The Grana Jews of Tunisia

Of all the Western Sephardi communities that took root in the Islamic world, the Grana of Tunis went furthest. Elsewhere — in Aleppo, in Damascus, in Mogador — Italian and Iberian Jewish merchants settled under European protection, kept their passports, married among themselves and stayed socially apart, but they never built a separate community. In Tunis they built one and kept it for 189 years.

From 1710 the Grana had their own council of notables, their own grand rabbi, their own rabbinical court, their own synagogues, their own schools, their own butcher and their own cemetery. They called themselves — and their registers still call them — the Portuguese Jewish community of Tunis.

For a genealogist that is an extraordinary gift, because it means Grana records are physically separate from those of the indigenous community. You can frequently tell which of the two a family belonged to from the register alone. No other community on this site offers that.

Who were the Grana?

Livornese Sephardim in Tunis — a small, wealthy, self-consciously European minority among a much larger Arabic-speaking Jewish population.

The name comes from Gorna — Livorno, which Europeans called Leghorn and Arabic speakers rendered as Grana; in Hebrew, Gorneyim (גורנים).

They were the descendants of Marrano families who had returned to Judaism at Livorno in the sixteenth century, under the Medici charters that made that city the great free port of the Sephardi world. From the early seventeenth century they began moving to Tunis to work the Livorno trade, and by the 1680s they were a recognisable community.

They were a small and wealthy minority. In the eighteenth century they were perhaps eight per cent of Tunisia’s Jews — but they paid one third of the community’s collective jizya. They spoke and wrote Tuscan Italian and sometimes still Spanish, dressed as European merchants in wigs and round hats, and went straight into the beylical court as tax farmers, treasurers, physicians, financiers and diplomats.

The indigenous community — the Twansa, from Tunsi, Tunisian — spoke Judeo-Tunisian Arabic, wore the black chechia required of Jews, and were on the whole poorer. Both groups lived in the Hara, the Jewish quarter of Tunis, and they did not get on.

A Maltese doctor who reported on Tunis in 1818 caught the relationship exactly. He counted about twelve thousand Jews in the city, of whom “the first, or Leghornese, do not exceed the number of seven hundred”, governed by three elected parnassim; their liturgy “is that of the Spanish synagogue”; they wore European dress, kept a separate market, had no dealings with the Tunisian Jews, and in seven generations counted four intermarriages — a party to such a marriage being excluded from their synagogues as one who had degraded himself. He also noted that corporal punishment, routine for the Twansa, was rare for the Grana, whose penalties were “generally of a pecuniary nature”.

And one line in his report matters more than the rest. He observed that the Leghornese were “for the most part natives of Tunis, and consider themselves the descendants of the exiled families of Spain”. By 1818 the Grana were not immigrants. They were a Tunisian-born caste sustained by an origin story and a legal status.

What happened in 1710 and 1741?

The Livornese seceded, and thirty-one years later the two communities signed a treaty setting out who belonged to which.

A century of friction ended in 1710, when the Livornese community effectively seceded — with the tacit agreement of the authorities — and created its own parallel institutions across the board.

The arrangement was formalised in a takkanah signed in July 1741 between the two grand rabbis, Abraham Taïeb for the Twansa and Isaac Lumbroso for the Grana. It was renewed in 1784 and finally annulled in 1899.

The Grana nevertheless remained under a single qaid of the Jews, who was always chosen from the Twansa — presumably so that no foreign-protected figure held the office.

Why does the 1741 rule matter more than surnames?

Because it assigned families by where they arrived from, not by descent — and it is constantly misunderstood.

The rule the takkanah laid down is the single most important thing on this page:

Every Jew arriving from a Muslim country belonged to the Twansa. Every Jew arriving from a Christian country belonged to the Grana.

That is an administrative test, not a genealogical one. Three consequences follow, and every one of them trips people up:

“Grana” does not mean “of Livornese descent.” The Grana registers contain families who arrived from Pisa, Genoa, Marseille, Gibraltar, Amsterdam, Portugal and elsewhere — anywhere in Christendom. Livorno was the largest source, not the only one.
An Iberian surname does not make a family Grana. A family of Spanish descent that reached Tunis by way of Morocco or Algiers was assigned to the Twansa, and its records are in the Twansa series.
The two communities were fiscal as well as religious units. Assignment determined which rabbi, which court, which cemetery and which share of the tax — so it was recorded, and recorded consistently.

What trade were the Grana actually in?

Coral and diamonds — the same Livorno-to-India system described on our India page.

Red coral was fished off the North African coast, shipped to Livorno to be worked, and carried east to India, where it was exchanged for diamonds from the Golconda mines. The Sephardi merchant houses of Livorno held both ends, and their agents in Madras lived and worked in what was actually called Coral Merchant Street.

Tunis was a node on that route, and the Grana were the people who ran it there. Francesca Trivellato’s reconstruction of the Livorno firm of Ergas and Silvera — the same firm whose partner Isaac Ergas appears at Surat under the name Fernão Mendes Henriques on our India page, and whose name reappears among the Francos of Aleppo on our Syria page — is the best account of how the whole system worked.

