Multilingual provision is cheaper than English-Only

Hiroshi Mikitani, Englishnization in Marketplace 3.0 (Source:

Hiroshi Mikitani, Englishnization in Marketplace 3.0 (Source:

The business and self-help section of my local Kinokuniya bookstore is currently featuring shelves and shelves of Marketplace 3.0: Rewriting the rules of borderless business by Hiroshi Mikitani, the founder and CEO of e-commerce giant Rakuten. I’m not a fan of books in the “how to improve your business” genre and mostly Marketplace 3.0 seems to be a (poorly written) run-of-the-mill text about empowering your business through KPIs. It is also the self-aggrandizing account of an egomaniac. However, the first chapter is absolutely unique to the genre as it is devoted to Rakuten’s language policy: entitled “Englishnization,” it describes how Mikitani decided in May 2010 from one day to the next, “without consulting anyone,” to change the company language from Japanese (and other national languages spoken in subsidiaries outside Japan) to English.

I had, of course, heard about the Rakuten English-Only language policy before in the media so I was keen to read the inside story and to learn what happened after the announcement. However, Mikitani tells the reader at length about his decision to introduce English but is rather coy about what happened next, particularly in the medium term. Mikitani imposed English so that communication within the company would be faster. Not surprising in a world where speed is money. But did it really work? Are they really faster at Rakuten because they speak English? Mikitani only says that the first board meeting after the introduction of English was twice as long as usual leaving the reader to assume that this slow-down was a one-off and communication has significantly sped up since.

How is communicating in English supposed to be faster than communicating in Japanese? To begin with, Mikitani argues that the need for linguistic mediation – translating and interpreting – falls away. Second, as employees gain confidence in English, they no longer send e-mails to the USA but pick up the phone and call. Finally, Mikitani believes that English communication is faster per se because it is supposed to be a more egalitarian language than Japanese and so employees are forced to stop prevaricating in the face of superiors.

“Englishnization” thus is a collection of more or less common language ideologies: linguistic diversity as a barrier to be overcome by an English-Only policy is a throw-back to the Tower of Babel myth and English as more direct is the stock-in-trade of popular forms of linguistic relativism.

So, how does such a radical language policy play out at Rakuten? Well, Mikitani is not saying. He says that Rakuten is much more widely known now because the English-Only policy caught the attention of the global media and even the Harvard Business School. But that is not really due to their English-Only policy per se but to a clever press release about it.

The only indicator of how English at Rakuten is going comes from the claim that 90% of employees had met the language test score required for their level within a year. The remaining 10% were excused because of the 2011 tsunami and given an extension. Personally, I’m guessing that most Rakuten employees had pretty good English to begin with and the few who hadn’t, well, they couldn’t do much about it. Language learning after all is time-consuming business.

That language learning takes time lies at the crux of evaluating a monolingual language policy: if you have a cohort of proficient speakers, then making them use the common language makes perfect sense but if you have to train them up first, then you are not only not speeding things up but slowing them down considerably. This point is unfortunately overlooked in many contemporary decisions to turn transnational institutions into English-Only institutions.

At a time when more and more transnational institutions, similarly to Rakuten, turn to English-Only an assessment of the effectiveness of the European Union’s multilingual provision is particularly timely. Having been committed to multilingual provision through translating and interpreting for many decades, the EU model is often dismissed as too costly and thus inefficient and unaffordable for other transnational organizations such as ASEAN.

Unfortunately, the dismissal of the EU’s multilingual language policy as costly and inefficient is wrong, as Gazzola and Grin explain in a new article in the International Journal of Applied Linguistics. They show that the EU’s language bill for translating and interpreting is Euro 1.1 billion per year. That may seem a lot until you see that this amounts to less than 1% of the EU’s annual budget of €147.2 billion. That means that the expenditure for the EU’s current multilingual language policy is €2.2 per person per year; about the price of a cup of coffee and clearly not so astronomical …

Despite this small amount, it is true that it would be even cheaper if the EU abolished all translating and interpreting and adopted an English-Only policy. Anyone advancing that argument has to bear in mind that the language costs would not be entirely eliminated because language services to make sure all those documents are well written and legally watertight would still be needed. Cost would shift from translators and interpreters to ghost-writers, copy-editors and proof-readers.

But the overall cost could still be brought down from one cup of coffee to say half a cup of coffee?

