�����Ƶ

Education is both the lynchpin of sustainable development and a fundamental human right guaranteed in numerous international covenants and declarations, but it is not equally accessible to all. The Incheon Declaration and Framework for Action adopted at the 2015 World Education Forum recognized inclusion and equity in and through education as the cornerstone of a transformative education agenda to be implemented in the United Nations’ 17 Sustainable Development Goals (SDGs) for 2030.1 The fourth goal (SDG-4) pledges to ensure equitable and inclusive quality education and lifelong learning for all. Countries committed to “making the necessary changes in education policies” are required to address exclusion, marginalization, and inequities. To ensure that no one is left behind, they promised that “no education target should be considered met unless met for all.”2 

However, we are already more than halfway to the 2030 deadline, and approximately 244 million children and youth worldwide between the ages of six and eighteen were still missing out on school in 2021. This includes 67 million children of primary school age (about six to eleven years old), 57 million adolescents of lower-secondary school age (about twelve to fourteen years old), and 121 million youth of upper-secondary school age (about fifteen to seventeen years old). Being in school, however, is not the same as learning. Over 600 million children and adolescents worldwide did not attain minimum proficiency levels in reading and mathematics due to the poor quality of schooling, even though two-thirds of them were in school.3 Meanwhile, the COVID-19 pandemic compromised the quality of education for all learners across all countries and magnified existing challenges. 

In this essay, I examine unmet promises and global challenges for achieving language equality in schooling. Although virtually everyone acknowledges clear links between good education and a broad range of benefits impacting poverty, health, and gender inequality, limited recognition of the seminal role language plays as an intervening variable prevents these advantages from reaching the most marginalized.4 With over seven thousand one hundred languages worldwide but only about two hundred countries, there are about thirty-five times as many languages as countries. Bilingualism or multilingualism is present in practically every nation, whether officially recognized or not. Nevertheless, national policies remain radically out of line with the realities of multilingualism in today’s globalized world. Most countries operate as monolingual either de facto (unofficially) or de jure (through legislation) in recognizing only one language for use in education, administration, and public-facing communications. Most languages are thus excluded from education and other higher domains of public life. Policies pursued within national boundaries elevate some languages (and their speakers) to the majority position, while others are relegated to minority status. When a multi­lingual country uses one or more languages exclusively in public schools, as well as in the administration of government services and activities, it makes a distinction based on language. In showing a preference for some language(s), whether designated as official or national or not, the state’s decision benefits those for whom the chosen language(s) is a primary language, to the detriment or disadvantage of others who either have no or lower proficiency and are denied the benefit of using and identifying with their primary language. 

Failure to take language into account means that the goal of education for all embodied in SDG-4 (and in the earlier Education for All agenda from 1990 to 2015) translates into schooling only for some. The poorest speak most of the world’s languages, but have the lowest rates of access to dominant languages at school. Nearly 90 percent of those lacking education in their own language live in economically less developed countries, and 60 percent of out-of-school children live in regions where their own languages are not used in the classroom.5 The result is lost generations of children in the poorest countries whose life chances are irreparably damaged by failure to protect their right to quality education. Enrolling the poorest students will not solve the problem without changing the language of instruction. Policies that discriminate against the languages of the marginalized poor severely compromise the power of education to improve their lives. Unless we change education policy and practice, language minorities will continue to constitute the majority of those still living in poverty beyond 2030. Speaking a minority language in effect creates economic, social, and health risks because ethnolinguistic minorities constitute a large proportion of the bottom 20 percent still living in extreme poverty and suffering from poor health, lack of education, and deteriorating environments. 

Language is the missing link in the global debate on equality and inclusion. Language and education inequalities intersect with socioeconomic status, sex, gender, location, religion, ethnicity, and migration, and accumulate through life and compound over time. As long as education relies mainly on international languages at the expense of local vernaculars, education will reproduce rather than reduce these inequalities, making sustainable and equitable development difficult, if not impossible, to achieve. I argue first language–based multilingual education (L1-based MLE) could be a pathway to schooling equality and sustainable development.

Language diversity is a critical but overlooked variable in understanding who got left behind by the unfinished business and unkept promises of the UN’s eight Millennium Development Goals (MDGs) that expired in 2015, and were later replaced by the SDGs.6 The related Education for All (EFA) agenda also ended that same year, after being superseded by the new Education 2030 Framework for Action, but most countries failed to meet the EFA goals.7 The outcome of both agendas illustrates that globally it is the poor who miss out on school. A vicious circle of intersecting disadvantages pushes language minorities into what economist Paul Collier called the “bottom billion” left behind by development.8 The overwhelming majority of the poor live in two regions—Southern Asia and sub-Saharan Africa—comprising about 80 percent of the global total of people living in extreme poverty. These same poorest regions, not coincidentally, also have the highest number of out-of-school children and lowest literacy rates. Sub-Saharan Africa has the highest out-of-school population and is the only region where this number is still growing; it increased by twenty million to ninety-eight million between 2009 and 2021. One in five primary school–age students are still not in school. Out-of-school rates for adolescents and youth have stagnated since 2010 at 33 percent and 48 percent, respectively.9

