late 1970s on, they begin to be more numerous and more accessible. (See my discography, under Bibliography.) Today, one can still find traditional singers of Judeo-Spanish songs, but rarely those who still know the old romances and wedding songs.

Generally speaking, Judeo-Spanish singing can be divided into two main regions: Morocco and the former Ottoman lands, with several subdivisions, especially Greece. One major musical difference between the two main areas is the use of the maqam system in the Ottoman area, a complex system of melodic progressions and patterns, which cannot really be compared to the Western scale system and which contains micro tonal intervals difficult to reproduce for those who have not grown up with them. The Moroccan songs occasionally use microtonal intervals, but to a much lesser degree. Vocal ornamentation in the Ottoman area tends to be more complex, though this distinction may have been somewhat exaggerated; many Moroccan Sephardic singers take justifiable pride in their vocal ornamentation, their "floreo."

Harmony is not a traditional aspect of Sephardic singing. There are a few exceptions: Flory Jagoda recalls her family in pre-war Bosnia singing in thirds; she is referring more to songs from the late 19th century on than to the old ballads and wedding songs. Women have played the stronger role in singing these genres, and usually a capella, while performing various domestic tasks, in the case of the romances, and accompanying themselves on frame drums for weddings (as do women in many Mediterranean cultures). When melody instruments are used -usually, though not always, played by men -it is traditionally a heterophonic accompaniment, as in North African and Middle Eastern music in general. However, the songs of the late 19th century on do lend themselves better to Western style accompaniment. Moroccan Sephardic women are skilled in the piercing ululation they call barwala or youyou (also common in African and Middle Eastern cultures, and as well in certain regions of Spain). This is usually heard at the end of songs for various festivities or celebrations.



Sephardic music -in fact, pan-Sephardic music, not only Judeo-Spanish -is primarily vocal: instruments, when they are used, are played to accompany songs. Instrumental practice among Sephardim has generally reflected that of the host culture: Greek, Turkish, Moroccan, etc. -plucked lutes (fretless: 'nd, the Middle Eastern lute; and in Turkey fretted saz or sometimes mandoline or the chumbush), kanun or santur (plucked or hammered Middle Eastern zither), violin, and hand drums (frame and goblet).

For weddings and other celebrations, musicians might also be hired from the Muslim community. On the other hand, skilled Jewish musicians would be hired by the Muslim community. Generally, Sephardic men played both local percussion and melody instruments, while women usually sang unaccompanied in domestic contexts, and at weddings accompanied their singing with tambourines and sometimes other percussion instruments. Molho describes Salonica Sephardic women using kitchen utensils as improvised percussion, in a manner reminiscent of Spanish and Portuguese village practice today. (Molho 2021) In the eastern Mediterranean, women musicians specializing in singing and drumming for weddings were known as tanyederas, and they played a central role in the wedding events. Some early 20th-century Ottoman-area Jewish schools taught 'udand mandoline to girls, and some women learned to play the piano. In any case, whether or not instruments are used, the main and always appropriate instrument in Sephardic music is the voice.

Medieval instruments as such are not used, except in cases such as the 'ud where the instrument has survived with minimal changes in traditional practice. Sephardim, like other traditional musicians, often adapt traditional instruments to current norms: at a Sephardic wedding one will definitely not find any medieval instruments, but will likely notice an electronic keyboard.


Song genres

The Moroccan and ex-Ottoman-area repertoires share the same general song categories, as described below. There is one important difference: the late 19th-early-20th century love songs and topical/recreational songs, often the best known Judeo-Spanish songs through recordings and concerts, are mostly from the ex-Ottoman areas rather than Morocco. In both Morocco and the ex-Ottoman areas, the music of the songs has been strongly influenced by the music of the local cultures. Greek popular songs, Turkish rhythmic patterns and light classical Turkish melodies, Moroccan rhythms, entire songs translated into Judeo-Spanish, and popular melodies from the early 20th century, including tangos and the charleston, have all played an important part in the development of what is now Judeo-Spanish song. While musical traditions have always moved around to some extent from one Sephardic community to another, the unprecedented technological changes of the late 20th century made the circulation of songs much more widespread and rapid. Recordings, increased travel, and internet communication, as well as new pan-Sephardic communities, have resulted in a much greater interpenetration of repertoires.

