When people came to the U.S. from Europe, Asia and the Middle East in the mid-20th century, they faced a question that would define them: Should they keep their foreign-sounding surnames or change them to blend into their new community?
Some opted for change: Schmidts switched to Smiths. Carusos became Carrs. A Rosenberg rebranded as a Ross, while a Papadopoulos pivoted to Pappas - or a more English-sounding Pope. But until recently, researchers lacked a clear understanding of how often people changed their last names on arrival.
New research from UC Berkeley indicates that in much of the 1900s, name changes were much less widespread than long thought: More than 80% of immigrants kept their surnames intact. About 15% of those who arrived from 1936 to 1970 made slight changes that preserved their ethnic roots, and only about 5% changed their last name to an English-sounding one that could mask their origins, according to a study of population-level data published last week in the journal Proceedings of the National Academy of Sciences.
"Our findings challenge the dominant assimilationist narrative that immigrants in this era had to shed their foreign identities to enter the American mainstream," researchers wrote in the new paper.
To quantify how common both subtle and significant name changes were, researchers combed through a Berkeley-built dataset of millions of Social Security records. They limited the review to men whose first applications for a Social Security number were filed between 1936-1970, a span from the beginning of such applications to the most recent batch of publicly released data. The records were particularly useful because they included the name of the applicant's father, allowing researchers to track surname transmission across generations. The study omitted women because of name changes that stem from marriage as well as immigrants from Spanish-speaking countries whose surnames are a combination of a mother's and father's and are thus more difficult to parse.
With complete records on about 11 million people, researchers identified 596,000 who were foreign-born. They narrowed their focus to 172,000 men from 11 countries that were subject to nativist or racial debates and discussions at the time: Italy, Germany, Poland, Russia, Greece, China, Hungary, Austria, Egypt, Japan and Iran. They then compared sons' surnames to their fathers'.
"Here we have a chance to look at one way the world was supposed to work: That when you came to America, it was bad to be seen as a new immigrant, and so you changed your name," said Joshua R. Goldstein, a Berkeley professor of demography and first author of the study. "But when we look at the entire population, we actually see the opposite."
Less than 1% of Italians Americanized their names, Goldstein said. And despite frequent anecdotes of Jews changing their names, even that pattern was far less pronounced. Jewish immigrants accounted for the highest rate of changes. But among men whose fathers had names like Shapiro or Cohen, nearly 90% kept them; only about 6.5% changed to English-sounding names.
The paper introduces a framework for thinking about the costs and benefits of changing names, consistent with their findings that visible minorities from Japan and China rarely changed their surnames. Goldstein said there was a psychological and social cost for giving up one's ethnic identity that often won out against the potential and uncertain gain of trying to blend in.
Goldstein and his co-author, Guy Stecklov at the University of British Columbia, also found that immigrants who moved to the Deep South had a name change rate that was double that of the Northeast. That could suggest that name changes were rarer when people lived in communities with large immigrant populations and perhaps more welcoming policies, Goldstein said.
"As demographers, we focus on trying to get the big picture, which is often hidden," Goldstein said. "But the story about how people didn't change their names is not something you would tell at the dinner table. So there's a certain asymmetry in what stories get told."
The researchers acknowledge that most of the immigrants in the study arrived toward the end of or after the previous century's major migration waves. A smaller analysis of people born before 1900 found that they were slightly more likely to adopt an English name.
National Archives
Still, it was far from common even then, despite a longstanding myth that workers at ports of entry like Ellis Island changed names that were unfamiliar to them. In reality, inspectors didn't actually write down immigrant names; they checked them off a steamship manifest. That's not to say immigrants didn't change their names at all, Goldstein said, but it's not something that happened at the scale many may believe.
The research matters for population scholars in another way. Understanding how rarely names changed - and how they were most commonly altered - allows researchers to use data and computational linguistics tools to more reliably link historical records across generations and study a range of population trends.
The new paper builds on Goldstein's previous work with immigrant names, showing that the children of immigrants who carried American-sounding names had higher occupational achievement - something he and Stecklov found was the case even for those with distinctively non-English last names.
At scale, both first and last names can help tell a story about society. "The normative way to become American was not to say that you weren't from somewhere else," Goldstein said. "It was to say you're trying to become an American. Taking on that American first name was enough of a kind of pledge of allegiance. But you can keep your foreign origin."