Despite the push toward universal primary education in both the MDGs and the EFA, only 52 percent of countries achieved this goal by the 2015 deadline.10 Achieving the even more ambitious SDG-4 goal of universal secondary completion remains challenging for North America and Europe, let alone for lower-income countries. Just before COVID-19 struck, over half of young people were completing secondary school globally. Nevertheless, half of those attending school did not meet minimum proficiency in reading.11 In North America and Europe, 96 percent of students achieve the minimum benchmark for reading by grade 4, but in sub-­Saharan Africa, fewer than 40 percent do. In Central, South, and West Asia and North Africa, fewer than 50 percent do. This points to a critical shortfall in a key learning indicator called “learning poverty,” introduced by the World Bank and UNESCO in 2019 as a measure reflecting the number of children unable to read and understand a simple text by age ten.12 With “business as usual” progress, it would take a century or more for many low-income countries to reach current means set by the Organisation for Economic Co-operation and Development in international assessments like the Programme for the Analysis of Education Systems, Program for International Student Assessment, and Trends in International Mathematics and Science Study. Some countries would never catch up. 

There is a substantial geographic overlap between poverty, educational disadvantage, and language diversity. Table 1 shows the ten countries with the highest numbers of out-of-school youth between ages six and eighteen alongside the number of languages spoken and mean years of schooling. Altogether, these countries host 60 percent of the 244 million out-of-school children, including some of the world’s most linguistically diverse countries such as India, Nigeria, and Indonesia. In Nigeria, Africa’s most populous and ethnolinguistically diverse country, the out-of-school rate has increased among adolescents and youth of secondary school age by 61 percent (from 6.3 to 10.1 million) over the past twenty years. Among primary school–age children, it has increased by 50 percent (from 6.4 to 9.7 million) since 2010. Were more accurate data available, the figures would probably be much worse for countries like Chad (112 languages), Central African Republic (66 languages), Equatorial Guinea (12 languages), Eritrea (9 languages), and South Sudan (62 languages), where estimates suggest that more than 50 percent of primary school–age children are out of school.13 In 2019, for instance, 79 percent of the poorest, 60 percent of girls, and 61 percent of rural children in Chad were out of primary school.14 These countries would add 261 languages, increasing the total to 2,863, or 40 percent of the world’s languages. 

Mean years of schooling is one of three basic dimensions (along with life expectancy and income) in the United Nations’ Human Development Index (HDI), a composite measure of average achievement used to rank countries. These ten countries show a range from 3.2 (Ethiopia) to 8.6 (Indonesia). However, disaggregated national data obscure systematic patterns of discrimination and marginalization for some groups based on sex, wealth, location, and other characteristics that intersect with language. This is true especially for the poorest, for girls, for ethnolinguistic minorities, and for those in rural areas. Indigenous peoples, who make up less than 6 percent of the global population but about 19 percent of the extreme poor and speak up to 60 percent of the world’s languages (many at risk of extinction) are also particularly vulnerable.15 

Gender parity in education has long been regarded as a crucial indicator of overall gender equality. Nevertheless, gender disparities are still among the most entrenched inequalities. The fifth Sustainable Development Goal (SDG-5) aims to achieve gender equality and empower all women and girls, building on targets beginning with the MDGs and EFA. There are also very strong interlinkages between SDG-5 and the other SDGs, especially SDG-4. SDG target 4.5 pledges to “eliminate gender disparities in education.” Despite considerable progress over the past twenty-five years in getting more girls enrolled in school, gender discrimination remains a pervasive problem and a threat to inclusive education. Perhaps the starkest example is Afghanistan, the only country in the world where girls are banned from going to school beyond primary level. Since returning to power in August 2021, the Taliban has imposed a series of increasingly restrictive decrees on girls’ education. 

Globally, nine million primary school–age children (75 percent of the worldwide total) who may never set foot in school are girls, with over four million in sub-Saharan Africa. Fifty million sub-Saharan African girls between six and eighteen years of age account for more than the total number of out-of-school girls of any other region. Women still account for almost two-thirds (515 million) of adults unable to read, a legacy of inequalities and restricted educational opportunities beginning in childhood. If all children entering school after 2000 had achieved basic literacy, adult illiteracy rates would have fallen. Instead, the share of women among illiterate adults has remained unchanged for twenty years and the cycle of intergenerational transmission of education inequality and poverty continues.16

Between 1995 and 2018, we collectively moved in the direction of gender parity in education, with the worldwide percentage of countries achieving this goal rising from 56 percent to 65 percent in primary schools, from 45 percent to 51 percent in lower-secondary schools, and from 13 percent to 24 percent in upper-secondary education. Progress has, however, been uneven across regions, especially among low-income countries in sub-Saharan Africa. Only 2 percent of the poorest rural females in low-income countries complete upper-secondary school.17 Gender disparity in primary school attendance among poor and rural children in low- and middle-income countries is higher than average, mostly at girls’ expense. Various local conditions such as lack of sanitation facilities, potential for sexual abuse or even kidnapping, early pregnancy and marriage, and domestic chores keep girls out of school and/or lead them to drop out.

