English belongs
to its users.
By Adam Frick · Illustrated draftWhy should a textbook make you decode an idiom?
Imagine a textbook that says randomisation “levels the playing field”. You came to learn a research method, not to interpret a sporting image. Why not explain how the method assigns participants to groups by chance? I want English teaching to take that responsibility seriously, especially when the reader does not share the writer’s first language. The worked comparisons below show the kind of practical choice behind my proposal.
See the supporting references.Make the proposal answer to the reader.
Start with the shared audience.
Constructed teaching exampleNative users + Additional-language users
Shared responsibility for understandable English
Teach the meaning of the choice.
Constructed teaching exampleTell me if the room is available.
Tell me whether the room is available.
Separate the finding from the proposal.
Constructed teaching exampleFinding: English has an additional-language majority.
Proposal: That majority should help shape shared teaching conventions.
What has actually changed?
English did not suddenly become international in the twenty-first century. In The Future of English? (1997, pp. 12–13)[9], David Graddol already described a non-native majority and growing communication between non-native users. His English Next (2006)[10] examined the changing demands of that shared language. My argument concerns the expanding settings in which English does everyday work, not the invention of a new majority.
Read the chart data.
| Group | Millions | Calculated share |
|---|---|---|
| Native speakers | 373 | 25.7% |
| Non-native speakers | 1,080 | 74.3% |
When did the majority change?
The historical estimates reach further back than 1997. In English as a Global Language (2003, p. 69, note 15)[14], David Crystal contrasts a 1962 estimate of 250 million first-language users and 100 million additional-language users with estimates from the 1970s of 300 million first-language users and 400 million additional-language users. His review also records much larger additional-language estimates by 1985. These historical estimates establish earlier recognition, not the year a consistently measured population crossed a threshold.
The professional history needs a separate question. John Swales wrote in 1985[15] that English carried most research publications while non-native authors contributed relatively few papers to the available record, especially from developing countries. English dominating publication does not establish a non-native majority among authors, readers or research conversations. In a 2012 paper[16], the research team explicitly describes international business communication as mostly conducted by non-native English speakers, but their research does not date a worldwide crossover. No single global date for a white-collar non-native majority follows from these sources.
By 2008, John Flowerdew described most of the world’s scholars as having a first language other than English[17]. His paper examines the disadvantages experienced by scholars writing in an additional language. The statement documents recognition of that academic majority, but the paper does not establish a worldwide crossover date through a census of scholars or their exchanges.
| Date | Evidence | Limit |
|---|---|---|
| 1970s estimates | Additional-language majority in Crystal’s historical review | No consistently measured crossover year |
| 1985 | English dominates research publication in Swales’s account | No non-native author majority established by that finding |
| 2001 | Seidlhofer challenges native-speaker teaching targets | A research and teaching argument, not an agreed alternative grammar |
| 2012 | Business researchers explicitly describe mostly non-native international communication | No worldwide white-collar census or crossover date |
The university offers a measurable example. A study by Wingrove and colleagues[11] identified 24,043 English-taught degree programmes in non-Anglophone Europe in 2023/24, compared with 8,089 in 2013. The study reports roughly threefold growth and a wider geographical spread. A programme count does not tell us how many people live or work in English, but the change shows why international communication belongs in the classroom.
Read the chart data.
| Period | Programmes |
|---|---|
| 2013 | 8,089 |
| 2023/24 | 24,043 |
A population survey gives a different kind of evidence. The European Commission’s 2024 Eurobarometer release[12], reporting fieldwork from 2023, puts conversational English as a foreign language at 47% of EU respondents, up five percentage points from the report’s 2012 comparison. That finding concerns reported ability, not daily workplace use. The separate increase in daily foreign-language use covers foreign languages generally, so I do not present that increase as an English-only result.
