The true cost of science’s language barrier for non-native English speakers(nature.com)
nature.com
The true cost of science’s language barrier for non-native English speakers
https://www.nature.com/articles/d41586-023-02320-2
49 comments
This problem is real. As a non native speaker myself, I only became reasonably reading fluent in English around 19 years old. I remember well how hard was to learn it. Some things a native speaker takes for granted seems ludicrous to those that don't know it well. For example, in programming, pythonistas take pride in saying Python is easy to understand because it reads like plain english, like pseudo code that runs. Guess what, for those that don't know English, it doesn't make any difference. The keywords are as cryptic to learn as assembly opcodes or any other random language syntax.
There must be a lingua franca and English is likely among the best choices, considering its widespread adoption and relatively low barrier of entry (try Japanese or Vietnamese and you will see what I meant).
But it is true that there is cost to non-English speakers. Maybe we can lower the standard a little bit? I don't mean broken Engrish but if the sentences make sense and the grammar is correct, maybe that is good enough?
But it is true that there is cost to non-English speakers. Maybe we can lower the standard a little bit? I don't mean broken Engrish but if the sentences make sense and the grammar is correct, maybe that is good enough?
Japanese the language qua language is relatively easy from a language agnostic perspective. Formality levels are tricky, but the phonology is fairly simple (very few "weird" sounds) and the grammar is on the regular side.
The writing system on the other hand... IMO the worst used by any major language in the world. Graft a logographic system (already a bad choice to start) from an isolating language onto a highly synthetic verb system, necessitating a whole extra syllabary. Evolve numerous "readings" of each logogram, depending on etymology and history. Then tack on a 2nd syllabary for abstruse historical reasons.
The writing system on the other hand... IMO the worst used by any major language in the world. Graft a logographic system (already a bad choice to start) from an isolating language onto a highly synthetic verb system, necessitating a whole extra syllabary. Evolve numerous "readings" of each logogram, depending on etymology and history. Then tack on a 2nd syllabary for abstruse historical reasons.
> If the seemingly arbitrary and unique assignment of pronunciations to common combinations of Kanji frustrates you—[...]—and leave you wondering why anyone would do such a thing, concluding that Japanese has the worst writing system ever, rest assured that they don't have the worst ever. They just have the worst writing system since the Hittites.
Cuneiform signs can be employed in three functions: syllabograms, Akkadograms or Sumerograms. Syllabograms are characters that represent a syllable. Akkadograms and Sumerograms are ideograms originally from the earlier Akkadian or Sumerian orthography respectively, but not intended to be pronounced as in the original language; Sumerograms are mostly ideograms and determiners. ... Thus, the sign GI 𒄀 can be used (and transcribed) in three ways, as the Hittite syllable gi (also ge); in the Akkadian spelling QÈ-RU-UB of the preposition "near" as QÈ, and as the Sumerian ideogram GI for "tube" also in superscript, GI, when used as a determiner.
https://en.wikipedia.org/wiki/Hittite_cuneiform
https://en.wikipedia.org/wiki/Hittite_cuneiform
> But it is true that there is cost to non-English speakers. Maybe we can lower the standard a little bit? I don't mean broken Engrish but if the sentences make sense and the grammar is correct, maybe that is good enough?
We can't demand things like that. If a paper reads like amateur English, reviewers are pretty much guaranteed to turn it down with higher probability, even if the factual quality is somewhat better than of a well-written paper. This is just how the mind works. Like wine from a cheap looking bottle tasting worse than the same wine from an expensive looking bottle.
We can't demand things like that. If a paper reads like amateur English, reviewers are pretty much guaranteed to turn it down with higher probability, even if the factual quality is somewhat better than of a well-written paper. This is just how the mind works. Like wine from a cheap looking bottle tasting worse than the same wine from an expensive looking bottle.
> There must be a lingua franca
Not true. The period of greatest scientific advancement in human history ( mid-19th to the mid-20th century ) had no lingua franca. Actually, the end of french as a 'lingua franca' in europe in the first half of 1800s can partly be credited with the burgeoning of science ( and other academic fields ) in britain, germany, russia, etc. Just like the ending of latin as 'lingua franca' ( beginning of the 'vulgar' period ) led to the burgeoning of religion, philosophy, literature, etc throughout europe.
> Maybe we can lower the standard a little bit? I don't mean broken Engrish but if the sentences make sense and the grammar is correct, maybe that is good enough?