The Grana also controlled the Livorno trade in general, despite paying customs at ten per cent against three for Muslim and Christian traders; ran credit-banking operations out of their trading houses; and participated in the ransoming of Christian captives taken by the corsairs.

Why are there two kinds of Grana after 1846?

Because a Tuscan amending act let later arrivals keep their Italian nationality indefinitely — which decides which archive holds a family.

A treaty with the Grand Duchy of Tuscany of 10 July 1822 limited a Grana’s stay in Tunisia to two years, after which he fell under the Bey’s sovereignty and was legally equal to the Twansa. In practice that made the existing Grana Tunisian subjects.

But an amending act of 2 November 1846 allowed Grana who settled after 1822, or who arrived subsequently, to keep their Tuscan nationality with no time limit — and those arriving after Italian unification benefited too.

So from 1846 there are two distinct Grana populations:

The old Grana — beylical subjects, living in the Hara, recorded in Tunisian and community sources.
The new Grana — Tuscan and then Italian nationals, ninety of them by 1848, living in the free European quarter of Tunis under the protection of the Tuscan consul, and recorded in Italian consular records.

If you cannot find a Grana family in the Tunisian material, this is usually why. Look in Livorno, Florence and the Italian diplomatic archives instead.

The distinction later had political weight: under the French Protectorate the administration resisted reforming the community institutions partly because it feared they would fall under the control of the Grana, who were regarded as sympathetic to Italy.

How did the split end?

From both ends at once — the Grana lost their Italian, and the French Protectorate imposed a single chief rabbinate.

By the close of the nineteenth century both communities spoke Judeo-Tunisian Arabic; the Grana had lost their Italian as a vernacular. The takkanah was annulled in 1899. And after an intermediate period the Protectorate administration recognised a single chief rabbi, from the Twansa, with authority across the whole country, which ended the institutional separation for good.

Intermarriage between Twansa and Grana became common in the twentieth century. By the time the community left, the distinction survived mainly in surnames, in family memory and in the older registers.

Why does this page stop where it does?

Because the Twansa and Djerba are large subjects in their own right, and they would be badly served as an appendix to this one.

The Twansa are the great majority of Tunisian Jewry and a two-thousand-year-old community, with a documented presence from the second century, a Roman-era necropolis at Gammarth, a synagogue mosaic at Hammam Lif, and a golden age at Kairouan in the ninth to eleventh centuries that made it one of the three great centres of the Jewish world.

Djerba is different again: an island community with its own rabbinic and printing tradition, its own liturgy, and the El Ghriba synagogue and its Lag Ba’Omer pilgrimage. It stayed outside the modernising currents that reshaped Tunis, which is why it is often described as the most traditional Jewish community in the Mediterranean.

Both deserve their own treatment, and Paul Sebag and Lucette Valensi are the places to begin.

What records can you actually reach?

The Grana registers

The core sources, and the limitation. Of ten registers of ketubot kept by the Portuguese community, three were rescued. They were edited by Robert Attal and Joseph Avivi and republished in Israel by the Ben-Zvi Institute. That is the hard fact behind everything else here: seven registers are gone.

What survives is exceptionally well served by the genealogical community:

Registres matrimoniaux de la communauté juive portugaise de Tunis, XVIIIe–XIXe — put online by Bob and Francine Cassuto.
Gilles Boulu and Alain Nedjar, La communauté juive portugaise de Tunis dite livournaise ou Grana: registres matrimoniaux 1812–1844 et 1872–1881, avec notices généalogiques — available from the Cercle de Généalogie Juive, which also publishes on the Protectorate period. The genealogical notices are what make this more than a transcription.
Alain Nedjar is at work on a book on the genealogy of the Jews of Tunisia, which will be the fullest treatment attempted and is worth waiting for.

A marriage register of this kind typically names four people per entry — groom and bride, each with a named father — which makes it one of the richest structured sources in Sephardi genealogy. Two cautions, both learned the hard way. The Tunis registers span roughly 1788 to 1881, long enough that a single name will belong to two or three different men, so never treat a name alone as an identity across the whole run. And volume markers and sequence numbers have a habit of being absorbed into surnames by transcribers; check anything that looks like a family name with a digit or a stray particle in it.

Surnames

Paul Sebag, Les noms des Juifs de Tunisie (Paris, 2002) is the Tunisian equivalent of Laredo for Morocco, and is the first book anyone working this community should obtain. It will usually tell you whether a name belongs to the Grana or Twansa stock, which is a shortcut to the right archive.

Finding aids

Thierry Samama, Sources for the Genealogy of Tunisian Jews (2020) is a dedicated finding aid and should be read before starting. Samama has also lectured for Sephardic World on Sephardi families in the diplomatic and consular archives, which is precisely the material the post-1846 Grana generated.

French Protectorate records

Archives nationales d’outre-mer (ANOM), Aix-en-Provence — the French civil records of Tunisia, 1881–1956, are now held here. Births, marriages and deaths in the standard French form.
Naturalisation files. The decree of 3 October 1910 opened French citizenship to Tunisian subjects on strict conditions, and the numbers were tiny at first — ninety-three between 1911 and 1914 — before the Morinaud laws of 1923 widened access. Each file names the applicant’s birthplace, date and qualifications.

