Most Common Jewish Names

I present here an analysis of the distribution of Jewish and non-Jewish names among immigrants who arrived at Ellis Island at the beginning of the twentieth century. It is a by-product of my work on the Jewish immigrants from the Pale of Settlement. The first sections explain a few technical details about the data. It is followed by the lists of most common Jewish and non-Jewish names and a few comments.

Jew or Not: First Name or Family Name?

Today, one would normally think that knowing the family name would be more informative than knowing the first name, in trying to guess whether a person is Jewish or not. Indeed, some first names sound more likely to be Jewish than others, such as Rachel or Rebecca for females, and Abraham and Isaac for males. Other names are still quite definitely non-Jewish. Try to think of anyone called Christian – I bet the last Jew who had the syllable “Christ” in his name lived a couple of millennia ago, probably in Jerusalem. But overall, first names normally do not tell us that much. If you need to guess, you’ll try using the family name first.

But this wasn’t always the case; it is a result of over a century of cultural assimilation of Jews within the American society, and of the fact that the distribution of family names is somewhat more persistent than that of first names. Before crossing the Atlantic, Jews who lived in the Pale of Settlement had remarkably different distributions of both first names and family names. In fact, I found that the first names made a better indicator of Jewish ethnicity.

In the project I’m working on I try to identify the places of origin of Jewish immigrants who came to the U.S. from the Russian Empire. But before following the requirement to know from where you came, I have to answer what may be the most existential question of all: who is a Jew?

To do that, I’m using the distribution of names of immigrants who came aboard ships arriving in Ellis Island. I will be explaining how I do that in more details elsewhere, but the important things to keep in mind are the following:

Identification of Jews on Ship Manifests

Since about 1899, Jews begun to be identified on the ship manifests submitted by the shipping companies to the U.S. authorities not only by their country of origin, but also by their ethnicity, with Hebrew being an officially recognized category. But not all Jewish passengers were identified as Jews. It turns out (and I can’t show this here) that the accuracy of the identification of Jewish passengers went as follows:

• Some ships did not identify Jews at all, even when there were Jews aboard
• Most ships identified Jews, and did that very accurately
• Some ships identified Jews, but did that poorly
• Even when identification was poor, very few non-Jews were mistakenly misclassified (or later coded on the computer file) as Jews

In short, among immigrants from the Russian Empire after 1899, excluding ships that did not identify Jews at all, approximately 90-95 percent of Jews and 0.5-1 percent of non-Jews were identified as Jews. This leaves many unclassified Jews – some of them came aboard ships that did not bother to identify Jews at all; others were aboard ships that had worked out a poor classification; and a few slipped through the net for various other reasons.

Now, since I know the immigrants’ names, I can use the names distribution of Jews and non-Jews to help me identify the remaining mass of unclassified passengers. Keeping the details for elsewhere, it is important for me to know how “Jewish” were each first and last name. The greater was the proportion of immigrants carrying a certain name that were identified as Jews, the more Jewish this name was.

Sample Definition

The sample includes all passengers who passed through Ellis Island, were coded onto the file as coming from the Russian Empire, and that the ship in which they came had reported at least one Jew aboard. In all, 788,666 immigrants aboard those ships were tagged as Jews and 979,713 were not. Almost all passengers in this sample have arrived between the years 1898-1924.

 

 

Top Jewish First Names: Males

The table above shows the most common first name among Jewish males. The columns report the following statistics:

• Frac: Fraction of identified Jews that were carrying the name
• Cum: Cumulative fraction of identified Jews  that were carrying any of the names up to the current rank
• Freq:  Number of identified Jews that were carrying the name
•  %Jews: Proportion of identified  Jews among passengers carrying the name

Abram tops the list, with 14,730 identified Jewish passengers which make up 3.6 percent of all identified Jewish males. Note that each name is in fact a particular variant of a name, and so we see that Abraham, a different variant of the same name, appears again as the ninth most common name. Next comes Chaim (read Kha’yim in Hebrew), which was given to 2.9 percent of identified Jewish Males.

Probably the most striking thing to note is how common were common names. The top 20 names covered more than 30 percent of Jewish males! As shown below, that was far from being the case for last names. The common repetitions of the same basic names in different variants make this even more striking: just within the top 20 there are 5 Hebrew-Yiddish names that repeat in two different variants: Abraham (Abram), Jacob (Jankel), Moses (Moische), Josef (Jossel), and Samuel (Schmul).