Minority girls in particular face numerous disadvantages, both as a group and subgroup of the disadvantaged. Nearly three-quarters of out-of-school girls belong to ethnic, religious, linguistic, or other minorities.18 Speaking a minority language and living in a rural area further compound female marginalization. In India, the country with the most out-of-school children, the so-called Scheduled Tribes (or Adivasis, the Indigenous people) are among the most economically disadvantaged and marginalized, encompassing over seven hundred ethnic groups speaking nearly four hundred languages in the seven northeastern states and the so-called central tribal belt from Rajasthan to West Bengal.19 These Indigenous peoples make up 8.6 percent of India’s population, but represent more than 21 percent of out-of-school girls.

Increasing linguistic and cultural diversity arising from migration creates challenges for schooling equality in many countries. The International Organization for Migration regards language as one of the most central aspects for migrants’ inclusion by both the receiving society and migrants themselves.20 Among the most disadvantaged, however, are refugees such as the Rohingya Muslim minority who fled from Myanmar to neighboring Bangladesh, where they now number nearly one million (about half of them children) in refugee camps in Cox’s Bazar District. This is the biggest and most dangerous cluster of refugee camps in the world, with frequent floods, fires, and gang wars. Bangladesh is a poor country with its own numerous linguistic minorities. Over four hundred thousand school-aged Rohingya children urgently need education; in 2019, at least one-third were not in any kind of school program. The Rohingya language lacks a widely accepted written standard and Rohingya people have low levels of literacy. In Myanmar, before arriving in Bangladesh, Rohingya people faced virulent education discrimination: Myanmar restricted primary and secondary education for Rohingya people and banned them from universities. Meanwhile, Bangladesh denies Rohingya people access to the national education system. Rohingya are also not allowed to use the Bangladesh national curriculum, use Bangla (Bangladesh’s national language) as a language of instruction, or provide any written material in Bangla. Their only option is unaccredited informal education services provided by international, nongovernment, and private organizations, some of which offer the Myanmar curriculum.

These policies leave the Rohingya people caught in a dilemma. The longer they stay in Bangladesh, the greater their need for Bangla and less for Myanmar (also referred to as Burmese) unless they return. Most do not want to return until their safety is guaranteed, which is unlikely following the military coup in 2021. This is another example of a perfect storm in which language inequalities tied to socioeconomic status, location, religion, ethnicity, and language accumulate through life and compound over time. In Myanmar, Rohingya people lack legal status and citizenship, while in Bangladesh, they are also marginalized, not recognized as refugees, and denied integration into society.21

Secretary-General of the United Nations António Guterres called the impact of the COVID-19 pandemic on schooling a “generational catastrophe.”22 Even before the onset of the most serious crisis ever to hit world education systems all at once, progress in reducing the out-of-school population had slowed. School closures impacted nearly 1.6 billion learners from pre-primary to secondary education in at least 194 countries, leaving education systems reeling from its effects worldwide.23 This unprecedented disruption exacerbated substantial pre-existing inequalities, both between and within countries. Those most vulnerable and marginalized before COVID-19 (that is, girls, the poorest, the disabled, those living in rural areas, refugees and migrants, and those speaking languages other than those used in formal education) face increased exclusion and learning loss compared to their more advantaged peers. An additional eleven million girls and young women may never return to school.24 Speakers of minority languages were also disproportionately affected because emergency education provision tends to be provided only in major national or international languages. The massive shift to remote learning also underlined inequities in access and the need for linguistic diversity in the digital domain.25 The lack of multilingualism in cyberspace poses significant barriers to digital inclusion. Wide gender disparities in information and communication technologies also remain.

It is still too soon to know the long-term effects of COVID-19 on global education and other development goals. Official SDG statistics still largely reflect the pre-pandemic situation. Although extreme wealth has been rising for many years, extreme poverty was also falling. The arrival of COVID-19 reversed this trajectory, marking the first increase in extreme poverty in over two decades. Indeed, HDI has declined two years in a row for the first time since calculation began thirty-two years ago. This not only erased gains from the preceding five years, but also reversed much of the progress toward the SDGs.26 It seems extremely unlikely that we will be able to achieve the first SDG—end poverty in all its forms everywhere—by 2030. The share of children living in learning poverty in low- and middle-income countries was already over 57 percent before 2020, and will probably rise sharply, potentially up to 70 percent. In poor countries, the level was already over 80 percent; in sub-Saharan Africa, it was 86 percent. As a result of school closures, this generation now risks losing USD 17 trillion in lifetime earnings, equivalent to 14 percent of today’s global gross domestic product.27