Read the chart data.
| Year | Share | Basis |
|---|---|---|
| 2012 | 42% | Calculated from the reported change |
| 2023 | 47% | Reported survey result |
Workplace research also shows what shared English looks like in practice. In a 2005 study by Louhiala-Salminen and colleagues[13], the researchers examined email and meetings in two corporations formed through Finnish–Swedish mergers. The study examines English used between colleagues with different first languages, rather than a rehearsal for talking to native speakers. The case studies illuminate the communication task but cannot establish a worldwide growth rate.
Can we prove a much greater global proportion than 30 or 50 years ago?
The research cited here does not provide a comparable worldwide series for the proportion of people living and working in non-native English contexts. The available measures count different things, including speakers, reported abilities and degree programmes. Combining those measures into one global percentage would make the argument look stronger while weakening the evidence. The supported claim is substantial expansion in documented settings, alongside an additional-language majority that was already recognised decades ago.
The programme study also warns that earlier growth estimates can be inflated by improved data collection and expanding database coverage. A larger programme total need not imply the same proportional increase in students. My conclusion concerns the need to teach for these shared settings, rather than a claim that every sector or country has changed at the same rate.
A proposal must survive disagreement.
I would keep a technical term when the term gives a subject the precision the reader needs. My proposal asks the writer to explain that term, not flatten the subject into a smaller vocabulary. English already has established grammar, and no universal non-native replacement has been agreed. The teaching conventions I propose should earn acceptance through examples and testing with readers, including readers who disagree with me.
Whose English should a textbook expect?
If a textbook teaches your subject, why should the same book quietly test your familiarity with British idioms or American cultural references? My proposal gives international comprehension a central place in educational writing. A textbook should retain the precise technical term, explain the term and show the relationship through a worked example. The reader should spend effort understanding the idea rather than deciphering an avoidable verbal puzzle.
The teaching debate has a substantial history. In her 2001 paper[7], Barbara Seidlhofer challenged the continued attachment of English teaching to native-speaker norms and called for a description of lingua franca use. Her argument supports examining the target we teach, rather than assuming that a native model answers every educational need. The paper does not supply a universally agreed alternative grammar.
The unequal burden deserves attention beyond the textbook. A 2023 study by Amano and colleagues[18] surveyed 908 environmental researchers and found additional effort for non-native speakers across scientific activities, including reading and writing papers and preparing presentations. The study also reports barriers to conference participation. These findings make the communication burden visible, but the survey does not test whether my proposed textbook conventions reduce that burden.
What should the reader have to learn?
Constructed textbook comparisonAt the end of the day, randomisation levels the playing field.
Randomisation assigns participants to groups by chance. The process helps reduce systematic differences between groups, although chance imbalances can remain.
A consensus should earn acceptance through evidence from its users. My proposed approach needs testing with readers from different language backgrounds, including native speakers. A useful test would ask whether the reader understands the concept and can apply the explanation accurately. A majority provides a reason to reconsider the target, but a majority alone cannot tell us which grammatical convention works best.
References and evidence
- an interview with Spotlight ↩
- British Council’s report The English Effect ↩
- LemonGrad’s English Language Statistics ↩
- 2026 Global Englishes research agenda ↩
- European Commission’s 2024 Eurobarometer publication ↩
- Trudgill. Standard English: what the standard dialect does and does not mean. ↩
- 2001 paper on describing English as a lingua franca ↩
- British Council’s Future of English discussion ↩
- Graddol (1997). The Future of English? Historical framing, especially pp. 12–13. ↩
- Graddol (2006). English Next. Shared English and changing teaching models. ↩
- Wingrove et al. (2025 online; 2026 issue). English-medium instruction in European higher education. Applied Linguistics, 47(2), 193–215. ↩
- European Commission (2024). Eurobarometer release on language ability and use; 2023 fieldwork. ↩
- Louhiala-Salminen, Charles and Kankaanranta (2005). English as a lingua franca in Nordic corporate mergers. English for Specific Purposes, 24(4), 401–421. ↩
- Crystal (2003). English as a Global Language, second edition, p. 69, note 15. Historical estimates, not a comparable census series. ↩
- Swales (1985). English as the International Language of Research. RELC Journal, 16(1), 1–7. ↩
- Louhiala-Salminen and Kankaanranta (2012). Language as an issue in international internal communication. Public Relations Review, 38(2), 262–269. ↩
- Flowerdew (2008). Scholarly writers who use English as an Additional Language: What can Goffman’s “Stigma” tell us? Journal of English for Academic Purposes, 7(2), 77–86. ↩
- Amano et al. (2023). The manifold costs of being a non-native English speaker in science. PLOS Biology, 21(7), e3002184. Survey of 908 environmental researchers. ↩
Read the evidence and its limits.