Or maybe people should speak their own language and 'do science' in their own language.
What's more preferable? Having one billion chinese learn english to study algebra with an english language textbook or translated that one algebra textbook into chinese and have one billion chinese learn algebra in chinese?
The dull and boring mono-linguistic and mono-culture world has led to stagnation. The world would be infinitely better off when that changes. When we have actual diversity, not the superficial diversity where we look different but all think and talk the same.
Look at the tech world today. Look at just europe alone. How great would it be if europe had a german, spanish, russian, french 'google/apple/microsoft/etc'? Now broaden that view to the arab, persian, chinese, indian, african, etc regions. So much diversity, knowledge, creativity, etc is lost because one group of people want to dominate the world. It's tragic.
Not true. The period of greatest scientific advancement in human history ( mid-19th to the mid-20th century ) had no lingua franca. Actually, the end of french as a 'lingua franca' in europe in the first half of 1800s can partly be credited with the burgeoning of science ( and other academic fields ) in britain, germany, russia, etc. Just like the ending of latin as 'lingua franca' ( beginning of the 'vulgar' period ) led to the burgeoning of religion, philosophy, literature, etc throughout europe.
> Maybe we can lower the standard a little bit? I don't mean broken Engrish but if the sentences make sense and the grammar is correct, maybe that is good enough?
Or maybe people should speak their own language and 'do science' in their own language.
What's more preferable? Having one billion chinese learn english to study algebra with an english language textbook or translated that one algebra textbook into chinese and have one billion chinese learn algebra in chinese?
The dull and boring mono-linguistic and mono-culture world has led to stagnation. The world would be infinitely better off when that changes. When we have actual diversity, not the superficial diversity where we look different but all think and talk the same.
Look at the tech world today. Look at just europe alone. How great would it be if europe had a german, spanish, russian, french 'google/apple/microsoft/etc'? Now broaden that view to the arab, persian, chinese, indian, african, etc regions. So much diversity, knowledge, creativity, etc is lost because one group of people want to dominate the world. It's tragic.
> Not true. The period of greatest scientific advancement in human history ( mid-19th to the mid-20th century ) had no lingua franca. Actually, the end of french as a 'lingua franca' in europe in the first half of 1800s can partly be credited with the burgeoning of science ( and other academic fields ) in britain, germany, russia, etc. Just like the ending of latin as 'lingua franca' (beginning of the 'vulgar' period ) led to the burgeoning of religion, philosophy, literature, etc throughout europe.
Huh?!? That's totally false.
The lingua franca changed from the mid-19th century to the mid-20th century. Depending on the field of science it was either French or German and transitioned to English.
For physics it was German moving into English over that 100 years. https://en.wikipedia.org/wiki/List_of_important_publications...
In math French was a bit more of a player. https://en.wikipedia.org/wiki/List_of_important_publications... Hilbert for example published in German until the ascendancy of English. Turing's famous work is in German.
No one serious was publishing in Russian, Italian, etc. There was a clear lingua franca.
Huh?!? That's totally false.
The lingua franca changed from the mid-19th century to the mid-20th century. Depending on the field of science it was either French or German and transitioned to English.
For physics it was German moving into English over that 100 years. https://en.wikipedia.org/wiki/List_of_important_publications...
In math French was a bit more of a player. https://en.wikipedia.org/wiki/List_of_important_publications... Hilbert for example published in German until the ascendancy of English. Turing's famous work is in German.
No one serious was publishing in Russian, Italian, etc. There was a clear lingua franca.
> Huh?!? That's totally false.
Lets see who is lying and who is telling the truth.
> Depending on the field of science it was either French or German and transitioned to English.
Charles Darwin did his work in french? Or was it german? What language did faraday, maxwell, dalton, etc use? French or german? What language did lord kelvin to give us the temperature scale? Can you point to a single scientific paper that an english or american scientist wrote in german or french?
> For physics it was German moving into English over that 100 years.
Yes I know. I'm not denying that english is lingua franca.
> In math French was a bit more of a player.
Bit of a more player? That's proving my point. French was not lingua franca. Nobody would say 'english is bit of a more player' in math, science, business, politics, etc. Lingua franca is obvious. You don't have to miscontrue it with such weasely language.
> Turing's famous work is in German.