In Jewish music, including the Sephardic tradition, there has always been considerable musical exchange between religious and secular genres, and between songs in Hebrew and songs in the vernacular. Many printed piyyutim texts cite Judeo-Spanish song incipits for their melodies; however, there is no guarantee that the melodies used for these songs today are the same as those used centuries earlier, when the piyyutim were printed. (See Avenary, Armistead and Silverman.) From at least the late 16th century, use of the Turkish maqam for piyyutim is attested to, in the mystic poet-musician Israel Najara's monumental collection Zemiroth Israel; in this collection Najara also gives incipits from Spanish, Turkish, Arabic, and Gleek songs. (See Seroussi 1999.)

There have been various approaches to classifying Judeo-Spanish songs, often as a combination of form and function. In terms of form, they can be divided into romances, canciones, coplas, and a very small category of oraciones (Judeo-Spanish sung prayers). In terms of function and context, I suggest the following framework:

(1) Songs used primarily in domestic and recreational settings:

(a) romances (old and new) and other narrative songs

(b) love songs (usually cancianes)

(c) topical and recreational songs (cancianes or sometimes coplas)

(2) Songs used in primarily in contexts of religious and ritual life (litnrgical and para-liturgical)

(a) life cycle songs: birth, marriage, coming of age,

death (often" cancianes," also certain specific romances)

(b) calendar cycle songs (usually coplas)

(c) songs on religions and moral themes (usually capias)


Sephardic romances

The classic definition of the romance is a Hispanic narrative ballad, most frequently in assonant couplets of 16 syllables each, divided into octosyllabic hemistichs. The romance first appeared in a recognizable form in 1421, though many romances trace their origins to medieval or early Middle Eastern themes and stories. Some are later compositions; there are also early 20th-century broadside compositions about local events. Sephardic traditional singers do not always follow standard classification definitions. Moroccan Sephardim often say simply, "antiguos cantares muestros" (our old songs), or " cantares judios" (Jewish songs); or cantares de matesha or de columpio (when girls sang them on the outdoor swing in their courtyards); or cantares de cuna when used as lullabies.

Among Ottoman-area Sephardim, "romanza" may refer to narrative or lyric songs, and, because of its similarity to romance, the terms are often confused. In Weich-Shahaq's interviews with eastern Mediterranean informants (1984), they describe the romanza as "an old song, about kings and queens." Eastern Mediterranean Sephardim I have interviewed over the years generally do not limit their definition of "romanza" to this kind of content; in fact, one defined a "romanza" as "a song that is romantic," and this is a fairly pervasive popular image, often reflected in album notes. "Cantica" refers to wedding songs, lyric songs, or, in fact, almost any song in the east, while in Morocco." Cantares de boda' or "cantares de casamiento" is common for wedding songs. Romances used for mourning are called "endechas" or oinas, though not all endechas are romances.


Life cycle songs

Each step of the life cycle and each holiday has its own songs. The wedding and courtship songs are the most numerous in the life cycle group, but there are also birth songs, circumcision songs and laments, and, for boys, only a very few coming-of-age (Bar Mitzva) songs in Judeo-Spanish, as most of these are in Hebrew. Birth songs served not only to celebrate the event but also to keep the women awake through the eight-day continuous vigil before the circumcision, to ward off evil spirits. In a few cases, wedding piyyutim may also be used for a religious celebration (Simchat Torah) using the symbolism of the bride and groom as Israel and God. Songs used for the life cycle may not be directly related to the event, but may have texts evocative of it. These songs can be seen as a mixture of religious (for their function) and secular (for some of their themes), or the separation between secular and sacred can be seen as artificial: it is more of a continuum.



Coplas are defined by their form, basically rhyming couplets. (See Romero.) The majority are associated with religious life, especially, songs of the calendar cycle and Sabbath; but other themes can also be expressed in coplas. Some of them are macaronic, that is, they use two languages, in this case, Judeo-Spanish and Hebrew. Coplas are often passed down by written as well as oral transmission, though not with musical notation.

Endechas are sung not only for occasions of personal mourning but also, as part of the calendar cycle, for the solemn mourning day of Tisha be' Av. "Don Gato" is usually sung as a light-hearted children's song among Moroccan Sephardim, but when sung on Tisha be'Av. It uses a slow lament tune, and they say that whoever laughs while singing it on Tisha be' Av will weep on Rosh Hashanah (New Year), soon afterwards.


Lyric, topical and recreational songs

In the first half of the 20th century, many new songs in a basically Western European musical idiom emerged, and several older texts were given new melodies and performance treatment. In former Ottoman areas, the older style was sung largely in the maqam system, with corresponding ornaments and vocal timbre. The newer style sounds very different, including its singing style, the origin of its melodies and the Western scale system. From the later 20th century, several new compositions of known authorship were (and are still being) added, and performance practice is influenced both inside the community and, especially, by performers from outside Sephardic or even] Jewish culture. The latter particularly have been largely responsible for shaping public perception of what constitutes Judeo-Spanish song.

