Are Americans more likely to be monolingual?

Question

From a comment on English Language & Usage, also mentioned in Wikipedia, and Chad Fowler's book The Passionate Programmer (Related blog post by the author: How Learning a Second Language Changed My Life, where he describes it as a joke he heard in India, which used to be a British colony)

What do you call someone who speaks three languages? Trilingual.

What do you call someone who speaks two languages? Bilingual.

What do you call someone who speaks one language? American.

Are Americans more likely to be monolingual than people of other countries?

The joke doesn't provide a definition of what counts as "American", in terms of whether immigrants to the United States count as American, or the children of immigrants, or only people whose parents were themselves born in the United States.

I suspect there is a more general claim regarding English speakers. Britons also think they are more monolingual than the rest of Europe. — Sklivvz, Aug 28 '14 at 05:05
@Sklivvz that would be very plausible, because English is the most common foreign language learnt. If you're a native English speaker, you wouldn't learn English as a foreign language. — Andrew Grimm, Aug 28 '14 at 05:08
It's more of a claim about India. English-speaking Indians are multi-lingual: English is the 'lingua franca' in India. — ChrisW, Aug 28 '14 at 08:41
http://www.nytimes.com/2012/01/15/opinion/sunday/are-we-really-monolingual.html has some thought on the subject. — liftarn, Aug 28 '14 at 10:00
Larian told me that joke on our first deployment together back in 1995! I think it's quite notable. — JasonR, Aug 28 '14 at 12:17
@ChrisW: and furthermore, Americans in India can know how to speak e.g. Spanish as fluently as they like without getting an opportunity to demonstrate their multi-lingualism ;-) — Steve Jessop, Aug 28 '14 at 12:27
OTOH, I'd count some not as monolingual, but semilingual, as they don't speak correctly even one language ;-) http://media.tumblr.com/505778faa9f68f886f12f2e196c75e4e/tumblr_inline_n0ee9ufKE51qawfnh.png — vartec, Aug 29 '14 at 08:49
@vartec I don't know for sure about this partciular tweet, but these tweets are often jokes, not real. — Andrew Grimm, Aug 29 '14 at 09:48
@AndrewGrimm even if particular ones are jokes, they are only funny because they are mimicking real ones. — vartec, Aug 29 '14 at 09:52
@flimzy the claim is notable not *because* it is a joke, but because a bunch of adults believe the claim is true — , Aug 29 '14 at 15:55
In general, Americans have comparatively few interactions with non-English speakers. One border is with Mexico, and many Mexicans speak sufficient English. Relatively few Americans need to speak with Mexicans in their native language. The other border is with Canada, and English is most common there. Similar border conditions don't hold in Europe nor in many other parts of the world. There simply is little use for and exposure to other languages. — user2338816, Dec 03 '14 at 06:45
@user2338816 this is exactly right. If Napoleonic France had maintained it's Continental Empire, we might be saying the same about French and Europeans. — RonJohn, Sep 10 '19 at 00:23
A better question is what percentage of people on the western edge of the North American continent can understand someone on the eastern edge of the North American continent, and likewise the western and eastern edges of the European continent. — RonJohn, Sep 10 '19 at 00:27
It's worth mentioning that a lot of Americans have Hispanic roots and Spanish-speaking family. — DJClayworth, Sep 11 '19 at 14:18
Expanding upon what user2338816 said: Look how far one has to travel before finding themselves in a place that doesn't speak the language--the farther that is, the less reason there is to know any other language. Americans either have to cross an ocean or head south--and many of those places to the south aren't the safest places and thus there won't be too much travel to them. No other country comes close to this linguistic isolation. — Loren Pechtel, Jan 17 '21 at 03:35

score 30 · Answer 1 · edited Jan 17 '21 at 00:29

30

SUMMARISED ANSWER

To be able to answer the question, a baseline measurement needs to be established. What does it mean to be multilingual/monolingual? How do you measure populations? Where does the data come from? Who is being compared in a "more likely" scenario? American to another Westerner, or to an average global citizen?

There are no reliable statistics to answer the question. So the only honest answer, unfortunately, is we don't know for sure.

That said, here are some points of reflection for those who want to explore the question from different angles.

DETAILED ANSWER
_{(source: dana.org)}

Viorica Marian Ph.D. and Anthony Shook compare US to EU bilingualism as part of a neurological study and created the graph comparison above. Based on statistics from the European Commission and the US Census, there is clearly a big gap, although there is also a distinction between speaking a language at home and having the skills to speak another language.

François Grosjean, Ph.D., Emeritus Professor of psycholinguistics at the University of Neuchâtel, Switzerland, reflects also on the challenge of counting bilinguals, including the definition.

I will present two examples of national censuses—the one in the United States and the one in Switzerland—and show how quite official data can sometimes produce very surprising results. In this instance, it would appear that there are proportionally more bilinguals in the United States than in Switzerland, even though the latter country is known the world over for the bi- or multilingualism of its inhabitants.

