Sephardic surnames are the subject we are asked about more than any other. This page sets out what a surname can and cannot establish.
The short answer is that a surname tells you where to look. It does not tell you who you are.
This is not an evasion. It is the answer, and the rest of this page explains why, and what to do instead.
Why Iberian surnames cannot carry this weight
There are five reasons, and they compound.
1. Iberian surnames were not fixed until about 1800
This point is not widely known, and it undermines the rest on its own.
In Spain and Portugal it is usual to carry two surnames, one from each parent. In the Spanish tradition the father’s comes first, in the Portuguese the mother’s. But which name was inherited from which parent was not settled until roughly the late eighteenth century.
Before then, a child might take a surname from a grandparent, from a godparent, from a place, or from a nickname. Siblings could carry different surnames. Children frequently carried different surnames from their parents. Women often took a husband’s name on marriage. People used different names in different jurisdictions. When a family moved between Spanish and Portuguese territory, the order of the two names did not always change with it.
So if you know the surnames of an ancestor alive in 1800, you cannot safely infer that either name descends from a particular ancestor alive in 1500. That is true whether the family was Jewish, converso or Old Christian. It is fatal to any argument running from a modern name back to the fifteenth century.
2. Jews in Iberia used the same names as everyone else
There was no separate Jewish naming system in medieval Spain and Portugal. Jewish families formed surnames as their neighbours did: toponymic (Toledano, Laredo, Espinosa, Curiel), patronymic (Rodrigues, Fernandes, Nunes, Lopes), occupational, and descriptive (Delgado, Moreno, Bueno, Franco).
Every one of these was borne by large numbers of Christians who had no Jewish ancestor. Toledano means “from Toledo”, and most people from Toledo were not Jews.
3. Converts took Christian names, probably their godparents’
At conversion — in 1391, in 1492, and in the Portuguese forced conversion of 1497 — converts took Christian names. The evidence suggests these were commonly the names of the godparent or patron who sponsored the baptism. That is how converso families came to be called Santa María, de la Cruz, Álvarez and Rodríguez.
The consequence inverts the exercise. A distinctively Old Christian name is often the mark of a converso family. A name that sounds Jewish may belong to a family that never converted, because they left instead and kept what they had.
There is a persistent belief in Spain and Portugal that Jews adopted surnames from trees and fruit. We have seen no evidence for it.
4. The Inquisition did not rely on people’s surnames
David Gitlitz, whose Secrecy and Deceit is the standard catalogue of crypto-Jewish practice compiled from the Inquisition’s own files, asked why certain family names are held to indicate Jewishness when the Inquisition did not consider them so.
The Inquisition spent two and a half centuries identifying Judaizers. It was expert, obsessive, well-resourced and, being funded from confiscations, financially motivated to convict. If surnames had worked, it would have used them.
5. Names split, merged and moved
Lopes and López. Rodrigues and Rodríguez. Nunes and Núñez. Gomes and Gómez. Dias and Díaz. The same family appears in Portuguese, Spanish, Dutch, Italian, English and Hebrew forms across four archives.
Two families with the same name are usually unrelated. One family with four spellings is not four families. Neither is visible from a name.
Where the Sephardic surname claim came from
The idea is recent and traceable.
It was popularised by a well-meaning but confused Ashkenazi amateur genealogist in the late 1990s, in the early days of the internet. The argument was that if a New Christian or a Jew is recorded using a surname, then the surname is Jewish, and people bearing it today are therefore of Jewish descent. Catholics using the same surname were not counted.
It was widely republished from about 2001, amplified by prominent genealogy blogs, and it has proved persistent.
The flaw is a base-rate fallacy. The reasoning takes a name attested among Jews and treats it as diagnostic of Jews, without asking how many non-Jews bore it. The figures below show how badly this fails.
Three lists, one set of names
The Inquisition’s list
The commonest surnames among people accused of Judaizing in Portuguese Inquisition records, as reported:
| Rodrigues | 453 |
| Lopes | 282 |
| Nunes | 229 |
| Mendes | 224 |
| Miranda | 190 |
| Gomes | 184 |
| Henriques | 174 |
| Costa | 138 |
| Fernandes | 132 |
| Pereira | 124 |
| Dias | 124 |
The congregational list
Separately, we have extracted surnames from the records of identified Sephardi congregations: Amsterdam, London, Livorno, Bordeaux, Hamburg, Curaçao, Jamaica, Suriname, Barbados, St Thomas, Charleston, New York, Tunis, Vienna and others. Ranked by the number of communities in which each appears, the top of that list runs: Henriques, Mendes, Da Costa, Rodrigues, Lopes, Nunes, Baruch, Jessurun, De Leon, Fernandes, Pinto, Pereira.