Well, no, actually not. Under an English-Only regime, most Europeans would be paying much, much more than the equivalent of a cup of coffee for linguistic provision. The British and the Irish would not be paying at all. Those 7% of continental Europeans who already speak “very good” English would not be paying, either. That leaves everyone else – around 80% of Europeans – out of pocket for English language learning if they wanted to exercise their democratic right to understand what is going on in the European parliament and to participate in the European project in any other way. The cost for those 80% of Europeans to bring their English up to scratch would be less for some (those who already have “good” or “modest” English) and astronomical for those adults who have no English – for all these individuals it would be much, much higher than is currently the case.

Turning language costs from a public expense to personal language learning expenses is, of course, totally unfair and undemocratic. Not only would it make participation in the European project contingent about an individual’s financial capacity to invest in language learning, it would also be unfair for continental vis-à-vis English native speakers in Ireland and the UK who would not have to invest in language learning. Already receiving huge language subsidies by everyone learning their language, they could completely withdraw from sharing the costs of linguistic provision in the EU.

Rakuten’s Englishnization includes a similar cost transfer from the company to employees: Mikitani made them learn English in their own time and at their own cost. Viewed this way, Englishnization is nothing more than a billionaire’s trick to socialize cost (language learning) to employees and privatise profits (derived from eliminating linguistic provision) to himself.

The evidence is clear: the EU’s multilingual provision is more cost-effective and fairer than an English-Only policy would be.

Every language policy maker in a multilingual environment should have a copy of Mazzola’s and Grin’s article on their desk. Gazzola, M., & Grin, F. (2013). Is ELF more effective and fair than translation? An evaluation of the EU’s multilingual regime International Journal of Applied Linguistics, 23 (1), 93-107 DOI: 10.1111/ijal.12014

Author Ingrid Piller

Dr Ingrid Piller is Professor of Applied Linguistics at Macquarie University, Sydney, Australia. Ingrid’s research expertise is in the fields of intercultural communication, bilingual education and the sociolinguistics of language learning and multilingualism in the contexts of migration and globalization.

More posts by Ingrid Piller
  • Using Esperanto instead of English would be both fairer and cheaper, since the language is very much more consistent and modular than English.

    It would probably be cheaper than a multilingual approach in some or most settings and, imposing less on the people involved, allows more time for maintenance of linguistic diversity outside of those settings.

    Thanks for an interesting story.

  • Rosemary Kuwahata

    What is this ‘ But the overall cost could still be brought down from one cup of coffee to say half a cup of coffee’ about?
    Translation and interpreting services are worth more than that!
    Having Japanese ability I have so far only used Rakuten in Japanese. The domestic site obviously operates in Japanese only, and it would only be the operators of site set up for overseas customers that are forcefully expected to be operating in English. In their dreams!

  • Excellent points!

    I would add, translation, and the labor of translation, never really go away. It’s not a matter of translation suddenly magically disappearing, translation is always present. It’s a matter of where the translation work is being done and who is doing the translation work — whether it’s an individual with Japanese as a mother tongue who’s become highly proficient in English and is, in her/his head, constantly translating, doing the work of translation, and who has had to invest a lot of time/money into getting to the point of being able to do internal translation, or an external translator, who is performing the labor/work of translation for someone else.

  • Li Jia

    What a thoughtful remark by Gazzola and Grin: the concept of “English as a lingua franca” (ELF) turns out to be “English Only” in terms of the socioeconomic implications! ELF as a new label also frequently appears in Asia (as indicated by the book “English as a Lingua Franca in ASEAN—A Multilingual Model”) especially after English became the “the sole working language” in ASEAN in 2009. However, I really wonder how many percentages of ASEAN citizens can have the full access to the content of documents and democratic participation of ASEAN affairs when fewer number of people’s first language is English compared to the number of EU? What’s more, the labour market rewards with the “native speaker norm” will only bring ELF varieties in ASEAN into a fantasy. Additionally, If I could have access to the figures of English language cost in ASEAN, I believe language budgets cost from “the extended circles” like Laos, Vietnam would be much more than those of “Outer Circles” like Philippine, Singapore. If so, how can citizens in ASEAN have fair play in dealing with their common concerns? How can ASEAN realize their dream of building up “an ASEAN identity”(

  • Pingback: Internationalization and Englishization in Higher Education | Language on the Move()