Hacker News new | past | comments | ask | show | jobs | submit login
Researchers discuss the challenges posed by science’s embrace of English (nature.com)
53 points by salutonmundo on June 16, 2019 | hide | past | favorite | 59 comments



I'm not sure I believe this paragraph:

> I’ve worked on many multinational collaborations, and I notice that European researchers often speak to each other in their native languages. However, it’s relatively uncommon to see Chinese or South Korean scientists talking to each other in their own language in an academic setting away from their home country. They just don’t feel comfortable.

In my experience, Chinese are one of the most likely to form little cliques with their compatriots when abroad (and speak their own language while at it). They're not the only ones to do so, but they definitely seem to do it more than average.


I hear equal amounts of Mandarin, Hindi, and Portugese around my current lab. Honestly, the people least likely to speak their native tongue are the few Germans floating around. So my personal experience is closer to yours than that expressed in the original article.


Germans (stereotypically) prefer to speak English poorly over speaking German if there is the possibility that a non-German listener might be interested, the other's German is not 100% perfect or the other is a german wanting to learn english.


In my first job I worked with a lot of Chinese Malaysians, and I remember one time we were in the car going somewhere for lunch. We'd all been conversing in English - for context, this is in Australia where English is the overwhelmingly predominant language, and I was the only non-CM in the car - but when a couple of them started talking directly to each other they started chattering away in... something else, and others joined in it was all incomprehensible.

I said something about them switching languages on me and they said, "No we didn't." And apparently they were still speaking English but when conversing more directly with each other that in the group their accent went so thick and fast it honestly sounded like a foreign language to my uneducated ear. They were totally unaware of this and were really surprised when I said I couldn't understand what they were saying anymore. Made for ongoing jokes for a month or two.


Ha ha, yes. I had a similar experience on Grand Cayman. My host seemed like a totally native english speaker -- until a few days in when a friend came by and they started a little good-natured ribbing over a recent fishing expedition -- they dropped into Caymanian, which is "English" of a very Scotts-influenced and evolved-in-isolation sort. I got nothing out of it except that they both seemed to be in good humor.


The language barrier is most likely the cause. For some reason people from some countries like Russia, China, France etc. struggle with English and they'll seek a way to communicate however they can, usually by finding fellow countrymen.


At least in my experience, in the Bay area, Russians are some of the best at English. There were a couple times where I was at a Russian social gathering and I thought I was around people born and raised in the US but it turned out they moved to the US from Russia when they turned 18.


You don't have to "believe" it, it's a personal anecdote, just like yours. No conclusive evidence has been entered either way.

I've worked with multiple groups of foreign nationals at multiple jobs, and no discernable pattern related to ethnicity emerged. Sometimes one group would talk in their native tongue, sometimes not. It turns out that stereotype isn't a reliable metric.


Academic settings like discussion panel in professional conference? No.

Private gathering during professional conference? Yes.


This article brings some pain points that many people might recognize, I know at least one brilliant chemist that had professorship delayed by about a decade or two. The thing is - that actually might be fair because if you need to teach your students in English you should learn it and be good at it, but how good speaker one has to be to tech chemistry, huh. Maybe the faculty was too hard there, who knows, but sometimes your country of origin might also play a role and not just the language proficiency. I think that some discrimination happens even at that high level of academia.


If people don't like English being the lingua franca of science, they should encourage translation of foreign language articles. Unfortunately scientific translation seems to have declined significantly since the 1950s and 1960s. I wrote a Stack Exchange post about locating translations of scientific articles and few of the sources I mention are currently active:

https://academia.stackexchange.com/a/93209/31143

The rise of English as the lingua franca of science certainly would lead to less translations being produced, but I don't think it should reduce to near zero as it seems to have. Today there's still valuable research that's not published in English.

During my PhD I published several translations (all produced via Google Translate and manual editing) and consistently people thought this was some weird quirk of mine. But I enjoyed it and learned quite a few things I would not have known otherwise.


It used to be required that PhD students (even in the sciences, and in the US) had to demonstrate written understanding of at least one foreign language. When I got my doctorate in microbiology in the 1990s that rule was still officially on the books, but not enforced. But certainly before WWII and even into the 1960s there was a lot of research being published in Russian, French and German besides English. Even today, China still publishes many Chinese-language journals which go mostly untranslated.


