The Roster on the Staff Room Wall Is the Real Curriculum

The duty roster in the staff room of a government upper-primary school I visit in the eastern part of the state is a sheet of ruled paper in a plastic sleeve, updated in three different pens. Last year it recorded, for one teacher, forty-one days: booth-level officer work for the electoral roll revision in three separate spells, a fortnight of census enumeration training and fieldwork, four days on a pulse polio drive, two days counting cattle, one day escorting children to a district function, and a scattering of single days whose reason column simply says office.

She teaches mathematics to classes six, seven and eight. There is no other mathematics teacher. Forty-one days is a fifth of the instructional year.

Every one of those days she was marked present.

That is the sentence I would put in front of anyone who wants to argue about Indian schooling, because it explains why the argument keeps missing. We have spent fifteen years building a national conversation about teacher absenteeism — surprise inspections, biometric attendance, dashboards, an entire literature. Absenteeism is a real problem and I am not going to defend anyone who takes a salary and does not turn up. But the largest single subtraction from teaching time in the government system is not teachers being absent. It is teachers being present, on duty, and elsewhere, and every measurement instrument we have built records that as attendance.

The state uses its teachers as its general-purpose field workforce. It always has. When the district needs someone literate, locally resident, trained to follow a form and answerable to a government hierarchy, in every village at once, there is exactly one such body of people and it is standing in front of classrooms. Elections need them. The census needs them. Vaccination drives need them. Cattle, apparently, need them.

I want to give this its full weight rather than the sneer it usually gets, because the alternative view is not stupid and I have argued it myself. An election in a country of this size runs on roughly a million polling stations. It runs on time, it runs honestly enough that the losing side concedes, and it runs because a schoolteacher in every village spends her February on the electoral roll. That is not a trivial good. It may be the most impressive administrative achievement of the republic, and it is purchased with teaching days. Somebody who wants those days back is obliged to say who does the roll instead, and the honest answer is that nobody has proposed one that would work by next February.

But look at how the cost is currently handled, which is: not at all. There is no line anywhere that says this school lost forty-one days of mathematics. The teacher is present in the attendance system. The school is open in the school system. The syllabus is unchanged. The examination in March asks the same questions of these children as of the children in the town school forty kilometres away whose mathematics teacher was not a booth-level officer, because the town school has four mathematics teachers and could spare one.

That last point is the part I have come to think is the actual scandal, and it took me years of classroom visits to see it, because it is invisible from any single classroom. Deputation is not distributed evenly. It falls hardest on exactly the schools with the fewest teachers, because the duty is assigned per school and per village, not per teacher. A school with one mathematics teacher and a school with four owe the district the same February. So the mechanism that produces our elections is also, quietly, a mechanism that transfers instructional time from the thinnest-staffed schools to nowhere at all. The gap between the village school and the town school widens every election year and no instrument we operate can see it happening.

Since this was published a teacher in a neighbouring district sent me her own roster, photographed, thirty-eight days, same shape: three spells of roll revision, a census fortnight, and four single days marked office. Two schools is not a study. It is enough to say that the plastic sleeve on the wall is not one school's misfortune.

The children see it. That is the thing about a fifth of a year: it is not spread thinly. It arrives in blocks. Class seven does trigonometry basics in the weeks the roll revision runs, and the substitute is the Hindi teacher, who is honest about it and gives them reading time. The chapter is not retaught. It appears in the March paper. The children who fail it are recorded as having failed it, and that record follows them, and at no point does any document connect the failure to the plastic sleeve on the staff room wall.

So my proposal is deliberately small, because the large one is unavailable. I am not asking for teachers to be exempted from election duty; I do not think that is winnable and I am not certain it is right. I am asking for the days to be counted. Publish, per school, per year, the number of instructional days lost to non-teaching deputation, in the same place we publish the results. An officer has since pointed out to me that the deputation orders are already numbered and filed at the block level, which means this is a reporting change and not a collection one, and which makes the case for it considerably stronger than I made it. Put it next to the pass percentage. Let a parent see that this school lost forty-one days and the one in town lost six.

Counting a thing does not fix it. It does, reliably, make it arguable, and right now this is not even arguable, because officially it did not happen.

I asked her whether she minds. She said the census fieldwork was interesting and she liked the walking, and that what she minds is a specific girl in class eight who was good at mathematics in class six and is not any more, and about whom she can produce no evidence of any kind.