Another noticeable feature is that most of these top names were given exclusively to Jews. Consider names such as Jankel and Moische. Within the sample, which includes only ships that had identified at least one Jew, 94 percent of passengers carrying these names were identified as Jewish. However, it is absolutely certain that every single person who carried such a name was Jewish. How do I know it? Well, I just know it… Other uniquely Jewish names show the same pattern – approximately 5-10 percent of persons carrying them were not identified as Jews. It seems reasonable to deduce that this was the lower bound of the rate of false-negative error, and that any Jew in the sample had at least this likelihood not to be identified as Jewish. This means that when a name is over 90 percent Jewish within ships that identified at least a single Jew, it should be read as an exclusively, or almost exclusively Jewish name.

Keeping that in mind, note that most of the names on the list are around 90 percent Jewish and above. This means that the most common Jewish names were almost perfectly Jewish. It sounds like a tautology, but it isn’t. It could have been the case that distinguishable Jewish names were not too common, and that most Jews opted for relatively more neutral names. This list, and other evidence that I can’t show here, indicate that the preference towards unmistakably Jewish names was overwhelming at that time.

But there are a few exceptions to this rule. It appears that Josef was a very common Polish name: more than 1 in 20 passengers who were identified as Polish were written as Josef. So although it was also a common Jewish name, only 15 percent of Josefs were identified as Jewish. The same, though not as strongly, applies for Jacob (54 percent), and David (85 percent), a fairly common Polish name, is also somewhat mixed.

Finally, it seems that the list is mostly comprised of proper names. Either proper Hebrew-Yiddish names (that is to say, names that are Hebrew or Yiddish of Hebrew origin) transliterated to English or German (e.g., Abraham, Chaim, David, Israel, Moses), or proper German-Yiddish names (Hersch, Leib, Wolf). Yiddish variants and diminutives of proper names (Jankel, Jossel, Leiser, Mendel, Moische, Schmul) are not yet the majority. This balance is broken as we look down at less common names. After all, there are only a few proper ways to write a proper name, but the number of ways in which proper names could be Yiddishized or turned into diminutives is infinite. Therefore, non-standard variants and Yiddish diminutive names dominate the rest of the list, and the proper names are only frequent at the top. One might suspect that even the passengers that reported names that were written as transliterations of proper names were actually going by some Yiddish variant of that name, and that they reported a proper transliteration only to make the registration easier on the clerk, or because that was how their names were written in their documents.

Top Jewish First Names: Females

All the patterns that were pointed above in regards to male given names also applied to females. The 20 most frequent names captured 25 percent of the total identified Jewesses, and the prevalence of repeated variants of the same name is even more striking: Within the top 20 we find 4 variants of the name Sara, 3 of Chaia, 3 of Rachel, and 2 of Chana. As with males, only part of the top names were Yiddish diminutives. There were some properly transliterated Hebrew-Yiddish names (such as Ester, Rachel, and Sarah/Sara), and proper Yiddish names of non-Hebrew origin (Feige, Beile, Rosa, and Golde).

Rosa, which is of a non-Hebrew origin, was clearly shared with non-Jews (80.6 percent Jewish), and to a much lesser degree probably also Sara and Sarah. But for the most part, the most frequent Jewish female names were overwhelmingly and unmistakably Jewish.

Top non-Jewish First Names: Males

The mirror image of the previous patterns can be seen when looking at the most frequent names among passengers who were not identified as Jews. First, note that most of the top names were typically Polish (such as Adam, Michal, Franciszek, and Kasimir), Russian or general slavic (Iwan, Piotr, Boleslaw, Wladislaw), or German or general European (Jan, Franz, Jacob, Stefan, Anton). This pattern reflects the ethnic composition of the non-Jewish migration from the Russian Empire.

Second, except for the two biblical names Josef/Jozef and Jacob, it seems that Jews took great care to avoid names that were common among non-Jews. Even the pre-Christian name Alexander, that became so common among Soviet-era Jews, was only 3.5 percent Jewish at the period of the great migration. And indeed, there’s little wonder that some reluctance was exercised in sharing children’s names with the notoriously Judophobic Tsar Alexader III.

Third, common non-Jewish names were very common, even more so than among Jews: the top 20 names cover almost 38 percent of the total population of males that were not identified as Jews.

Additionally, the distribution informs us about the probabilities that a non-Jew would have been mistakenly identified as a Jew in the data. Distinctively Christian names that did not go back to the old testament, such as Franciszek, Iwan, Pawel and Piotr, can be seen as the least likely to be given to Jews. Indeed, around 1% of the passengers reporting them were identified as Jews. Even if we take the radical assumption that no Jews were named by these names, we find that a false-positive identification (i.e., identifying non-Jew as a Jew) was very unlikely; no more than one in a hundred non-Jews was mistakenly identified as a Jew.

Finally, we can infer that names that were significantly above 1 percent Jewish were indeed shared by some Jews. Among the non-Old Testament names, only Alexander comes to mind, although as we’ll see below that was more common among females.

