In the Programme for International Student Assessment (Pisa) international competition for children’s learning, India came 72nd out of 73 countries. The Annual Status of Education Report (Aser 2011) reveals that the proportion of Class 5 children able to read a Class 2 text has fallen from 53.7% in 2010 to 48.2% in 2011. The proportion of Class 3 students able to do simple subtraction sums is down from 36.3% to 29.9%. India is bidding to be a super-dunce rather than superpower.

The government’s two flagship programmes, Sarva Shiksha Abhiyan and the Right to Education Act, emphasize huge spending on school infrastructure and teacher training. But learning outcomes have not improved at all despite Rs 100,000 crore of extra spending in the last five years.

One answer is to make teachers more accountable for results. But politicians of all parties fear teachers’ unions, and dare not discipline them. Instead, politicians pretend that higher spending is the solution.

Desperate parents find free governments schools so bad that they have shifted massively to private schooling. The proportion of children in private schools is up from 18.7% to 25.6%, with another 26% going for private tuitions. The puzzle is that not even the huge shift to private paid education shows up in improved learning outcomes.

Many research studies (eg. Muralidharan and Kremer, Tooley and Dixon) have found that private schools yield better learning outcomes. But this is not vindicated at the national level in ASER’s surveys. Can so many researchers be wrong? Can millions of parents be wrong?

We need more research to find answers. Maybe research studies so far have focused on relatively strong states. Maybe outcomes have improved in higher classes but not yet in elementary schools. Maybe the shift to private schools will improve outcomes with a lag.

A fourth possibility is that states and private schools have shifted prematurely to teaching English, or teaching in the English medium. Poor parents desperately want their children to learn English. Thousands of private schools have come up, especially in the Hindi belt, with outlandish names like Saint Convent School, or even Popatlal Convent School. But not even the teachers in such schools know decent English.

Some state governments have made English compulsory in Class 1. But premature teaching of English may worsen all learning.

Educationist Helen Abadzi writes “people must be able to read one word per second, or per 1.5 seconds at the outside, to be functional readers. If they read more slowly than that, they find that they have forgotten the beginning of their sentence by the time they reach the end… If they cannot read fast enough, then all their mental attention is taken up in decoding the letters…If a child cannot read quickly, it cannot follow what text books or teachers are conveying. All schooling can bypass such children. They can spend eight years in school and remain functionally illiterate.” This sound so much like India!

Elite children enter school with a vocabulary of 3,000 words, and can read easily. But poor children start with a vocabulary of just 500 words, and struggle to read. Teaching English in Class 1 compounds the problem. Here again elite children speaking English at home can easily cope. But a new language devastates poor children, who struggle to read even in their mother tongue. They may fail to acquire the minimum required speed (one word per second) for comprehension. Such children will obviously have poor learning outcomes.

But children who learn to read fluently in their mother tongue can easily learn English later. In an experiment in Zambia, some children were taught both English and the local language in Class1, while others started English writing only in Class 2. The difference was astounding. The first group had reading scores two grades below the standard benchmark in English, and three grades lower in the local language. But where English was introduced later, English reading and writing scores shot up 575% above the benchmark in Class 1, 2,417% in Class 2, and 3,300 % in Class 3. Scores in the local language showed leaped up too. This system was extended to all Zambian schools.

Lesson: start with mother tongue, and introduce English later. The issue in India is not Hindi vs English. Rather, a good Hindi foundation improves English learning later. The trap to avoid is premature introduction of English, or teaching in the English medium to children who don’t speak it.

Is premature English teaching an important reason for the failure of high government spending and private education to improve learning outcomes? I suspect so. But we need fresh research and experimentation to find out.