Italian and Tuscan records

For the post-1846 Grana, and for the Livorno end of every Grana family:

Archivio di Stato di Livorno and the records of the Livorno Jewish community
Italian consular archives for Tunis, and the Tuscan diplomatic material for the period before unification
The Italian state archives in Florence

Alliance Israélite Universelle

The first AIU school in Tunisia opened in 1878, three years before the Protectorate. The archives of the AIU schools in Tunisia, 1876–1970, are held at the CAHJP in Jerusalem — a specific and accessible holding, with reports that describe the communities as well as the pupils.

Cemeteries

The Borgel cemetery in Tunis, named for the Borgel rabbinic family, is the principal burial ground and has been the subject of a documentation project. Remember that the Grana maintained a separate cemetery from 1710 — so for the eighteenth and nineteenth centuries, which cemetery a person lies in is itself evidence of which community they belonged to.

Community and press

Over twenty-five periodicals appeared in Judeo-Tunisian Arabic, mostly around the turn of the twentieth century, alongside the French-language Jewish press. Harissa.com is the long-running French-language Tunisian Jewish website and remains the community’s principal gathering point online.

Where did they go?

Overwhelmingly to France — despite the Italian roots.

Tunisia held about 105,000 Jews before independence in 1956; there were around 1,500 by 2017.

Departure came in waves: after 1948 and the Hafsia attacks of 1952; after independence, as the judiciary and administration were Arabised and Jewish-owned businesses were squeezed; after the Bizerte crisis of 1961, which alone prompted some 4,500 departures in 1962; and after the Six-Day War in 1967, when the Grand Synagogue of Tunis was burned and a further seven thousand went to France and some 2,400 to Israel.

The pattern of destination is consistent and useful. The poorer and more traditional communities — the south, Djerba, the provincial towns — went to Israel. The educated, French-speaking urban class went to France, and the Grana, being both, went overwhelmingly to France; only a few settled in Italy. Paris (Belleville, and later Sarcelles), Marseille and Nice hold the largest concentrations, with a further community in Montreal.

Sephardic World lectures

Thierry Samama has lectured on Sephardi families in the diplomatic and consular archives — the exact material the post-1846 Italian-national Grana generated. Our video pages also carry lectures on Livorno, which is the other half of this page, and on the Livorno–India coral and diamond trade.

What has and hasn’t been researched

Seven of the ten ketubot registers are lost, and nobody knows whether copies, extracts or notes were made before they went. Somebody should ask.
The Livorno end has not been joined to the Tunis end. Both communities kept good records; the Grana are the link between them; and no systematic matching of Livorno families to Tunis families has been done. This is the largest opportunity here, and it needs someone comfortable in two archives.
The post-1846 Italian-national Grana are under-researched, and their consular records are accessible.
The provincial Grana — there were Livornese families at Sousse, Sfax, La Goulette and elsewhere — are barely studied at all.
The 1741 takkanah itself deserves a proper genealogical reading. It is the document that assigned families to one community or the other, and it, and the disputes that followed it, will name people.
Nobody has compared the Grana with the Francos of Aleppo, though they are the same phenomenon at different degrees of institutionalisation and in several cases involve the same families.

Organisations and repositories

Ben-Zvi Institute, Jerusalem — the surviving ketubot registers.

Central Archives for the History of the Jewish People, Jerusalem — the AIU school archives for Tunisia, 1876–1970.

Archives nationales d’outre-mer, Aix-en-Provence — French Protectorate civil records.

Archivio di Stato di Livorno and the Livorno Jewish community; Italian consular and diplomatic archives.

Cercle de Généalogie Juive, Paris — publications and the North African group.

Alliance Israélite Universelle, Paris.

Harissa.com; the Society for the History of the Jews of Tunisia (founded Paris, 1997).

Help us find records

The Grana are unusually well served by comparison with most of this site, so what is missing is specific:

Any trace of the seven lost ketubot registers — copies, extracts, photographs or notes taken before they disappeared
Livorno records naming families who went to Tunis, and the reverse
Italian consular registrations of Grana families after 1846
Records of the Grana cemetery before its records were merged
Family papers documenting the Livorno-to-Tunis move, and protection or nationality patents of any kind
Material on the provincial Livornese families at Sousse, Sfax and La Goulette

If you hold any of this, or know where it is, please email [email protected]. Material contributed is credited, and we are happy to work under any conditions you set.

Research services

Grana research is Italian research with a Tunisian address, and it runs in three archives: the surviving Portuguese community registers, the French Protectorate civil records at Aix, and the Livornese and Italian consular material. The first question is always whether the family was Grana or Twansa, because the answer sends you to different records — and the second is whether they held Italian nationality after 1846, because that sends you to a different country. We will tell you honestly which you are facing. If you would like help, including for a Portuguese or Spanish nationality application, you can hire a professional Sephardic genealogist.

If you have found this page useful, please consider making a small donation to support this site.

On this page