Providing quality education to the poorest children requires teaching them through the language they understand best. Nevertheless, this commonsense principle is still the exception rather than the rule worldwide. For decades, UNESCO has been at the forefront of advocating first-language education, but estimates suggest that at least 40 percent of children worldwide still lack access to education in their own languages.28 Literacy provides an indispensable foundation for lifelong learning, and is therefore key to sustainable development. SDG indicator 4.6.1 aims to “ensure that all youth and a substantial proportion of adults, both men and women, achieve literacy and numeracy.”29 Learning to read through an unfamiliar language imposes a double burden because learners must acquire the new language and content simultaneously. Children can more easily acquire literacy in languages they already know. L1-based programs can produce competent readers in two to three years, rather than the five or more years typical of many second-language programs. In countries where children average five or fewer years of school, such as Pakistan (4.5 years), Ethiopia (3.2 years), and Sudan (3.8 years), and the poorest average even fewer, L1-based programs represent the only option for the majority of nondominant-language speakers attending school to achieve even modest levels of literacy. In Guinea, only 26 percent of girls whose home language was not the same as that used at school reached the minimum proficiency in reading by the end of their years in primary school in 2019.30

Research shows that the more developed children’s literacy skills are in their first languages, the more prepared they will be to acquire second languages successfully. After only three years of L1-based schooling, children are not ready to learn through a language they do not understand. Many are still learning the alphabet in grade 3, so the first two grades are in practice lost years for learning the content of the curriculum. Early exit models of L1-based instruction that transition to English or other international languages in grade 3 do not give students enough time to develop literacy skills in their own language that can later be transferred to learning other languages. In addition, many textbooks are written for fluent native speakers and are not adapted to the special needs of those learning English or other dominant languages as second languages. To be maximally effective, L1-based MLE must be high in quality and adequately resourced with well-trained teachers and materials.

Educating children in languages they do not understand results in poor outcomes. Consider Africa, the linguistically richest but economically poorest region on earth, with one-third of the world’s languages and nearly one-third of its population in extreme poverty, surviving on less than USD 1.90 per day. With the exception of Eritrea, Ethiopia, and Tanzania, no country in sub-Saharan Africa provides the entire span of primary schooling in a local language, let alone children’s first languages. None offers secondary schooling or higher education in an indigenous language. While multilingualism prevails, colonial languages still persist.31 Twenty-seven of the thirty-one countries with the lowest HDI are in sub-Saharan Africa.32 Here students receive on average only 4.9 years of schooling, representing a range of values from six to seven years in Tanzania, Lesotho, Democratic Republic of Congo, and Nigeria, and fewer than three years in Burkina Faso, Chad, Guinea, Mali, Niger, and Senegal. With the exception of Lesotho and Mali, all of these countries also show gender disparities, with girls receiving fewer years of education than boys. 

Education scholar Birgit Brock-Utne regards language of instruction in Africa as the most important and least appreciated education issue.33 Across a continent with very high repetition and drop-out rates and fewer than 50 percent of African pupils remaining to the end of primary school, more than five decades of instruction through English (or other European languages like French and Portuguese) has done and can do little for most students: 80 percent to 90 percent of the population still has not learned European languages. Even in South Africa, where English has been a school subject for more than one hundred years and is widely spoken in larger society, proficiency is still very low among the poorest, predominantly Black populations speaking African languages.34 Countries that do not provide access to L1 education experience the lowest levels of literacy and educational attainment worldwide.35 Income inequality is also significantly higher in countries using colonial languages as the medium of instruction.36 Data from UNESCO indicate that use of an international language of instruction instead of local languages is “associated with higher inequality in the distribution of learning outcomes and lower performance of learners from the poorest households.”37

The effectiveness of L1-based programs represents only half the story of remedying deficiencies in delivering quality education to linguistically diverse populations. The other half is the continued failure of English to achieve the promises and hopes held out for it as a language of opportunity. The global rush to adopt English as a medium of education at increasingly earlier ages virtually guarantees that most children in the poorest countries, especially in Africa and South Asia, will be left behind.38 Even in multilingual countries in the Global South currently implementing some form of multilingual education, early exit is the most common choice.39 Although even a few years can give some students an advantage, programs most likely to facilitate successful transition to learning in a second language in secondary school require a minimum of six years of instruction through the first language.40 In sub-Saharan Africa, with school conditions far from optimal, as many as eight years of instruction in African languages may be needed. English falls short of being the promised gateway to the global knowledge economy in countries where few know English and the chances of acquiring it at school are limited due to inadequate resources and teaching. Switching to English as the sole language of instruction will not guarantee the supposed benefits of participation in the global economy for the majority of students from impoverished rural communities. Submersion models plunging children into a second language with no instruction or support in their first language are a recipe for persistent, if not permanent, underdevelopment. They will continue to produce a large underclass of almost 90 percent who will finish below the mean, with insufficient skills for most work but manual labor.41