Top non-Jewish First Names: Females

Looking at the most common non-Jewish feminine first names, most of the patterns follow. The top 20 names cover a third of the population, and the names are mostly given in Polish, other Slavic languages, or German. However, the sharing of those names by Jewish females is somewhat more common than with male names. This is true for non-Yiddish or Hebrew variants of Old Testament names (Anna, Maria), and other names that have a lesser Christian identification (Helena, Rosalia, Sofia). An exception to the rule is Paulina (6.8 percent Jewish), which may have been coupled with the similar sounding Hebrew name Peninnah.

Top Jewish Family Names

No, Cohen was not the most common Jewish last name (unless it is bundled up together with its two frequent variants, Kagan and Kohn). But otherwise, the composition of top 20 Jewish family names brings no surprises.

As with the given names, the most common variants are all properly, or close to properly written in German. The suffix –stein is written in German, and not according to other possible Yiddish or English variants such as shtein, stin or stine. This shows in the common writing of the name Lewin, in which the letter w stands for the sound v as in German. This probably follows from the fact that most of the Jewish immigrants have embarked from German ports, such as Hamburg and Bremen, and thus the lists compiled by the shipping companies were written by German speaking officers

Indeed, 19 names out of the top 20 are either of Hebrew-Yiddish, or German-Yiddish origin, the only exception being Gordon which is of a Slavic origin. Six names are indicators of priestly lineage (Lewin, Kaplan, Katz, and the three variants of Cohen). Twelve other names are of German-Yiddish origin. Out of them, eight are associated with nature; namely mountains, stones, and colors (Bernstein, Epstein, Finkelstein, Goldberg, Goldstein, Rosenberg, Schwartz, and Weinstein), and the other four are occupations or personal traits (Feldman, Friedman/Friedmann, and Schneider).

Interestingly, there are two types of common surnames that are under-represented in the top list. Only two names indicate a geographic place (Gordon and Schapiro) and there are no patronymic names (as would be, for example, Abramovich). 

As with the first names, the most common Jewish names are overwhelmingly Jewish, around 90-95 percent, and as discussed above, given the inaccuracies in the data this actually means almost 100 percent. The only exceptions to this rule are the German-Yiddish names Schneider and Schwartz.

A major difference compared to the lists of top first names is that the most common given names do not cover such a great proportion of the population. The top 20 names cover less that 6 percent of the Jewish population, as opposed to the 25-30 percent coverage of the top 20 first names. It turns out that the distribution of family names is much more spread out than that of first names, and this difference still stands as we move down towards the less frequent names.

Additionally, both first and last names can be written in several variants, but first names are prone to diminutions and transformations to nicknames, whereas last names are not. This further highlights the difference between Jewish first and last names: Most of the population is associated with a relatively small number of first names, and a much larger number of last names. While both first and last names appear to be good identifiers, in the sense that there’s little overlap with non-Jewish names, this feature makes first names the primary and most efficient tool for identifying Jews at that period.

Top non-Jewish Family Names

Among the most common last names of non-Jews there are five German names, all indicating professions (Miller/Muller, Schmidt, Schneider, Schulz), and almost all the rest are Slavic names indicating professions or geographic locations.

Interestingly, non-Jewish family names were far more sparsely distributed than Jewish names. The top 20 names captured only 1.4 percent of the non-Jewish population, as opposed to 5.7 percent of the Jewish population. As we saw above, no such difference existed for first names. This pattern follows also to the less common family names, and it is fair to make the following generalizations:

• Both Jews and non Jews have had a similarly restricted set of first names to pick from
• For both Jews and non-Jews, the distribution of family names was more sparse than that of first names
• The distribution of non-Jewish family names was more sparse than the that of Jewish family names

The separation of Jews from non-Jewish names still shows up, but it is not as sharp as in the distribution of first names. The most problematic cases are the German professional names mentioned above, where Schneider and Miller are in fact mostly Jewish. But even the Slavic names are not quite clearly non-Jewish, and it is obvious that almost each of these names was carried by a substantial number of Jews. This strengthens the insight that last names are not as efficient as first names in telling who’s a Jew and who is not.

Further Reading and Resources

For details on how I use the names distributions to identify Jews aboard the ships, see the Who-is-a-Jew Algorithm.

Alexander Beider is probably the foremost scholar of the history of Jewish Ashkenazy names. If you’re interested in this topic I strongly recommend reading his article on Names and Naming in the  YIVO Encyclopedia of Jews in Eastern Europe. A series of books he has written on the topic, as well as a few books published by other writers are referenced at the end of the article.

To browse and search the Ellis Island data that was used above, go to the Ellis Island website, or better, use Steve Morse’s terrific external search engine for the Ellis Island data. Searching and browsing the data is free, but seeing the scans of the original manifests requires a paid user account.

Acknowledgements

The data that was used in this post was kindly given to me for research purpose by the Statue of Liberty Foundation.