The Portuguese telephone book
Compare both with the commonest surnames in Portugal, Brazil, Angola, Mozambique and Cape Verde today: Silva, Santos, Ferreira, Pereira, Oliveira, Costa, Rodrigues, Martins, Jesus, Sousa, Fernandes, Gonçalves, Gomes, Lopes, Marques, Alves, Almeida, Ribeiro, Pinto, Carvalho, Teixeira, Moreira, Correia, Mendes, Nunes.
They are the same names. The people accused by the Inquisition, the people in the Sephardi congregations, and the general Portuguese population all carry the same surnames, in roughly the same order. They were all drawing on the same stock of Portuguese surnames.
Is there an official list of Sephardic surnames?
No. And it has been claimed on various websites that all 100 of the 100 commonest Portuguese surnames are Jewish, and 99 of the 100 commonest Spanish surnames.
Followed through, this produces a dilemma. Either the concept of the Sephardic surname is meaningless, or there were almost no Christians in Spain and Portugal.
The Sephardic Genealogical Society’s Code of Conduct states the position directly. Surname reports are studies showing that surnames in an applicant’s family were previously used by Sephardic Jews or New Christians. In the absence of genealogy, such reports can be used to imply a familial connection between unconnected individuals of the same surname, even when divided by centuries and continents. The Code concludes that surname reports have no place in Sephardic genealogy, and provides that clients who have been sold them should be offered full refunds.
Common misconceptions
There is an official Spanish list of Sephardic surnames. There is not. Lists circulated claiming to be accepted by the Spanish government under its former citizenship scheme. This was a hoax.
Portuguese law recognises Sephardic surnames. Portuguese nationality legislation has referred to surnames among other indications. A surname alone is not evidence, and a state processing an application is not a state certifying descent. See our page on Spanish and Portuguese nationality.
Our family has a coat of arms, so we can trace the line. A coat of arms belongs to a family, not to a surname, and anyone sharing the name has no claim to it. There is a specific trap here: some Sephardi families in Amsterdam adopted the heraldry of unrelated Spanish aristocratic houses whose names they happened to share. An Iberian coat of arms in a Sephardi family may be an aspiration rather than a descent.
The name is unusual, so it must mean something. Rarity is not distinctiveness. A rare name may be rare because it was formed once, in one village, by one Christian family.
There is no serious scholarship on Sephardic surnames. There is. Alexander Beider has published two substantial scholarly dictionaries covering Sephardi names, and is at work on a third. What that scholarship shows, however, is not that surnames prove descent — it is the opposite. See below.
The alias problem
This is missed almost universally, and it matters more than any list.
Western Sephardi merchants routinely used two names: a commercial or civil name, and a Hebrew or synagogue name. Both are real, both are documented, and they appear in different bodies of record.
- Jaques de Paiva, the first Jew of Madras, was Moses Zagache
- Domingo do Porto was Abraham do Porto
- Fernão Mendes Henriques, of the Livorno firm trading coral to India, was Isaac Ergas
- Gomez Rodrigues of London was Abraham Israel de Sequeira
- Mosseh Pereyra de Paiva, author of the 1686 account of the Jews of Cochin, was Pedro Pereyra of Surat
Commercial records use one name and congregational records the other. A researcher who knows only one will conclude that the person does not appear in the other archive.
Váez, Báez, Vaz and Vázquez are likewise one surname, used interchangeably by the same Inquisition scribes within a single file.
Five regional naming regimes
Sephardi surname is not one category. It is at least five, and confusing them is the commonest analytical error in the field.
Western Sephardi — Amsterdam, London, Hamburg, Bordeaux, Bayonne, Livorno, the Caribbean and the Americas. Overwhelmingly Portuguese rather than Spanish, and drawn from the commonest Portuguese name stock.
Eastern Sephardi — the Ottoman world. Iberian names survive (Franco, Pardo, Sarfati, Benveniste) alongside Hebrew, Greek, Italian and Ottoman formations: Behar, Alhadeff, Capon, Saltiel, Romano. Eskenazi is a common Sephardi surname in the Ottoman lands.