Investing in the development of local languages in the context of high-quality, well-resourced L1-based MLE lays the foundation for sound economic policy for promoting long-term sustainable development. At first glance, it might seem easier and more cost-effective to immerse children as early as possible in the national and/or international languages they will eventually need for accessing wider opportunities and participating in national life beyond their communities, especially when school provides the only context for learning them. The added expense of moving from monolingual to multilingual education is much less than commonly believed. Where evaluations exist, they suggest additional costs of 3 percent to 4 percent above that of monolingual schooling. This estimate does not take into account that poverty is also expensive in terms of human costs and lost resources. Using more of children’s first language in school is likely to lead to more effective learning of additional languages and to reduced repetition and dropout rates, resulting in significant cost savings to education budgets. Political scientist David Laitin and economist Rajesh Ramachandran estimate that if a country like Zambia adopted Mambwe instead of English as its official language, its Human Development Index ranking would move up forty-four positions to reach a level of development similar to that of Paraguay. In 1994, Ethiopia introduced first-language instruction, which has had a positive effect at all levels of schooling, leading to a 12 percent increase in the number of students completing six years or more of schooling.42 

Just and socially inclusive language policies will generate economic benefits. One year of schooling increases earnings by 10 percent on average. In sub-Saharan Africa, returns are highest on average (12.5 percent) and even more for girls (nearly 14 percent).43 Educating girls and women is one of the best investments a country can make to break the intergenerational poverty cycle. Indeed, developing countries can gain the largest economic and social advantages. Where income and school levels are lower, girls and women potentially reap greater benefits, especially from completing secondary education. Conversely, failure to educate girls can lead to substantial losses between USD 15 trillion and USD 30 trillion in national productivity and wealth.44 

Misguided policies preventing multilingual education from reaching the most linguistically diverse populations have never realized a positive return on investment in educational, social, or economic terms despite significant financial resources funneled into them. Low proficiency in the language of instruction is associated with poor attendance, lower learning outcomes, higher dropout risk, and lower transition rates to higher grades.45 Countries tolerating high levels of educational and gender inequality ultimately pay a high price. Failing to educate large numbers of young people results in unemployment, lost earnings, hopelessness, and instability. Being out of school has repercussions through the lifetime of individuals and across generations, as educational disadvantage is transferred from parents to children. 

I have focused primarily on inequities tied to language in developing countries in the Global South, but poor school achievement of speakers of languages other than the official and national languages recognized for instruction is well documented in virtually all nations.46 Full exercise of the right to education depends on the right to language. Failure to recognize language and language diversity as an equity issue during both the formulation and implementation of the SDGs (and the expired MDGs and EFA) has disproportionate effects on vulnerable populations, key stakeholders for successful achievement of these agendas. Although countries recommitted to achieving progress by 2025 and 2030 on seven SDG-4 benchmarking indicators (early childhood education attendance, out-of-school rates, completion rates, gender gaps in completion rates, minimum proficiency rates in reading and mathematics, trained teachers, and public education expenditure), SDG-4 will not be achieved by 2030 even if countries meet their benchmarks.47 Priority should be given to SDG indicator 4.5.2, which is “the percentage of students in primary education whose first or home language is the language of instruction.”48 Simply allocating more resources to education without ensuring that they are equitably spent will not suffice. Reaching the most marginalized will also cost more. We need to prioritize poor countries and earmark funds for multilingual education. 

Overall, aid to education has been declining and is far too low to meet SDG targets. Donors do not give enough; nor do they allocate funds to those needing them most. Estimates of the gap between what developed countries provide now and what is needed by 2030 are as high as USD 97 billion a year. Educational policy scholar Stephen Klees considers this an underestimate because it does not include all SDG-4 targets, nor does it consider the amount needed for other SDGs.49 Indeed, he concludes that all the SDGs are already failures due to the unwillingness of the international community and national governments to finance them. While fulfilling all SDGs would require between 1 percent and 4 percent of global GDP, this will not happen without a drastic alteration of neoliberal capitalism. This echoes my previous call for radical rethinking of the SDGs and prevailing models of development. Relying on economic growth to eliminate poverty is environmentally unsustainable. Increasing the share of the benefits of global growth to the world’s poorest would require dramatically curtailing the consumption of the rich. Instead, inequality is increasing at exponential rates.50

Meanwhile, there is no critical information concerning potential impacts of different educational policies and choices upon children, community welfare, and national development, and therefore, popular myths about multilingualism being divisive, expensive, and detrimental to economic development prevail. Many countries continue to make poor policy choices through ignorance, misguided political ideologies, poor governance, corruption, and military conflict. In 2019, for instance, Rwanda, an extraordinarily linguistically homogeneous country by African standards, where nearly all speak Kinyarwanda, the national language, changed its language policy abruptly for the third time in eleven years. Until 2008, schools used Kinyarwanda for instruction during the first three primary years before switching to French, the former colonial language. Then the government implemented a sudden change to English as the sole medium of instruction, leaving schools with fewer than four months to prepare. Only forty-seven hundred (15 percent) of the country’s thirty-one thousand primary school teachers and six hundred (5 percent) of its twelve thousand secondary school teachers had been trained in English. At the time, French and English were spoken by only an estimated 8 percent and 4 percent of the population, respectively. By 2011, when it was clear that children and teachers were struggling, schools were allowed to return to Kinyarwanda for the first three years. In 2019, however, the ministry of education decided to introduce English as a language of instruction from grade 1 in all schools rather than continuing Kinyarwanda as the main language of instruction at least through the primary school years, with English gradually phased in as a subject, which would make far more sense in both the short and long term. The new policy has already had a detrimental effect, with a drop in school enrollment rates at both primary and secondary levels. The completion rate for primary school declined sharply six years after the language change when the first cohort affected by the new policy should have finished primary school.51 The quality of schooling has also suffered, with 85 percent of students ranked “below comprehension” in reading at the end of grade 3.52