In several cases, the songs have a mixed musical identity, combining Judeo-Spanish words and Jewish customs with rhythms adapted from local Moroccan or Ottoman music. Proper names, and specific situations may also reflect Jewish life: Esterina Sarfaty is an obvious example, from Turkey. In some cases, these songs also exist in Spain or Portugal (e.g. La mosca y la mora, Hemsi #134), though they may also be relatively recent compositions; in others, they were composed or adapted by Sephardim.


Portuguese Crypto-Jewish traditions

In the case of the Sephardic diaspora, it is fairly easy to see how songs in Judeo-Spanish maintained a Sephardic, a Spanish-Jewish identity in a culture so different -Turkey, the Balkans, North Africa. In the case of those who remained in the Peninsula as forced converts, conversos, the role of songs as identity markers had to be different. The vernacular language was the same, but in Inquisition Portugal and Spain, Hebrew was dangerous, and could not be used as an open identity marker. In Portugal, the songs and prayers of Crypto-Jews who have continued to maintain their Jewish identity throughout the centuries following the Expulsions, are in Portuguese, with Adonay (God) virtually the only remaining Hebrew word.

A few Biblical ballads, now recited and not sung, exist in the Crypto-Jewish but not elsewhere in the Portuguese tradition: The Sacrifice of Isaac, Jonah, Daniel in the Lions' Den, The Crossing oj the Red Sea, and In Heaven there is a Castle. Other correspondences between Portuguese and Sephardic ballads need further research, which has been initiated by scholars such as Manuel da Costa Fontes and Pere Ferre. The Crypto-Jews in Portugal also sing a narrative ballad to a simple dance tune; the text, music and general style of the song do not appear to be older than perhaps the early 20th century. A song sung at Passover (" A Santa Festa") about the Exodus has older words, but the melody is by no means archaic. (See Cohen 1999.) Aside from these, and a translation of the Israeli anthem "Hatikvah," the Crypto-Jews I have interviewed, fairly extensively, know the songs of the regions they live in and sometimes adapt them to their own purposes.


Invented Jewish festivals

Over the past decade, the Red de juderias (Network of Jewish Quarters) has been established in Spain, and at least two small towns have created festivals containing several components related to Jewish culture. I have described these events in detail elsewhere. (Cohen 1999) In both towns, the fact that a neighborhood and some building remnants are known to have been Jewish several centuries earlier somehow seems to have led many people to conclude that the people who live there now -and, by extension, their songs -are too, a rather mystifying logical leap. Judeo-Spanish songs that are heard were learned from commercially available recordings, and, by coincidence, in both cases many of the songs chosen were from CDs of my own Moroccan Sephardic ensemble, Gerineldo and learned before the participants actually knew me, though later on, I gave workshops in Sephardic song to participants in both towns. (1995, 1999) The songs are, as we have seen, "medieval," but visitors to the festivals with often see them as just that. By extension, local songs have come to be seen by some -and even, rather cynically, sometimes presented by local officials -as "Sephardic" (interviews, 1998-2001), simply because they are sung by people who happen to live in the area designated as the "Old Jewish Quarter" -even when the songs in question are part of the Christmas season repertoire or late 19thcentury broadside ballads.



Since its musical precursors set out on the various roads of exile, Judeo-Spanish song has reflected its singers' processes of learning music from its environment, through all available media, and harvesting and adapting what appeals to its singers. It has been reclaimed in Spain and Israel, re-affirmed in Turkey and adopted by North Americans always on the lookout for a new musical mascot. After insiders earlier in the century westernized it to various degrees, outsider performers and media have added new images: medieval, romantic oriental, western art song, Middle Eastern, elements of jazz, and others, e.g., re-orientalization. Judeo-Spanish songs are performed more by artists from outside the communities than from within them. Once a private, domestic repertoire sung mostly by women, it is now heard largely through mediated performances, and often accompanied or even sung by men, though women still predominate. From a unidirectional move away from Spain to other areas Judeo-Spanish song movement has become multi-directional, originating in and moving among geographically and culturally far-flung countries and cultures. However one defines it, explains it, reacts to it, Judeo-Spanish songs and their singers maintain and re-affirm their identity with both aspects of this hyphenated term, centuries after the notorious attempts to sever them from each other.


This article belongs to the following subjects:

•Copyright 2007, The Hagshama Department