In short, this is because of the way questions are phrased (specifically related to German vs. Swiss German dialect).

The outcome was that Swiss Statistics stated that a mere 15.8 percent of the Swiss population is bi- or multilingual (less than the 20 percent found for the U.S.) when, in fact, most Swiss people know several languages that they use frequently.

In this piece, he also discusses the drivers and influencing factors in bilingualism. Asia as a general rule is pointed to as a place where multilingualism is a daily fact. Americans have less incentive/need to learn another language because English is the current global language of trade/commerce (as Greek and Latin have been in the past).

With these facts and the phrasing of the question "more likely", the answer nudges to "Yes, Americans are more monolingual", but Michael Erard writing for the NYT in 2012 believes that it's still a toss up.

Recently, the Stockholm University linguist Mikael Parkvall sought out data on global bilingualism and ran into problems. The reliable numbers that do exist cover only 15 percent of the world’s 190-odd countries, and less than one-third of the world’s population. In those countries, Mr. Parkvall calculated (in a study not yet published), the average number of languages spoken either natively or non-natively per person is 1.58. Piecing together the available data for the rest of the world as best he could, he estimated that 80 percent of people on the planet speak 1.69 languages — not high enough to conclude that the average person is bilingual.

Multilinguals may outnumber monolinguals, but it’s not clear by how much. The average American may be no more monolingual or less multilingual than any other average person elsewhere on the planet. At the very least, we can’t say for sure — not in any language. [emphasis added]

EDIT: Check out this Wolfram Alpha-generated overview of the spread of languages spoken in the USA at home. It has been noted by several that the home bias is a core weakness to the census data.

edited Jan 17 '21 at 00:29

Glorfindel

1,452
1
17
28

answered Aug 28 '14 at 09:57

MMPA

554
1
4
8

3

Well referenced statements that something is hard to measure are useful, so I've upvoted this. – Andrew Grimm Aug 28 '14 at 10:30
16

I'd like to point out the graph about the United States does not specifically speak to multilingualism; it addresses whether English is the language used at home. Somebody could speak a non-English language at home and also not know English. – SocioMatt Aug 28 '14 at 12:13
@SocioMatt - Good point. The three questions asked in the US Census are "Does this person speak a language other than English at home? What is this language? How well does this person speak English (very well, well, not well, not at all)?" ... the researchers apparently chose to use the responses to the first question without washing out "not at all" answers for the last question. – MMPA Aug 28 '14 at 12:24
26

And for that matter an American could be multi-lingual and nevertheless speak only English at home (by choice or because that's the only language they have in common with whoever they live with). Sounds like those researchers basically found *no data* on multi-lingual Americans, if that's the closest they came up with. – Steve Jessop Aug 28 '14 at 12:25
3

Agreed. I speak fluent Spanish as well as English... but not at home. – Mason Wheeler Aug 28 '14 at 13:00
@SteveJessop There's a lot of data, but it's tough to wash. That's kind of the conclusion of the post. That data is from the US Census, which at least has historical consistency going for it. I have to take personal responsibility for repurposing their apples-to-oranges graph: the original article is about the neurological impact of bilingualism rather than a sociolinguistic analysis. Here's a good Wolfram Alpha deep-dive, but still with the "home" bias: http://www.wolframalpha.com/input/?i=how+many+spanish+speakers+are+in+usa – MMPA Aug 28 '14 at 13:11
@MMPA: well, I mis-stated a little. They had data about certain categories of Americans who are multi-lingual, they lacked data about the group "multi-lingual Americans" en masse... – Steve Jessop Aug 28 '14 at 14:04
Uh, this has lots of upvotes, but does it answer the question? I can't see where it does. – Oddthinking Aug 28 '14 at 14:54
@Oddthinking - It is tricky, but the "definitive" answer is in bold at the end. In short, the lack of good-quality, comparable data makes it impossible to accurately answer today. So the only honest answer is "We can't say for sure". – MMPA Aug 28 '14 at 15:00
3

So, to summarise: The first graph about Europe looks interesting, but the Swiss think it doesn't measure the right thing. The second graph about the USA is not comparable, and is not helpful for this question. You think the answer might be true based on some theory. A Swede measured something completely different which didn't help. A linguist in the NYT says he couldn't find any useful stats. Might I suggest you scrub the answer down to that last point? – Oddthinking Aug 28 '14 at 15:14
What's the data for the UK? AFAIK, it's said to be the most monolingual country in Europe. – vartec Aug 28 '14 at 15:16
1