They had this requirement at KU in the 90’s, but a computer “language” satisfied the requirement.


> If people don't like English being the lingua franca of science, they should encourage translation of foreign language articles. Unfortunately scientific translation seems to have declined significantly since the 1950s and 1960s.

Really? I don't think there was ever much translation.

Before WWII, a LOT of science was in German (arguably more of a scientific lingua franca than English at the time), and it seems like conferences (ie. face-to-face communication) were the primary way that scientific advances crossed languages.


> Really? I don't think there was ever much translation.

Look at my Stack Exchange post. I know from the "Consolidated Index of Translations into English" that there were ad-hoc English translations of very roughly 300,000 articles produced from 1953 to 1986 (that were reported to the index). That's ignoring the cover-to-cover translation journals too. This isn't everything, but it's a lot. In my experience the ad-hoc translations tended to pick the most important articles that the cover-to-cover translation journals missed.


But wasn't that basically compiled by the US intelligence agencies? So, those translations are a unique artifact of the Cold War, no?

It appears that there was never any drive for translation from the scientific side.


While I think the Cold War increased interested in translations, I don't think US intelligence agencies were the driving force. If I had to guess I'd say that the driving force appears to have been scientific interest in what "the other side" was doing. That combined with the fact that English wasn't the lingua franca of science then as far as I'm aware.

The index I mentioned was produced by the Special Libraries Association, which is an organization that still exists today and has no link to US intelligence as far as I'm aware. The CIA has their own public translations index circa the 1960s (with a similar name) but that largely duplicates this one, and as far as I'm aware rarely mentions actual CIA translations.

The most common type of translation listed is the TT series. The TT series had nothing to do with intelligence agencies as far as I'm aware. Looks like it was funded by the National Science Foundation and the Special Libraries Association: https://www.loc.gov/rr/scitech/trs/trstt.html

The UK translations series I've seen were entirely motivated by scientific interest. In fact, I recall that while looking through a UK translations index I found a form to request a translation. Seems to have been addressed to scientists as intelligence agencies presumably would have the resources to translate articles in-house.

Also, the large number of cover-to-cover translation journals, many of which continue today, would suggest there is appreciable scientific interest in translations.


Angewandte Chemie is an internationally significant scientific journal published in German.


A fair amount of good recent robotics papers are also published in German. My daughter was at an internship and found a good robotics paper in German relevant to her project and actually found it reasonably easy going with high-school AP-German level of language skill. Of course a robotics paper is half math anyway, and German isn't shy about borrowing technical terms from other languages.


> English speakers have become the gatekeepers of science. By keeping those gates closed, we’re missing out on a lot of perspectives and a lot of good research.

I'm always baffled with how disjointed and chaotic the global science community seems. Why there isn't something like Facebook for science? Are there grants for solving meta problems like this?

One interesting fact about using a foreign language: There is a research saying that lying in a foreign language is easier. English is not my mother tongue so don't take this for granted :D


Math, and Science in general, is slow. Arxiv, mathexchange, ncatlab, have each changed how scientists collaborate, as have so many other projects. Wikipedia itself is way way better than a "facebook" for Science. Science is best practised in person

To elaborate, and mis-quote a general, He wins who gets there fastest with the mostest.

Imagine if China had developed digital computers first and had to map their writing to n-ary. It would not be extensions of ASCII upon which we still rely. Imagine instead an algebra of notation based on composition of analogies. Which is why English is not entirely unusual as a Lingua Franca.

English is easy to learn, incorporates a smorgasbord of techniques from other languages, has a rich literature (there has to be more than one good reason to learn a language) and so has ample comparatives. It uses an alphabet, has a lot of law, and is subject to no central authority.

English is a committee of language camels trekking through our desert of ignorance. Kinda like C++

Flame on!


Connecting empirical studies to epistemological formulations is not easy, getting unconnected scientist to exchange ideas is not easy. I think that collaborative potential of the internet for the science as a whole is yet to be realized, I sure hope it won't end up like "Elsevierbook" or some shit like that.