I don't know german and I've read turing's original works on computation, AI, morphogenesis, etc. Have you? They were most definitely not written in german. What's with people who know absolutely nothing writing with such authority?
> There was a clear lingua franca.
So tell me what 'lingua franca' turing was publishing his work in. It wasn't german like you claimed so german wasn't lingua franca. It surely as hell wasn't french.
The only lingua franca in europe before english was latin. Neither french nor german ever was 'lingua franca' as you've clearly proved with turing. At best you could argue that english, french and german were major players in science, math, etc. But certainly there was no lingua franca until english took over post-ww2.
Lets see who is lying and who is telling the truth.
> Depending on the field of science it was either French or German and transitioned to English.
Charles Darwin did his work in french? Or was it german? What language did faraday, maxwell, dalton, etc use? French or german? What language did lord kelvin to give us the temperature scale? Can you point to a single scientific paper that an english or american scientist wrote in german or french?
> For physics it was German moving into English over that 100 years.
Yes I know. I'm not denying that english is lingua franca.
> In math French was a bit more of a player.
Bit of a more player? That's proving my point. French was not lingua franca. Nobody would say 'english is bit of a more player' in math, science, business, politics, etc. Lingua franca is obvious. You don't have to miscontrue it with such weasely language.
> Turing's famous work is in German.
I don't know german and I've read turing's original works on computation, AI, morphogenesis, etc. Have you? They were most definitely not written in german. What's with people who know absolutely nothing writing with such authority?
> There was a clear lingua franca.
So tell me what 'lingua franca' turing was publishing his work in. It wasn't german like you claimed so german wasn't lingua franca. It surely as hell wasn't french.
The only lingua franca in europe before english was latin. Neither french nor german ever was 'lingua franca' as you've clearly proved with turing. At best you could argue that english, french and german were major players in science, math, etc. But certainly there was no lingua franca until english took over post-ww2.
That said there gotta be some balance though. It is much more effort to translate something into every language that interested people might use, that is, if people who can translate are willing to spend the effort at all. Learning a common language has opened up an enormous realm of knowledge that many other languages can't come close (see for example the English Wikipedia vs. other languages' Wikipedia, both in article count and in the length of each article).
I’d imagine for most languages people are going to be using for scientific dialog already have pretty robust translation tools. You can probably put a Chinese math paper through google translate and get decent output. This sort of thing could be built into google scholar, maybe even finer tuned for the exact idioms and jargon used by each field in that language.
This is true, and I can tell you that we're actually missing out on a lot of interesting Chinese research in mental disorders and therapy.
> Maybe we can lower the standard a little bit?
Lower them more you mean?
As a native English speaker having learnt a foreign language, I can tell you that we're pretty goddamn lax in comparison. The English have long since made peace with the fact that their language has been usurped. They're okay with it, and make concessions for foreign language speakers; as long as the meaning is understood, nobody bothers correcting you.
Now try speaking French in France, or German in Germany. I don't know about Japanese in Japan, but I imagine it's the same or worse (from what I've heard second hand).
Lower them more you mean?
As a native English speaker having learnt a foreign language, I can tell you that we're pretty goddamn lax in comparison. The English have long since made peace with the fact that their language has been usurped. They're okay with it, and make concessions for foreign language speakers; as long as the meaning is understood, nobody bothers correcting you.
Now try speaking French in France, or German in Germany. I don't know about Japanese in Japan, but I imagine it's the same or worse (from what I've heard second hand).
I can't see this as the injustice the article makes it out to be. Not standardizing on a single language would just cause more problems. If it wasn't English it would be something else.
The author isn’t arguing against English as a lingua franca, but it seems to me suggesting that there should be more accommodations for those who aren’t native English speakers because they are at a distinct disadvantage.
I have so much respect for students, researchers, and professors whose native language is not English. Not only have they reached a level of English proficiency that allows them to converse and share viewpoints with their colleagues, but also they can understand and express themselves using academic English styles specific to their fields.
The point is such respect is no longer necessary or required. It'll make science evolve faster, which is the whole idea.
For tech and science, english must be used as lingua franca/default.
This has nothing to do with any property of english or it's speakers, it is simply the path of least resistance for everyone involved. Due to media and existing materials, english (even badly worded/spoken) is much easier to pick up than most languages and opens up the most opportunities to the most people.
Learning Chinese is for example is much more difficult for most non-chinese with no added benefit unless they travel to china or work for a chinese company.