What's the rationale for using Europe as the basis for comparison, instead of areas with significantly larger population, such as China? – Ben Voigt Aug 28 '14 at 15:37
Aside from the previously-mentioned holes in the 'language spoken at home' statistic, someone could conceivably not speak any language regularly at home. I, for instance, live by myself. Since I'm not in the habit of talking to myself, I don't tend to frequently speak any language at home unless I have company or I'm on the phone. – reirab Aug 28 '14 at 20:00
So to summarize the US census data: Despite me speaking four languages (three fluently), I'd be considered monolingual because I generally only speak German (or English when in the US) at home? Really doesn't make sense to use that data to show that Americans are mostly monolingual. So we're down to anecdotes (lots of those though) and the fact that the school system in Europe forces students to learn English besides their native tongue [and EU directives to learn a third living language) while you can finish high school in the US without having to learn another language. – Voo Aug 28 '14 at 21:04
@BenVoigt Because that's thirty or more countries with good data. China is only one country. – Relaxed Aug 28 '14 at 22:44
I believe European data is mostly self-reported ability, which might or might not reflect actual ability to speak a language. – Relaxed Aug 28 '14 at 22:45
I disagree with the assertion that the only honest answer is "we don't know," at least in the comparison of the U.S. to Europe (admittedly, statistics for some parts of Africa and Asia would be much more difficult to ascertain.) I have added an answer below that does compare directly comparable and reputable statistics of the U.S. vs. the E.U. Both surveys asked the question of whether the respondent is sufficiently fluent to hold a conversation in a second language, which seems like a very reasonable (probably the most reasonable) definition of multilingualism. – reirab Aug 29 '14 at 15:30
The data for EU is actually *percentage of people who claim to know at least one foreign language*, which is absolutely not comparable to *percentage of people who speak another language at home*. For example Sweden might have 97% of people fluent in English, but I doubt if anyone significant number of them speaks it at home. – vartec Feb 01 '18 at 21:26

score 22 · Accepted Answer · answered Aug 29 '14 at 15:27

22

At least in the case of Europe and the U.S., statistics do exist that directly answer this question. Those statistics simply aren't the ones from the censuses.

In 2013, a Gallop poll found that 34% of Americans could hold a conversation in at least one second language. In 2012, a European Commission survey found that 54% of Europeans could hold a conversation in at least one second language. So, at least when compared to Europeans, yes, Americans are somewhat more likely to be monolingual.

answered Aug 29 '14 at 15:27

reirab

4,095
20
25

4

The Gallup poll was a phone interview of people aged 18 and older. The European poll was face-to-face interview of people aged 15 and older. The Annex which describes the European survey is [found here](http://ec.europa.eu/public_opinion/archives/ebs/ebs_386_anx_en.pdf). – ChrisW Aug 29 '14 at 16:12
1

Good catch. That might skew the statistic slightly in Europe's favor, since school-age children would be more likely to be taking a second-language class. Still, assuming the age distribution of those polled is representative of the populations studied, 15-17 year olds would only represent a few percent of the population, so the overall statistics wouldn't be affected by more than a few percent by that discrepancy, at most. The measured difference is sufficiently large that the ultimate conclusion of Americans being more monolingual would still hold. – reirab Aug 29 '14 at 16:19
2

I agree (and upvoted your answer). I posted that comment because I wanted to verify whether and how much the two surveys were alike, and it took me a slight amount searching (whose result I wanted to share) to find the Annex in which the European survey was described. – ChrisW Aug 29 '14 at 16:25
Yes, I appreciate the added information and upvoted your comment. I just wanted to make its ramifications clear to others who might read it. – reirab Aug 29 '14 at 16:44
Subtracting the 1 American in 6 with Hispanic roots, the numbers look even bleaker for those American with English as mother tongue. – ZeroTheHero Jan 19 '21 at 14:24
@ZeroTheHero It would make a difference to be sure, but perhaps not as much of one as you might think. That poll actually did break down the results by white/black./Hispanic with 30%, 31%, and 54% claiming to be able to hold a conversation in a second language, respectively. There are quite large portions of recent Hispanic immigrants who aren't conversational in English and also a significant amount of 2nd+ generation Hispanic immigrants who aren't conversational in Spanish. – reirab Jan 19 '21 at 16:52
@ZeroTheHero Also, just to clarify, while most Hispanic people in the U.S. are normally considered "white," the methodology section at the end there says that their "white," "black," and "Hispanic" categories are actually "non-Hispanic white," "non-Hispanic black," and "Hispanic," as three disjoint sample groups, so it's not including Hispanic white respondents among the "white" numbers. – reirab Jan 19 '21 at 17:11
Yes probably some fraction (that I would not know) of the total population with Hispanic roots would still be uniligual. I would not have guesses a "significant" portion but you could be right. I also heard the original OP question for the first time many years ago (in Paris) as a joke on "anglophones" broadly. The current question is a little more specific. – ZeroTheHero Jan 19 '21 at 18:31
1

@ZeroTheHero According to that survey, 46% of American Hispanics surveyed were monolingual (many of those only Spanish, many only English.) And, yes, I'd assume that monolingualism is much higher in the UK than in the remainder of Europe, too. Same for Australia, NZ, etc. Similarly, I'd assume that multilingual percentages in continental Europe would drop significantly if you didn't count English. – reirab Jan 19 '21 at 18:40

Are Americans more likely to be monolingual?

2 Answers2

Linked