The earlier version of this essay recorded LemonGrad’s figures attributed to Ethnologue’s 2022 edition. LemonGrad has since updated the linked page to a newer edition. The historical figures below remain dated estimates and are not a current census.
The historical claim concerns that combination and the worldwide scale of English, not the mere existence of a lingua franca. In an interview with Spotlight[1], linguist David Crystal describes the global spread as unprecedented and states: “No language has ever been spoken so widely.” The British Council’s report The English Effect[2] documents the breadth of that international role and the substantial non-native majority. I draw those observations together as the starting point for this manifesto. English has a worldwide constituency, and that constituency should shape what we teach.
LemonGrad’s English Language Statistics[3] reports the following estimates, attributed to Ethnologue’s 25th edition released in 2022. The table preserves the date of those estimates rather than presenting the figures as a new count. The independent sources discussed below provide a broader basis for the manifesto’s argument about a shared audience.
The historical LemonGrad table is a secondary account of figures attributed to Ethnologue. For a more recent research context, the 2026 Global Englishes research agenda[4] locates most English users outside traditional native-speaker contexts and discusses communicative effectiveness as a teaching goal. That source supports the relevance of a shared international audience without providing a new global census. I use the dated table to illustrate the scale of the earlier estimates and the newer research to examine the implications for teaching.
The European Commission’s 2024 Eurobarometer publication[5] reports that 47% of Europeans speak English as a foreign language. The survey concerns the European Union and records reported language ability, so the figure must not be treated as a global speaker share. In relation to my teaching in the Netherlands, the regional finding provides a further reason to prepare learners for communication across first languages. The point is the audience the writing serves, rather than the impression created by a large number.
Standard English does not mean one accent or one uniform way of speaking. In his account of Standard English[6], Peter Trudgill distinguishes the standard dialect from pronunciation and describes most native speakers as acquiring a nonstandard variety. A speaker can use standard grammar with a regional accent, and a regional dialect has grammatical patterns of its own. Everyday variation is not evidence that almost everyone speaks defective English. The problem is the claim that one undifferentiated native model can guide the whole world.
Barbara Seidlhofer’s 2001 paper on describing English as a lingua franca[7] challenged the continued dependence of teaching targets on native-speaker norms. That paper called for research treating lingua franca users as language users in their own right. The subsequent 2026 Global Englishes research agenda[4] emphasises adaptation to interlocutors and communicative effectiveness. My proposal for more explicit teaching conventions is a contribution to that debate, not a claim that the research has endorsed one universal rulebook.
A language used across different communities develops through use. The British Council’s Future of English discussion[8] places English within a multilingual world and relates proficiency to particular contexts and tasks. That perspective changes whose experience deserves attention. Successful communication between additional-language users should inform teaching directly, rather than appear only as a departure from a native model.
The research agenda Global Englishes: Research agendas and tasks in English as a lingua franca[4] (2026) discusses teaching directed towards communicative effectiveness in multilingual settings. My proposal builds on that communicative goal by emphasising shared conventions and explicit meaning. English teaching should prepare learners to communicate with people who bring different first languages to the exchange. The practical goal is to convey an intended meaning, understand another person’s meaning and recognise when further explanation is needed.
| Speaker group | Estimate |
|---|---|
| Native speakers | 373 million |
| Non-native speakers | 1.080 billion |
| Total | 1.453 billion |
The non-native share calculated from these figures is 74.3%. Dividing 1,080 million by 373 million gives approximately 2.9 non-native speakers per native speaker. Neither calculation measures the effectiveness of a proposed grammar rule.