Perhaps if english was renamed to internationalese it would make things easier and more palatable.
This has nothing to do with any property of english or it's speakers, it is simply the path of least resistance for everyone involved. Due to media and existing materials, english (even badly worded/spoken) is much easier to pick up than most languages and opens up the most opportunities to the most people.
Learning Chinese is for example is much more difficult for most non-chinese with no added benefit unless they travel to china or work for a chinese company.
Perhaps if english was renamed to internationalese it would make things easier and more palatable.
Conversely there is the advantage of the narrow waist. the state where everything has to fit in the same container.
The biggest example of this is the internet, where we have many types of cable to carry information, and many applications that want to talk to each other, but everything speaks IP.
see also: containerized shipping, nema motor mounts.
So true, it is hard if english is not your first language, but there is this huge advantage to having a single common language that everyone speaks. today it is english, before that french, and before that latin.
The biggest example of this is the internet, where we have many types of cable to carry information, and many applications that want to talk to each other, but everything speaks IP.
see also: containerized shipping, nema motor mounts.
So true, it is hard if english is not your first language, but there is this huge advantage to having a single common language that everyone speaks. today it is english, before that french, and before that latin.
Broken English in scientific papers are, if anything, a problem. I have a brother-in-law who doesn't speak English, but gets his papers translated into English to publish. There is no way for him to really know if they precise, accurate, and correct; as the person who knows his own research the best, he simply cannot review the paper.
As it might be possible for us to eventually improve our machine translation, it might be more valuable to keep his papers in Japanese.
As it might be possible for us to eventually improve our machine translation, it might be more valuable to keep his papers in Japanese.
This is a fine example, because the text has a few grammatical errors but they do not impair intelligibility so shouldn't matter.
The challenge comes where errors may have impacted intelligibility or worse still changed their meaning. This kind of misunderstanding is common and I'm not surprised it needs extra effort - I had a major disruption this week when an ESL coworker misunderstood the meaning of a particular ambiguous word in a project even though this exact difficulty was anticipated and time was spent up front to clarify it.
The challenge comes where errors may have impacted intelligibility or worse still changed their meaning. This kind of misunderstanding is common and I'm not surprised it needs extra effort - I had a major disruption this week when an ESL coworker misunderstood the meaning of a particular ambiguous word in a project even though this exact difficulty was anticipated and time was spent up front to clarify it.
Learning English is priority #1 for any human being right now. If you are able to communicate in english, so many opportunities and sources of information open up to you. It shocks me that some countries aren't making it mandatory, but that's due to nationalism and not wanting too much emigration/brain drain I suppose.
Not being moderately fluent in english is inexcusable in the modrn world, in the same that not being tech savvy, or not able to do math is inexcusable. There are basic skills that one needs to operate in a modern world. English is one of them. I appreciate the sentiment, but it's akin to complaining that "In the star wars universe, scientists who don't speak Galactic Basic Standard are at a huge disadvantage." Yes, they are, and it's their fault for not learning the common language. Same with English in 2023.
Not being moderately fluent in english is inexcusable in the modrn world, in the same that not being tech savvy, or not able to do math is inexcusable. There are basic skills that one needs to operate in a modern world. English is one of them. I appreciate the sentiment, but it's akin to complaining that "In the star wars universe, scientists who don't speak Galactic Basic Standard are at a huge disadvantage." Yes, they are, and it's their fault for not learning the common language. Same with English in 2023.
I don’t think this is as true as you make out. For the vast majority of people (who are born, live and die all in the same country/language community) should it really be their #1 priority to learn English? For example, a street food vendor in Chengdu, a phone salesman in Belo Horizonte, or a roofing contractor in Avignon. Or even a professor at a small university in Kagoshima. Learning English would probably open up some new opportunities for all these people. But to the extent they should dedicate years of study time to it? I’m not sure.
The same can be said about a cell phone, a car or electric. It's just a matter of time before the globalization and modern way of life hit everybody. But it's always subject to individual who's decided to go offgrid, one can always say No to English and happily ever after. It's a choice with hefty consequences, that's all one needs to know.
It’s effectively a branch. It determines some things you can do and can’t so.
There have been different language barriers in the past—most notably Latin which wasn’t used natively by pretty much any speakers after a point and then German in some fields and French in European courts.
