The fat that replaced ghee, and then lost the argument

Look at a tray of mithai and pick the oldest thing on it. The jalebi looks like the answer: an orange coil, syrup-heavy, sold by weight, the sort of sweet that seems to have come out of the ground. It has one ingredient younger than the shopkeeper's grandfather. The fat it is fried in is not ghee and usually never was. It is vanaspati — hydrogenated vegetable fat, a factory product that entered the Indian kitchen in the 1930s, took the sweet shop in about one generation, and has spent the last decade being legislated out of it.

This is the one place in the pillar where the argument runs backwards. Elsewhere the plate records what a place could not get: no salt in the eastern hills, no fresh fish in Manipur, no wine vinegar in Goa. Here the plate records something a place was given — cheaply, deliberately, with an advertising budget — and then had to be argued out of.

A factory that could undercut a cow

Before the 1930s, hydrogenated fat in India was imported. An importer named Hussein Dada sold it under the name Dada Vanaspati, and the volume was small, because the product was a novelty and ghee was the real thing. In 1931 Lever incorporated Hindustan Vanaspati Manufacturing Company to make the stuff locally. In 1937 a plant at Sewri, in Bombay, began turning out a brand called Dalda — Dada's name with the L of Lever pushed into the middle of it, a deal the importer agreed to and presumably regretted.

The pitch was substitution, and it was honest about being substitution. Desi ghee comes from milk, and milk in India is seasonal: there is a flush season when buffalo and cow give heavily and a lean season when they do not, and the price of ghee swings with it. Ghee also goes rancid, sets hard in winter, and costs enough that a household used it by the spoon, not by the pan. Vanaspati did none of that. It was the same every month of the year, held its shape at room temperature, did not smell of anything until it was hot, and cost a fraction.

For a sweet shop, that last set of properties is worth more than flavour. A halwai making ladoo, imarti and jalebi needs a fat that behaves identically on a Tuesday in August and a Friday in December, because his customers expect the same product. Ghee is a variable. Vanaspati is a constant. The same logic sold it to bakeries, where shortening has to laminate into layers for puff pastry and khari biscuits, and to restaurants frying at volume.

What it actually changed

FSSAI's own regulatory documents list the uses without embarrassment: vanaspati in mithai — ladoo, imarti, jalebi — and in deep-fried snacks such as aloo tikki; bakery shortening in cakes, pastries and puffs; margarine in the same family. That is a precise map of what became cheap in the twentieth century. The festival sweet became a daily purchase. The puff, which is a French technique rebuilt on laminated fat, became a bakery item sold by the piece at a railway stall.

None of this was a trick played on a passive public. Vanaspati was sold as modern, hygienic and convenient, in sealed tins, in a country where loose ghee was routinely adulterated and where "pure" was a claim you could not verify. A sealed tin from a factory in Bombay was, for a generation of housewives, the safer purchase. The product won on trust before it won on price.

The cost sits in the molecule

Hydrogenation pumps hydrogen into liquid oil to make it solid, and the process does not fully control which way the remaining double bonds point. Some of them end up in the trans configuration, which is a shape the human body has no enzyme for. The clinical result is not subtle: trans fat raises LDL, lowers HDL, and does both at a lower dose than saturated fat does either.

FSSAI's own numbers put the Indian cost at roughly 71,000 deaths a year — about 5% of an estimated 1.5 million annual coronary deaths — attributed to trans fat intake. The figure is an estimate from 2017 dietary and mortality data rather than a count, and it is worth treating as an order of magnitude rather than a ledger. The order of magnitude is why the regulation exists.

Four steps down, in nine years

The legal history is unusually legible, because FSSAI published each step. In 2013 the limit on trans fat in margarine, bakery shortening and vanaspati was 10% by weight. From 27 August 2016 it was 5%. From 1 January 2021 it was 3%, and from 1 January 2022 it was 2%. In February 2021 a further amendment extended the 2% ceiling beyond fats and oils to any food in which oils or fats are used as an ingredient, measured against the fat actually present in the product.

What each step actually meant inside a factory is worth pausing on, because "reformulate" sounds easier than it is. A vanaspati plant buys a blend, hydrogenates it to a target melting point, and sells the result as a semi-solid that behaves on a hot pan. Lower the trans fat and the melting point moves; the fat goes soft at room temperature; the laminated layers of a puff collapse. The industry's answer was interesterification — rearranging the fatty acids on the glycerol backbone without creating trans bonds — and blending with palm fractions. The product survived. The puff survived. What changed is that the fat is now assembled rather than converted.

The part the label cannot reach

Here is the gap, and it is not a small one. FSSAI's 2018 note says it plainly: trans-fatty acids are also formed during repeated heating of fats and oils while deep frying, at home and in restaurants. That is a different mechanism from hydrogenation — it happens in the pan, to any oil, with heat and time and oxygen doing the work. It produces no ingredient list. A tin of vanaspati at 2% and a fryer that has been topping up the same oil since morning are governed by two different systems, and only one of them has a number on it.

This is why the regulation reads, from a street level, as a rule about factories at the same time as it is being sold as a rule about food. The sweet shop that buys a compliant shortening has complied. The shop that fries in oil it will not change until the colour turns has not broken the rule, because there is no rule for it to break — only a recommendation, and a recommendation is not a test.

What the brand became

Dalda became a generic. For decades "vanaspati ghee" was the category name and Dalda was what people said, the way a photocopy is a Xerox. Then the market moved: refined sunflower, soybean and palm oil took the frying, ghee came back as an aspiration, and in 2003 Unilever sold the brand to Bunge for reportedly under ₹100 crore. It still exists. It is no longer the centre of anything.

The general lesson is worth carrying out of the pillar. Of the three variables that decide a regional kitchen — grain, fat, and what the place could not get — fat is the one that can be standardised, and therefore the one that regulation reaches first. Nobody can write a law about a spice blend, because a spice blend has no single molecule to cap. A fat does. That is why the most heavily legislated ingredient in the Indian kitchen is not the chilli, which arrived in the sixteenth century and rearranged everything, but the fat, which arrived in 1937 in a sealed tin and is now being taken back out of it one percentage point at a time.

This is entry 17 of Food in About India. The pillar's argument — grain and fat, not spice blends — is set out on the pillar page.