My uncle's field flooded in June, which is what it is for. It flooded again in the second week of August, which is not, and the second flood took the crop. He replanted. The replanting failed too, because the water came back before the seedlings had set, and by September he had spent two rounds of inputs on a field that produced nothing at all.
He is planting the same variety again this year. So is almost everyone in the panchayat. I want to explain why, because the explanation is the actual story, and it is not the one about farmers being resistant to change.
The variety he plants is high-yielding, it responds to fertiliser, and it was bred for a monsoon that arrived in June and tapered through September. That monsoon is gone. It has been visibly going for the twenty years I have been paying attention: fewer rain days, more rain per day, the tail end moving later and hitting harder. In coastal Kerala this is not a projection anybody has to be persuaded of. It is a thing you can see from a doorway.
There are varieties bred for exactly this. Submergence-tolerant lines that can sit under water for two weeks and come back — the Swarna-Sub1 family and its relatives — have existed for well over a decade and do, demonstrably, work. My uncle knows they exist. He has seen a demonstration plot. He is still planting the old one.
Here is why, and it took me an embarrassingly long time to understand because I kept looking for the answer in agronomy when it was in procurement.
The mill buys on a grain standard set by the state procurement operation. That standard is written around the old variety's grain length and cooking behaviour. Bring the tolerant variety to the mill and it is not rejected, exactly; it is graded down. The difference is a few rupees a quintal, which sounds trivial written out and is not trivial against a margin this thin. Against a two-in-five chance of losing the crop entirely, the tolerant variety is obviously the better bet. Against the certainty of a lower price every single year including the good ones, it is not obvious at all. My uncle is not gambling badly. He is being paid, in a small and reliable amount, to gamble.
The second reason is credit. The input dealer who extends him seed and fertiliser on account until harvest stocks what he can move, and what he can move is what the mill pays best for. Ask for the other seed in May and you can get it, if you ask early, if you pay cash, and if you are prepared to be the only one. Farming is not an individual activity at that scale — labour is shared, transplanting is synchronised across adjacent fields, water moves between them — and being the only person on a different calendar is a real cost that no extension pamphlet has ever costed.
So the picture is: a technically solved problem, a farmer who knows the solution, and a price signal pointing firmly the other way. The seed is not the bottleneck. The bottleneck is that we are still paying for a grain standard written for a climate we no longer have.
I want to give the counterargument its due, because it is not stupid. Procurement standards exist to protect buyers from adulteration and to keep quality legible across thousands of transactions; loosening them casually would create a different and possibly worse mess, and the people who administer them are not fools. Changing a grain standard is genuinely hard, involves several ministries, and would be exploited by somebody within a season. I accept all of that. It is still the thing that has to change, and the fact that it is administratively unpleasant is not a reason to keep discussing seed adoption as though the problem were farmer psychology.
What annoys me most is the framing I hear at every workshop: that adoption is slow because of awareness, and what is needed is more demonstration plots. There have been demonstration plots in this district for eleven years. Everybody is aware. Awareness was never the constraint, and continuing to fund awareness campaigns is a way of doing something visible about a problem whose actual solution sits in an office in Thiruvananthapuram and involves rewriting a schedule that nobody gets promoted for rewriting.
The August flood is going to happen again. It is the most predictable thing about the coming decade. The seed to survive it is sitting in a research station forty kilometres from my uncle's field, and the reason it stays there is a few rupees a quintal, which is a policy decision, which means somebody could decide differently.
He asked me last month whether the standard was likely to change. I said I did not know. He said the seed dealer had asked him the same question, which I found more interesting than anything in the literature: the dealer is waiting on the mill, the mill is waiting on the standard, the farmer is waiting on the dealer, and every one of them can describe the loop accurately. Nobody in the chain is confused. They are all just correctly waiting for somebody else.
Update. The procurement schedule was amended in a subsequent revision cycle and the tolerant varieties are now graded on their own line rather than against the old grain length. The dealer in my uncle's block still does not stock the seed, because the mill has not yet changed what it tells him it wants, which is the next link in the loop and is the one I did not write about.