Maghrebi — Morocco, Algeria, Tunisia, Libya. Two stocks side by side: Iberian Megorashim names (Toledano, Corcos, Pinto, Almosnino, Benattar), and indigenous Arabic and Berber names, many toponymic (Gerbi from Djerba, Tajouri from Tajura, Chetrit, Abitbol, Meghnagi). Both are Sephardi communally; only the first is Iberian by descent.
Levantine — Aleppo, Damascus, Beirut. Yedid, Sutton, Dwek, Antebi, Tawil, Safdieh, mostly not Iberian.
Italkim — the Italian Jewish communities, neither Sephardi nor Ashkenazi: Calò, Di Segni, Piperno, Sonnino, Ascoli, Ancona, Morpurgo, Finzi, Bondi, Meldola, Treves. These circulate on Sephardi lists because Italian and Sephardi families lived together at Livorno, Venice and Rome. The names are Italian.
Laniado defeats every method based on the shape of a name. It is the great Sephardi rabbinic dynasty of Aleppo, which supplied five consecutive chief rabbis, and the name is not Iberian. A Sephardi family with a non-Iberian surname is common.
Rigorous onomastics: the work of Alexander Beider
Alexander Beider is the leading authority on Jewish onomastics. He trained as a mathematician — a doctorate in applied mathematics from the Moscow Institute of Physics and Technology — and took a second doctorate in Jewish studies at the Sorbonne in 1999.
His Ashkenazi work is the standard reference: dictionaries of Jewish surnames from the Russian Empire, the Kingdom of Poland, Galicia and Prague, a dictionary of Ashkenazic given names, and Origins of Yiddish Dialects (Oxford, 2015).
He has since turned to the Sephardi and Mediterranean world, and has published two dictionaries covering it:
- A Dictionary of Jewish Surnames from Maghreb, Gibraltar and Malta (Avotaynu, 2017)
- A Dictionary of Jewish Surnames from Italy, France and “Portuguese” Communities (Avotaynu, 2019)
The second of these is the one that matters most for readers of this site. Its full subtitle is including surnames of Jews from continental Italy, non-Ashkenazic communities in France, and Sephardic communities in Western Europe (after the 1490s) and the Americas. In 882 pages it treats more than 3,150 root surnames, from which nearly 10,000 variants are derived, drawing on over 560 bibliographic sources. It covers the “Portuguese” communities of Amsterdam, Hamburg, London and Bordeaux, and the Caribbean and American congregations. Each entry gives the attested spellings and their sources, the community group, the etymology, and the regional variants.
Beider is currently at work on a third volume, expected to cover the Eastern Sephardim of the Ottoman world, and possibly other communities including the Mizrahim. His recent papers — on the Italian Jews of the Ottoman Empire, on the Karaites, and on the names and origins of the grâna community of Tunis — point in that direction.
Between them, these volumes are the only rigorous, source-referenced reference works covering Sephardi surnames as a whole. Anyone doing serious work on these names should use them.
What Beider’s method actually establishes
Beider’s dictionaries are indispensable, and they are also routinely misused. It is worth being precise about what they do.
His method rests on six principles, and every one of them is violated by the surname claims described earlier on this page.
- Primary sources only. Actual documents naming actual people in specific places at specific dates: censuses, tax lists, community registers, notarial acts. Never secondary compilations or circulating lists.
- Dated, localised attestation. For every name, where and when it is first documented.
- The same name does not mean the same family. Names were formed independently in many places. Beider emphasises this repeatedly, and it is the point most often ignored.
- Linguistic rigour. Etymologies tested against actual phonology, orthography and dialect, which eliminates most folk etymology.
- The name’s origin is not the bearer’s origin. A German-derived surname does not imply German descent, and an Iberian surname does not imply Iberian descent.
- Establish the moment of adoption. When, and under what circumstances, a family came to bear the name.
A dictionary built on these principles tells you that a given name is attested among Jews in a particular place at a particular date, what language it came from, and how its spellings relate to one another. That is exactly what a researcher needs in order to know which archive to open and which variants to search.
It does not tell you that a living person bearing that name is descended from those people. Beider’s third principle rules that inference out directly. His own work on the grâna of Tunis is a good illustration: about half of that community’s surnames are Iberian names inherited from former Portuguese and Spanish New Christians, and the other half are not — the name stock is mixed, and no single name settles the question for a family.