In India, education for Indigenous, tribal, and other minority children operates in contradiction to both India’s constitution and its 2020 national education policy. Article 350-A of the constitution exhorts states to provide instruction through the mother tongue for primary education, but article 351 recognizes only twenty-two languages for official use.53 The national education policy affirms a commitment to using children’s home languages and mother tongues as the medium of instruction preferably until grade 8 and beyond. In practice, however, a gap between policy and practice deprives most Indigenous, tribal, and other minority children of education in their own languages.54 The actual number of languages used for teaching/learning, medium of instruction, or school subjects has declined over the years. In higher education, even regional majority languages are only minimally present; tribal languages are completely absent. University and technical education are almost exclusively in English.55

Large military budgets and defense spending in both rich and poor countries also divert money that could have been more wisely spent on education. International military spending dwarfs the amount of development assistance for education. One-fifth of Pakistan’s military budget would suffice to finance universal primary education.56 Nigeria is Africa’s richest country. A greater investment in education could yield a higher return for peace and stability than equivalent military spending. More than half of Nigerian girls in the less developed and conflict-ridden northeast and northwest of the country are not in school. The northeast region (in particular Borno State) is one of the poorest, with nearly 75 percent (or just over 1.4 million) out-of-school children. Boko Haram (usually translated as “Western education is forbidden”) is the main driver of attacks against education. A UNICEF-supported intervention providing first-language instruction to over nine thousand students across grades 4 to 6 achieved impressive results in both reading and mathematics in only three months. The baseline of 14 percent of children in grades 4 to 6 able to read a paragraph of four lines in Hausa, the largest minority language, increased by 31 percentage points to 45 percent.57 Programs such as these need to be scaled up across Nigeria and the African continent.

There may be grounds for optimism, both regionally and internationally. In 2022, Nigeria approved a new national language policy providing for first-language instruction throughout the six years of primary school. Given the country’s size and ethnolinguistic diversity, implementation will be challenging, but could benefit millions of children and the country as a whole. Similarly, Botswana’s ministry of basic education promised to introduce eleven new local languages for instruction (in addition to English and Setswana) in January 2022, but still has not yet implemented the policy. Unlike Nigeria’s policy, however, Botswana’s is an early exit model, providing only three years from pre-primary level until grade 2. Meanwhile, a recent World Bank policy paper recommended actively championing and leading the way on good language-of-instruction policies because they promote human capital accumulation and are therefore of acute concern to national policymakers and development partners.58 As the largest funder of education in the developing world, the World Bank could prioritize allocation of resources for L1-based MLE and put pressure on ministries of education to adopt sound language policies. 

In 2015, countries pledged to make changes in education policies to address exclusion, marginalization, and inequities as part of a transformative education agenda to be implemented in the United Nations’ SDGs. Despite encouraging developments in some countries, education in many parts of the world still operates in ways that contradict best practices recognized more than seventy years ago by UNESCO, supported by a substantial body of research on the benefits of L1-based MLE. I have provided empirical evidence in support of a significant geographic overlap between poverty, educational disadvantage, and language diversity. There can be no true development without linguistic development. Use of local languages is inseparable from participatory development. Exclusionary policies, no matter how well funded, will not work. The continuation of educational policies favoring international languages at the expense of local ones is part of the development fiasco. The social and economic costs of inequities in differential access to good-quality education are high indeed, with the heaviest burden falling on the poorest, girls, ethnolinguistic minorities, and those living in rural areas. Achieving equality and inclusion will not be possible so long as development agendas continue to ignore language of instruction.

author’s note

I would like to thank Carol Benson for comments on an earlier version of this essay.

