Most 20th-century immigrants to the US kept their last names, Social Security records show

Social Security applications from 1936 to 1970 show that more than 80% of immigrants from 11 countries kept their surnames, and only about 5% switched to an English-sounding name that hid their origins.

Share this:

The image of the immigrant who arrived in America and swapped a foreign surname for an English one is mostly a myth. A study of Social Security applications filed between 1936 and 1970 finds that more than 80% of immigrants from 11 countries kept their last names, and only about 5% changed to an English-sounding name that hid where they came from.

The study, titled “How 20th-century immigrants did not have to change their last names in order to become American,” was written by Joshua R. Goldstein of the University of California, Berkeley, and Guy Stecklov of the University of British Columbia. It was published in the Proceedings of the National Academy of Sciences on November 3, 2026.

A record that shows both names

Famous name changes are easy to list. Marilyn Monroe was born Norma Jeane Mortenson and Kirk Douglas was born Issur Danielovitch. But earlier research on the subject mostly looked at people who did change their names, such as court petitions in New York City, and left out everyone who kept theirs. Counting how common changes really were has been hard because most name changes in the US never went through a court.

Goldstein and Stecklov used a source that records both names in one place. When workers applied for a Social Security number, the form asked for their own name and the names of both parents. For an immigrant man, a last name that differs from his father’s is evidence of a change. The researchers studied about 172,000 men born in Italy, Germany, Poland, Russia, Greece, China, Hungary, Austria, Egypt, Japan and Iran, drawn from the Berkeley Unified Numident Mortality Database.

They then used name-frequency comparisons with the 1851 British census, plus a letter-pattern method from computational linguistics, to sort each change into two kinds. An ethnic change kept a recognizably foreign name, such as Papadopoulos becoming Pappas. An English change hid the origin, such as Papadopoulos becoming Pope.

Simplified, not disguised

More than 80% of the men had the same surname as their father. About 15% made an ethnic change, often a shorter or simpler spelling. Only about 5% took an English-sounding name. That figure already includes an upward adjustment for fathers and sons who may have changed their names together.

Even those who changed usually kept a link to the old name: the first letter, a syllable or a translation. Rosenberg, for example, more often became Rosen or Berg than Ross.

Italian, Chinese and Japanese immigrants had the lowest rates of English changes. Hungary, Russia, Poland and Austria, origins with many Jewish immigrants, had the highest. Among men whose fathers carried one of 36 distinctive Jewish surnames, such as Cohen, Goldstein or Shapiro, 88.3% kept the name, 5.2% changed to another ethnic name and 6.5% switched to an English one.

Where and why names changed

English changes were rarest in the Northeast. In New England only 3.2% of immigrants hid the origin of their name, while in the Deep South the rate was more than twice as high.

The authors propose a framework in which an immigrant weighs the possible economic gain of passing as unmarked against the loss of belonging to an ethnic community and the daily hassle of an unusual name. The data fit it. Men living among more people from their own origin changed less. Men in wealthier ZIP codes changed slightly more. A long or hard-to-spell surname, measured by its Scrabble score, raised the chance of an English change by about 1.7 percentage points across the middle range of names. Japanese and Chinese immigrants rarely changed, which fits the idea that a new name could not hide a visible difference.

Why it matters

The results challenge the long-standing story that immigrants of this era had to shed their foreign identities, or “become White” through an English surname, to join the American mainstream. The authors argue that the ethnic revival of the 1970s was less a recovery of lost identities than a rediscovery of names that had been in plain sight all along.

There is also a practical payoff for researchers. Because most immigrants kept their names, linking historical census and administrative records across generations should work better than many had feared.

The study has limits. It covers men only, since women usually changed their surname at marriage. Most of the men were born after 1900 and died between 1988 and 2005, so earlier arrivals are underrepresented, and those born before 1900 were slightly more likely to adopt an English name. The links between neighborhood, income and name change are correlations, not proof of cause.

Study Details:

  • Title: How 20th-century immigrants did not have to change their last names in order to become American
  • Authors: Joshua R. Goldstein and Guy Stecklov
  • Journal: Proceedings of the National Academy of Sciences (vol. 123, no. 45)
  • Publication Date: November 3, 2026
  • DOI: 10.1073/pnas.2614209123