But yeah, English has become pretty much the standard language for many purposes and if you didn’t learn it as a native it’s something you have to learn as a non-native for many purposes.
But yeah, English has become pretty much the standard language for many purposes and if you didn’t learn it as a native it’s something you have to learn as a non-native for many purposes.
Manuscripts from non-native speakers place a considerable burden upon unpaid peer reviewers. Going through 10,000 barely comprehensible words is not fun.
I think it's extremely unfair how much of an advantage us native anglophones have in the scientific realm. Doing anything interesting enough to get published is hard enough as it is, and I have to assume that it's substantially harder for people who are also fighting a language barrier as well.
A part of me wants something like Esperanto to take over this space.
A part of me wants something like Esperanto to take over this space.
Esperanto just creates the same language barrier for everyone.
English is an easy language to learn. I'm ESL and it's no big deal.
English is an easy language to learn. I'm ESL and it's no big deal.
Not all ESLs have it equal, both due to differences in standardised education, and how close your primary language is to English.
For example, a Japanese person probably would find learning English to the level required to publish academic papers to be a “big deal”.
For example, a Japanese person probably would find learning English to the level required to publish academic papers to be a “big deal”.
There’s probably a distinction between speaking/writing as a fluent and idiomatically interesting well-versed native writer and competent prose for a paper—which is fine for most purposes; an editor will fix you up. Most native speakers aren’t great writers either. Trust me. It’s a lot of my job these days.
I've heard it said that "English is an easy language to speak badly and still be understood". Most Europe, even the French, muddle through in a sort of ESL because it's the most common intermediate language.
If you're Dutch, you're not going to learn Italian just to holiday in Rome.
If you're Dutch, you're not going to learn Italian just to holiday in Rome.
> I'm ESL and it's no big deal.
Not everyone would be learning English as a second language. In many parts of the world you may need to know some regional language in addition to your native one, so English would be your third, not second language.
Not everyone would be learning English as a second language. In many parts of the world you may need to know some regional language in addition to your native one, so English would be your third, not second language.
> Esperanto just creates the same language barrier for everyone.
Yes, that's kind of what I was getting at; I think everyone should go through the same hurdles instead of creating a huge advantage for about ~400 million people while the remaining 7.6 billion people have to do extra work.
If we chose a language that isn't really spoken natively, then in theory we could create a system that doesn't give an inherent advantage to any one group.
Obviously that can't be strictly true in practice; a language like Esperanto would give advantages to people more used to European languages, and other conlangs or abandoned languages would have other issues.
> English is an easy language to learn. I'm ESL and it's no big deal.
Is that true? I've always heard that English is a hard language to learn.
Yes, that's kind of what I was getting at; I think everyone should go through the same hurdles instead of creating a huge advantage for about ~400 million people while the remaining 7.6 billion people have to do extra work.
If we chose a language that isn't really spoken natively, then in theory we could create a system that doesn't give an inherent advantage to any one group.
Obviously that can't be strictly true in practice; a language like Esperanto would give advantages to people more used to European languages, and other conlangs or abandoned languages would have other issues.
> English is an easy language to learn. I'm ESL and it's no big deal.
Is that true? I've always heard that English is a hard language to learn.
I grew up with francophone parents in the suburbs of Montréal, which is in the majority french speaking part of Canada. The context is special in that I live within an hour's drive of the border of places where english is the official language and that bilingualism is fairly standard in Montréal, I have loads of friends with whom conversation involves seamlessly switching between english and french. I got my first experience of full immersion at age 4 a few afternoons a week in pre-school, at age 7 I was in an hockey exchange with a new england family for a weekend, miming and pointing at things to communicate with people when my limited vocabulary didn't cut it, at age 12 started streaming TV shows in english and by age 14 I was reading Mary Shelley's Frankenstein.
There's something to be said about my educational privilege and the geographic context of my being ESL. Obviously it's probably harder to learn english if your first language isn't that of a country that has colonized England at one point and you don't grow up within walking distance of the USA, but considering that billions of people have english as a second language and totally manage to convey whatever idea they want when they speak or write it I don't think my point that it's no big deal to learn it is outlandish. It's the most spoken language and it's speakers are all over the globe, there isn't a language that's as accessible to practice as english.
I think levelling by the bottom by forcing everyone to learn a second language (which would become a lot of privileged kids' first language after a few generations anyway) isn't a very interesting idea.