Endnotes

• 1United Nations, “” (accessed June 21, 2023).
• 2UNESCO, Education 2030 Incheon Declaration and Framework for Action (UNESCO, 2015), iv.
• 3UNESCO Institute for Statistics, “New Estimation Confirms Out-of-School Population Is Growing in Sub-Saharan Africa,” Fact Sheet No. 62/Policy Paper 48 (UNESCO, 2022), 3.
• 4Suzanne Romaine, “Keeping the Promise of the Millennium Development Goals: Why Language Matters,” Applied Linguistics Review 4 (1) (2013): 1–21; and Suzanne Romaine, “Global Language Justice Inside the Doughnut: A Planetary Perspective,” in Global Language Justice, ed. Lidia L. Liu, Anupama Rao, and Charlotte A. Silverman (Columbia University Press, 2023), 68–95.
• 5Stephen L. Walter, “The Language of Instruction Issue: Framing an Empirical Perspective,” in The Handbook of Educational Linguistics, ed. Bernard Spolsky and Francis Hult (Blackwell, 2008), 129–146.
• 6United Nations, “” (accessed July 7, 2023).
• 7UNESCO, “” in Glossary of Curriculum Terminology (accessed July 7, 2023).
• 8Paul Collier, The Bottom Billion: Why the Poorest Countries Are Failing and What Can Be Done about It (Oxford University Press, 2007).
• 9UNESCO Institute for Statistics, “New Estimation Confirms Out-of-School Population Is Growing in Sub-Saharan Africa,” 4.
• 10UNESCO, Education for All 2000–2015: Achievements and Challenges (UNESCO, 2015), 225.
• 11Global Education Monitoring Report Team, Non-State Actors in Education: Who Chooses? Who Loses? (UNESCO, 2012), 212.
• 12This measure can be criticized for its emphasis on deficit rather than achievement. The findings are not surprising because students are assessed in the language of instruction, which for most is different from their first language. See The World Bank, “” (accessed July 7, 2023).
• 13UNESCO Institute for Statistics, “New Estimation Confirms Out-of-School Population Is Growing in Sub-Saharan Africa,” 7.
• 14UNESCO, “” (accessed July 10, 2023).
• 15United Nations Human Rights Office of the High Commissioner, “” (accessed July 7, 2023).
• 16Global Education Monitoring Report Team, A New Generation: 25 Years of Efforts for Gender Equality in Education (UNESCO, 2020).
• 17Ibid., 6, 9, 11.
• 18Maureen Lewis and Marlaine Lockheed, Inexcusable Absence: Why 60 Million Girls Still Aren’t in School and What To Do about It (Center for Global Development, 2006).
• 19Susmita Mitra, Sudipta Kumar Mishra, and Rajesh Kumar Abhay, “Out-of-School Girls in India: A Study of Socioeconomic-Spatial Disparities,” GeoJournal 88 (2023): 341–357. For discussion of the marginalization of Indigenous/tribal minorities in India, see Ajit K. Mohanty, The Multilingual Reality: Living with Languages (Multilingual Matters, 2018).
• 20

  International Organization for Migration, World Migration Report 2020 (International Organization for Migration, 2019), 192.

• 21

  Mahbub Alam Prodip, “Education for What? The Politics of Education for Rohingya Refugees in Bangladesh,” Politics & Policy (2023): 1–18; and United Nations High Commissioner for Refugees, “Rohingya Refugee Response in Bangladesh: Education Fact Sheet,” December 31, 2022.