Biology has "ResearchGate" which, ironically, is a bit like facebook and is somewhat proprietary. It is, however, incredibly useful for solving common lab problems like "I stuck all my histo slides together, how do I dissolve the mounting media?".

Bioinformatics has BioStars, stack exchange, and bioarvix.


The fastest developing branch of AI NLP could help with translation, classification and reorganization of knowledge. Fast internet links, HQ videos, VR and simulation software could accelerate the adoption of new concepts and provide possibilities for remote collaboration. Getting up to speed on a new project would be much more interesting and inviting than sifting through a see of bland PDFs and grainy 2D images.


Given the number of speakers of English in the world there really is no alternative to English - maybe Chinese, but I doubt anyone in the world outside of China thinks that it is easier to read Chinese than English.

Of all the problems Science faces the language used is the least of our problems. Little things like funding, insecure tenure, and the immense publish or perish pressure are what we need to be looking at.


No one thinks Chinese is easy even the chinese themselves. One almost as easy language to learn would be Spanish, or even Hindi for goodness sake.


English has additional advantage of being painfully simple logical, grammar-wise. There's only a handfull of grammar rules to learn, and they are pretty consistent, exceptions are quite rare. This helps to write clear and concise text, unlike languages like, say, Polish, which evolved when Poland was under foreign rule for over a century, and for that reason it is deliberately cryptic and ambigous :)


> exceptions are quite rare

... are you sure we speak the same language?


English has less exception than say French but far more than Esperanto. It is a question of reference frame.


Yeah but two of the languages with exceptions are in active use by millions or even billions, while the cohesive one is used by a sum total of... A few dozen? Should Esperanto enter any active usage, however unlikely that is, the exceptions will come.


BTW, Esperanto has a few _hundred_ native speakers and tens or hundreds of thousands of speakers. I agree that it's not comparable to a natural language, but there are many, many natural language communities that are much smaller.


English painfully simple, grammar wise? Wait until you tried constructed language like Esperanto or Ido.


Eh the hard parts of English arise because it likes to mug languages in dark alleys and pick their pockets for spare grammar and vocabulary.

This makes it difficult to form a unified theory of the language. The more elegant your rules, the more exceptions there are.

But it’s also a super power. It makes the language feel more familiar to just about anyone who speaks an Indo-European language. There’s something homely and welcoming in english for anyone.

And the readiness with which english adopts foreign vocabulary when practical is simply inspiring. You can introduce basically any word from any language and make it stick.

And then there’s Slavic languages where at least in mine, every verb or pronoun can come in one of 20 or so different forms. Nouns have 20+ different forms too.

Not expecting Slovenian to become lingua franca of anything any time soon. It’s just too hard.


I can't speak from any experience outside Romance languages, but English verb conjugation is significantly simpler, along with a lack of gendering of nouns. Those two points alone go a long way.


Mandarin the spoken language has quite simple grammar and is pretty easy to learn as languages go.

The written form is painful though.


I was amazed to learn that even native Chinese readers forget the written language if they don't use it for a few years.

The one major thing English has going for it is that native English speakers are pretty tolerant of non-native speakers getting the grammar wrong. With just a little effort you can get people to understand you which is not the case for many languages.


I would say that instead of logical, it is instead familiar and flexible/redundant. You can use essentially whatever structure you like - it may be incorrect, but you have to mess up fairly significantly for English to be unintelligible.


> Why there isn't something like Facebook for science?

I hope not, why would there even be something as dreadful. Interdisciplinary research and science communication are great things and yes there are those who pursue PhDs in that but it's very difficult career wise.

One huge problem is the amount of jargon each and every field accumulates, there are also a too many gibberish meta research papers written and published, they're using the right scientific words and packed with clever explanations but are mostly utterly worthless. And the worthlessness doesn't end there, splitting one subject and it's findings into different articles is just one of those anti-patterns made to show the number of papers to receive more grants.


Thanks for the insights. Facebook for science is a bad metaphor, maybe the Slack for science or something like that is more appropriate. How to build a platform for scientists with incentive structures that reward meaningful research instead of grant fueled hyper-production of fluff?