There's something to be said about my educational privilege and the geographic context of my being ESL. Obviously it's probably harder to learn english if your first language isn't that of a country that has colonized England at one point and you don't grow up within walking distance of the USA, but considering that billions of people have english as a second language and totally manage to convey whatever idea they want when they speak or write it I don't think my point that it's no big deal to learn it is outlandish. It's the most spoken language and it's speakers are all over the globe, there isn't a language that's as accessible to practice as english.
I think levelling by the bottom by forcing everyone to learn a second language (which would become a lot of privileged kids' first language after a few generations anyway) isn't a very interesting idea.
No one cares about abstract let’s make everything equally hard fairness And, as you say, Esperanto is still Indo-European and probably Romance so why not just go back to using Latin or French.
That's fair! I was just using Esperanto as an example; I'm not opposed to going back to Latin.
English has some niceness to it that modern Latin derivatives lack. Specifically, the ability to verbificate anything, the lack of having to learn everything’s made-up gender, and the lack of a centralized authority which allows slang to propagate quickly and usefully.
Noun gender is really a non issue. It's something native anglophones complain about when learning other languages. ESL people are more likely to complain about just how many verb tenses English has.
> ESL people are more likely to complain about just how many verb tenses English has.
Also the completely irregular spelling. You need to learn every word twice, first how to write, then how to speak. You can't just infer the sounds from the written word. Languages like Esperanto or German are better in this regard, you learn how to write a word and the odds are you'll know how it sounds as well.
Also the completely irregular spelling. You need to learn every word twice, first how to write, then how to speak. You can't just infer the sounds from the written word. Languages like Esperanto or German are better in this regard, you learn how to write a word and the odds are you'll know how it sounds as well.
English is hard to pronounce and speak correctly. But it's pretty easy to obtain basic proficiency in English - and given how widespread it is, no language beats it in value per time invested.
Maybe you should let that part of you say its piece and move on.
>“Non-native English speakers constitute almost 95% of the world’s population,” Amano says. “If we don’t support those 95%, I’m sure we can’t solve many global challenges.”
Is 95% an accurate number? What percentage are not native speakers, but have near native fluency? I've been to Iceland and Denmark and everyone I spoke to in those places could speak English with near native fluency. I remember meeting one guy in a tiny village in northern Iceland. He sounded like a completely native American English speaker. It turned out that he had never even left his village to travel to Reykjavik, but still had me fooled into thinking that surely he had spent significant amount of time in America.
Is 95% an accurate number? What percentage are not native speakers, but have near native fluency? I've been to Iceland and Denmark and everyone I spoke to in those places could speak English with near native fluency. I remember meeting one guy in a tiny village in northern Iceland. He sounded like a completely native American English speaker. It turned out that he had never even left his village to travel to Reykjavik, but still had me fooled into thinking that surely he had spent significant amount of time in America.
Norse people are pretty amazing when it comes to English speaking, next comes German and Dutch. It seems EU people are generally potent in multilingual. Granted English may be like a dialect to them.
Iceland and Denmark combined have a population of ~6 million. My hometown has almost twice as many people.
> Is 95% an accurate number? What percentage are not native speakers, but have near native fluency?
You are right to wonder about that, but even when we assume every ESL speaker to be fluent enough the proportion only drops to 80%.
You are right to wonder about that, but even when we assume every ESL speaker to be fluent enough the proportion only drops to 80%.
I wonder where that statistic came from too. According to https://en.wikipedia.org/wiki/List_of_languages_by_total_num... the number of L2 speakers of English is ~1b or 12.5% of the world population (~8b).
But one thing that's missed is the advantage ESL speakers have: they by definition speak at least one other language. English speakers either don't have to or find it too hard to find someone to practise on even if they want to learn.
The fact that English is the lingua franca of science has another advantage even to ESL speakers: they only have to learn English to communicate with just about any other scientist in the world. Imagine having to pick between learning English, French, German, Japanese, Latin, depending on who you wanted your audience to be?
But one thing that's missed is the advantage ESL speakers have: they by definition speak at least one other language. English speakers either don't have to or find it too hard to find someone to practise on even if they want to learn.
The fact that English is the lingua franca of science has another advantage even to ESL speakers: they only have to learn English to communicate with just about any other scientist in the world. Imagine having to pick between learning English, French, German, Japanese, Latin, depending on who you wanted your audience to be?