• 22UNESCO, “,” August 6, 2020, last updated April 20, 2023.
• 23UNESCO, When Schools Shut: Gendered Impacts of COVID-19 School Closures (UNESCO, 2021), 29.
• 24Global Education Monitoring Report Team, A New Generation, 6.
• 25International Bank for Reconstruction and Development and The World Bank, LOUD AND CLEAR: Effective Language of Instruction Policies for Learning (World Bank Group, 2021), 28; and UNESCO, When Schools Shut, 36–37.
• 26Pedro ��DzԳ���çã��, Uncertain Times, Unsettled Lives: Shaping Our Future in a Transforming World (United Nations Development Program, 2022).
• 27International Bank for Reconstruction and Development and The World Bank, LOUD AND CLEAR, 5; and UNICEF, The State of Global Learning Poverty: 2022 Update (UNICEF, 2022).
• 28UNESCO, The Use of Vernacular Languages in Education (UNESCO, 1953); and UNESCO, “If You Don’t Understand, How Can You Learn?” Policy Paper 24 (UNESCO, 2016).
• 29Ze Yar Min, “,” UN Statistics Wiki.
• 30UNESCO, “World Inequality Database on Education.”
• 31Michael H. Kretzer and Russell H. Kaschula, “A Comparative Analysis of Language Policy and Planning in the SADC-Region,” in Handbook of Language Policy and Education in Countries of the Southern African Development Community (SADC), ed. Michael H. Kretzer and Russell H. Kaschula (Brill, 2022), 11–40.
• 32��DzԳ���çã��, Uncertain Times, Unsettled Lives, “Table 1,” 274.
• 33Birgit Brock-Utne, “Language of Instruction in Africa—The Most Important and Least Appreciated Issue,” International Journal of Educational Development in Africa 1 (1) (2014): 4–18.
• 34Kathleen Heugh, “Multilingualism and Education in Africa,” in The Cambridge Handbook of African Linguistics, ed. Ekkehard Wolff (Cambridge University Press, 2019), 577–640.
• 35Walter, “The Language of Instruction Issue.”
• 36Gary Coyne, “Language Education Policies and Inequality in Africa: Cross-National Empirical Evidence,” Comparative Education Review 59 (4) (2015): 619–637.
• 37Global Education Monitoring Report Team, Inclusion and Education: All Means All (UNESCO, 2020), 221.
• 38Suzanne Romaine, “Linguistic Diversity and Global English: The Pushmi-Pullyu of Language Policy and Political Economy,” in Language Policy and Political Economy: English in a Global Context, ed. Thomas Ricento (Oxford University Press, 2015), 252–275.
• 39Barbara Trudell, Early-Exit Language Transitioning Programming: The Rationale, the Benefits and the Limitations (UNESCO, 2023).
• 40Leila Schroeder, Megan Mercado, and Barbara Trudell, “Research in Multilingual Learning in Africa: Assessing the Effectiveness of Multilingual Education Programming,” in Multilingual Learning and Language Supportive Pedagogies in Sub-Saharan Africa, ed. Elizabeth J. Erling, John Clegg, Casmir M. Rubagumya, and Colin Reilly (Routledge, 2022), 33–61.
• 41Stephen L. Walter and Carol Benson, “Language Policy and Medium of Instruction in Formal Education,” in The Cambridge Handbook of Language Policy, ed. Bernard Spolsky (Cambridge University Press, 2012), 278–300.
• 42David D. Laitin and Rajesh Ramachandran, “Language Policy and Human Development,” American Political Science Review 110 (3) (2016): 458.
• 43Henry A. Patrinos, “Estimating the Return to Schooling using the Mincer Equation,” IZA World of Labor 278 (2016): 1–11.
• 44Quentin Wodon, Claudio Montenegro, Hoa Nguyen, and Adenike Onagoruwa, Missed Opportunities: The High Cost of Not Educating Girls (The World Bank, 2018).
• 45Lizzie O. Milligan and Laela Adamson, eds., Girls’ Education and Language of Instruction: An Extended Policy Brief (University of Bath Institute for Policy Research, 2022); and Carol Benson, Girls, Educational Equity and Mother Tongue–Based Teaching (UNESCO, 2005).
• 46James W. Tollefson and Amy B. M. Tsui, eds., Medium of Instruction Policies: Which Agenda? Whose Agenda? (L. Erlbaum, 2004).
• 47Global Education Monitoring Report Team, Setting Commitments: National SDG 4 Benchmarks to Transform Education (UNESCO, 2022), 2, 13.
• 48Global Education Monitoring Report Team, Measuring Thematic Indicator 4.5.2: Challenges and Alternatives (UNESCO, 2021).
• 49Steven J. Klees, “Why SDG-4 and the Other SDGs Are Failing and What Needs to be Done,” International Journal of Educational Development 104 (2024): 1–4.
• 50Ibid.; Global Education Monitoring Report Team, “Can Countries Afford Their National SDG-4 Benchmarks?” (UNESCO, 2023), 7; Suzanne Romaine, “Global Language Justice Inside the Doughnut: A Planetary Perspective,” in Global Language Justice, ed. Lidia L. Liu, Anupama Rao, and Charlotte A. Silverman (Columbia University Press, 2023), 68–95; Martin-Brehm Christensen, Christian Hallum, Alex Maitland, et al., “Survival of the Richest: How We Must Tax the Super-Rich Now to Fight Inequality” (Oxfam, 2023); and Daniel A. Wagner, Learning as Development: Rethinking International Education in a Changing World (Routledge, 2017).
• 51Dieudonné T. Munsi, “Language of Instruction as Education Policy and Its Effects on the Socio-Economic Outcomes of Young People” (PhD diss., Syddansk Universitet, 2023).
• 52The World Bank, Future Drivers of Growth in Rwanda: Innovation, Integration, Agglomeration, and Competition (International Bank for Reconstruction and Development and The World Bank, 2020), 24.
• 53From Article 350-A: “It shall be the endeavour of every State and of every local authority within the State to provide adequate facilities for instruction in the mother-tongue at the primary stage of education to children belonging to linguistic minority groups.” The Constitution of India, Article 350-A, “” (accessed February 10, 2024).
• 54The National Education Policy Provision 4.11 Multilingualism and the power of language states: “Wherever possible, the medium of instruction until at least Grade 5, but preferably till Grade 8 and beyond, will be the home language/mother tongue/local language/regional language. Thereafter, the home/local language shall continue to be taught as a language wherever possible. This will be followed by both public and private schools. High-quality textbooks, including in science, will be made available in home languages/mother tongue. All efforts will be made early on to ensure that any gaps that exist between the language spoken by the child and the medium of teaching are bridged.” The National Education Policy 2020 (Ministry of Human Resource Development, Government of India, 2020), 13.
• 55For a discussion of Indian language policies, see Ajit K. Mohanty, The Multilingual Reality: Living with Languages (Multilingual Matters, 2018).
• 56UNESCO, EFA Global Monitoring Report 2011, The Hidden Crisis: Armed Conflict and Education (UNESCO, 2011), 148–149.
• 57UNICEF, Every Child Learns:Global Annual Results Report (UNICEF, 2019), 38; and UNICEF, Nigeria: Every Child Learns, UNICEF Education Strategy 2019–2030 (UNICEF, 2019).
• 58International Bank for Reconstruction and Development and The World Bank, LOUD AND CLEAR.