Maybe form an incubator a la YC but for scientist in cross disciplinary fields (or involve existing ones), then connect the scientist on the interest basis and they maybe deliver some platform for online collaboration, it's the collaboration part online that's the difficult part. Anyone with some programming skills could throw together a django install with orcid.org login but what do with the platform if the most important. Most of these efforts end up like Figshare, holding the data but not encouraging anyone to go outside their boundaries of the field.


It seems it's about discovery and incentives. Which interesting/important project to join and why. Something like GitHub and Open Source Software is possible because programmers can easily make money on the side while donating their time to something they believe in. Plus being successful at that will bring them even more high paying gigs.


Would there be enough engagement in Facebook for Science to sell meaningful amount of ads?


You could target the ads really well, at least—lasers for the physicists, pipettes for the biologists, chalk for the mathematicians…


Facebook and social media in general is already horrible as is, why would you want to infect academia with it as well? The community does have a lot of issues but a social network and more blog posts are not the solution.

The traditional way is to organize meetups in the way of conferences, workshops, symposia, etc. It has a number of issues since some labs can't afford sending students or postdocs overseas and some countries are disadvantaged (for geographic or visa restriction reasons) but the organizers are usually well-aware of it and e.g. give out grants to cover expenses for students who apply for it.

I think the onus is on English speakers to reach out for the hidden knowledge out there, and not for the majority of the community to adapt to the native English-speaking minority. Right now they don't feel challenged because Anglo-Saxon univeristies are pretty much 'where it's at' and European ones more or less follow suit, but with the rise of Chinese research the situation may be about to change.


> There is a research saying that lying in a foreign language is easier.

I totally agree. I guess since you have to use more those areas involved in language and rational thinking, I find much more easy to say emotional content in a foreign language. You feel more detached of what you are saying.

Tldr: It's easier to say "I love you" or "I hate you" in a foreign language.


So being super rational is not a good thing. Found the research https://www.sciencedaily.com/releases/2018/07/180719142154.h...


I've written various things in English (papers, thesis, abstracts, etc.) and I loathed using that language every step along the way. English is such a clunky language for communication, especially in science.

-It has loads and loads of words, many of them overlapping with subtle differences that go over the head of many non-native speakers yet still carry some kind of difference between intended and perceived meaning. Some of them are just synonymous, such as Latin/Germanic cognate couples (godly vs. divine, tiredness vs. fatigue, etc.), and you just sort of have to learn them all if you want to understand everything you read or hear. So you apply yourself and it turns out that you're still not understood because the vast majority of the community doesn't speak English as a first language either and also has trouble understanding one or the other form you're using.

-Its pronunciation and spelling is completely wild and inconsistent. You can't guess how to pronounce a word you first encounter. And in scientific and technical presentations there are a lot of such words! This doesn't even take into account the various accents within native English speakers (I'm somehow supposed to be able to understand both the Californian and Australian accents no matter how nerve-rattling they sound) and non-native speakers (who also have the same trouble pronouncing new words as I do, except they'll mis-pronounce it differently than I would).

-The way sentence structure is designed favors the apposition of many different words without any preposition, leading to ambiguities. This is of course exacerbated in papers with absurd word count constraints that must somehow fill the needs of print journals despite the fact that 99% of people read them online. It also makes for very terse, opaque and generally tedious reading. The only thing more annoying than reading a paper is writing one, because you're also going to have to write in the same opaque style in order to squeeze as much information as you can in order to fit your 3 year project's worth of discoveries into a 1500 word Nature letter.

-It's unfair that non-native speakers be discriminated on peer-review based on the quality of their English. Reviewers and editors like to pretend they're unbiased but they're not. The quality of the work is independent on the author's ability to write in a foreign language. I'll readily admit that judging someone based on their mastery of a language is not unique to English speakers but this particular situation is especially egregious since most native English speakers do not, in fact, speak any foreign language.