Why the Iberian case is harder than the Ashkenazi one
Beider’s Ashkenazi dictionaries are unusually powerful because of a historical accident. Most Ashkenazi surnames were adopted in a narrow window, roughly the 1780s to the 1850s, under state compulsion: Austria in 1787, Prussia in 1812, the Russian Empire from 1804 and 1835, Poland in 1821. The process was administrative, documented and localised. A name can often be dated, placed, and tied to the moment a specific family in a specific town was required to choose one.
Nothing like that happened in Iberia. Jewish surnames there were formed early, gradually, and inside the same system as everyone else’s, so they were never distinctive in the first place. The one event that might have made them distinctive — mass conversion — did the opposite, attaching Old Christian names to converso families. By the time Iberian surnames became fixed and heritable, around 1800, the Jewish and Christian name stocks had been indistinguishable for three centuries.
This is why Beider’s Sephardi volumes are structured as they are. They record where and when names are attested and what they derive from. They are not, and do not claim to be, lists of names that identify their bearers as Sephardi. The existence of a rigorous dictionary is not a licence for the surname report; read properly, it is the clearest demonstration of why the surname report does not work.
What to do instead
Use the surname as a lead. A name attested in a known Sephardi community is a good reason to open a particular archive. That is how professional research begins.
Then test it. The hierarchy of evidence, in order:
- A documented line to a named individual in a Sephardi congregational record — a circumcision, marriage or burial register, a ketubah, a community tax list. This is what establishes descent.
- A named individual identified as a New Christian in a contemporary record — an Inquisition file, a limpieza de sangre investigation, a notarial act.
- Y-DNA placed in a known Sephardi lineage by someone competent to do it. See our page on DNA and Sephardic ancestry.
- A surname, as a pointer to where to look for the first three.
What does not work, in any combination: a surname on a list; a surname plus a family tradition; a surname plus an ethnicity percentage; a surname report. Four weak indicators do not make one strong one.
Practical advice
- Join your local genealogical society. They teach method, and the important parts are free.
- Follow the evidence where it leads, including where it leads away from what you hoped.
- Consider historical and geographical context before drawing conclusions from a name.
- Rely on archival records rather than internet claims.
- Start with our guides to getting started and genealogical standards.
Sephardic World lectures on surnames
- Jews in the Maghreb: their surnames and their roots — Alexander Beider
- The Story of Sephardic Surnames: fruit trees, Catholic themes and expert lists — Michael Waas
- Cardinal Errors in Sephardic Genealogy — Julian Land
Reference works
Alexander Beider, A Dictionary of Jewish Surnames from Italy, France and “Portuguese” Communities (Avotaynu, 2019). The single most important reference for Western Sephardi names, covering the “Portuguese” communities of Amsterdam, Hamburg, London and Bordeaux, the Caribbean and the Americas, alongside continental Italy and non-Ashkenazi France.
Alexander Beider, A Dictionary of Jewish Surnames from Maghreb, Gibraltar and Malta (Avotaynu, 2017). The companion volume for North Africa, Gibraltar and Malta.
Abraham I. Laredo, Les noms des Juifs du Maroc (Madrid, 1978), the outstanding regional work, giving the probable origin of each name and its notable bearers. Joseph Toledano’s later dictionary supplements it.
Paul Sebag, Les noms des Juifs de Tunisie (Paris, 2002), the Tunisian equivalent.
Guilherme Faiguenboim, Paulo Valadares and Anna Rosa Campagnano, Dictionary of Sephardic Surnames, the largest compilation and the one most widely used. It is best treated as an index of where names occur rather than as a determination of who is Sephardi.
Avotaynu’s Consolidated Jewish Surname Index, an index of indexes covering some 699,000 surnames across 42 databases. It records which published sources mention a name, makes no claim about what the name means, and is useful for generating spelling variants when searching material recovered by optical character recognition.
SephardicGen, the work of Jeff Malka, and Mathilde Tagger’s indexes.
Professional help
We do not undertake speculative research. If you have a solid reason to believe you have Sephardi ancestry — an ancestor from a known Jewish community, a documented New Christian, a family in a place where records can be searched — please get in touch.
If all you have is a surname, we will say so, and explain what would need to be true for it to lead anywhere. That conversation is free.
You can also email us at [email protected].
If you have found this page useful, please consider making a small donation to support this site.
See also: Crypto-Judaism, New Christians and Bnei Anusim · Genealogical Standards · DNA and Sephardic Ancestry