-It's generally a constant reminder of the cultural hegemony held by the US and Anglo-Saxon countries in general. Native speakers are advantaged since they have to learn less and are still less discriminated; scientists flock to Anglo-Saxon universities despite other ones being as competent if not more, but less well-known; the top journals such as Nature or Science are either British or American, reinforcing this bias. Univerisities ratings are often based on publications in these Anglo-Saxon journals, further deepening the bias. And lastly, Anglo-Saxons in general (and Americans in particular) have a very strong Not Invented Here syndrome whereby anything that was discovered outside the anglosphere is disregarded or met with much more skepticism than usual. Every so often an American paper comes out touting a new innovative "amazing" method that will completely revolutionize a field despite the principles having been discovered years before but outside the land of the free. It's frankly annoying.


What is your first language?


Anglo-Saxon? Have you not heard about the Norman conquest?


Other countries can adopt English as an additional official language, and allow children to be educated with English as the language of instruction.

This would be particularly advantageous in Europe, where the population of many smaller countries with their own language is falling. It would also attract inter-EU migrants.


Is there any benefit to having competitive languages?

It seems it would only make communicating ideas harder.

Culture? At what expense?


Competitive languages would hugely hinder the progress. Consider this: Mendel discovered the field of genetics but Darwin wasn't even aware of his work mostly because it was published in German, so we should be glad that the lingua franca of today is a relatively simple language as English, although I might had preferred Anglish if that was available - https://anglish.fandom.com/wiki/Main_leaf


Darwin could definitely read German -- there are quotations from German sources in his work that are left untranslated as he expected that any educated scientist would be able to read it. In the case of Mendel, the issue wasn't that it was in German, but it was in a very obscure journal of a local science society.


Really? You think "uncleftish beholding" [1] is an improvement over "atomic theory"?

[1] https://en.wikipedia.org/wiki/Uncleftish_Beholding


Whatever the problems of having a lingua franca for science are, the problems of not having one are much worse. Weird that Gordin is the only respondent to even mention this.


That is the obvious advantage that everyone knows, only tangentially related to the issue at hand: the challenges of being forced to work in a non-native language.


It's not at all tangential. It is the core of the problem. Any 'solution' or 'improvement' needs to understand and acknowledge the benefits of monolingualism. Look at Dharwadkar, who is calling for breaking up monolingualism without any consideration of the costs, or Sheridan, who advocates for more handholding by professors (because they have so much free time as it is?), or her example of an Indian professor: if the paper is so badly written that it cannot be understood despite many revisions, then how is it supposed to add to scientific knowledge? (It's not like scientific journals have very high standards for prose as it is, so that paper must have been gibberish at the start.) Consider this quote:

> English speakers have become the gatekeepers of science. By keeping those gates closed, we’re missing out on a lot of perspectives and a lot of good research.

OK, so let's say we switch to having everyone publish in their own native language because gosh we wouldn't want to be gatekeepers. Now instead of one 'gate', we have... hundreds, because everyone has to learn every language or else they are being 'gatekept'. Oops.

This pervasive error, this nirvana fallacy, of praising only the benefits of multi-lingualism, renders the entire discussion moot. It's a tissue of complaints and buzzwords.


We have the ability to learn multiple languages, and to work in them. In the history of science, working in multiple languages is the norm.

The idea that every piece of discourse should be in English is a modern phenomenon. It ironically has arrived just as the automated translation renders the interlinguistic literature gap irrelevant.

People should be talking about forcing English speakers to work in their non native language, not just about the woes of a single common language for science.

It is precisely at this moment that we have the ability to gain from cultural diversity without losing intelligibility. Will we go for this possibly? Probably not....


>People should be talking about forcing English speakers to work in their non native language, not just about the woes of a single common language for science.

There are thousands of languages in use worldwide. If English weren't the modern lingua franca, what would stop people from having these same complaints about any other "universal" language?

> It ironically has arrived just as the automated translation renders the interlinguistic literature gap irrelevant.

>It is precisely at this moment that we have the ability to gain from cultural diversity without losing intelligibility

Modern translators have come a long way, but they're far from able to translate scientific texts into intelligible English, especially without losing meaning, particularly from languages which have vastly different grammatical and conceptual styles, like Chinese, for example.

I think it's just en Vogue right now to complain about the Anglo-Saxon hegemony; these problems are not caused by English per se, and the rate of progress is substantially better with this current standard than the alternative you propose of keeping international research effectively siloed by language.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: