All dishes / 南アジア
Rice and native-grain upma layer (sada upma, pre-semolina) Period COrigin theory CVerified
Some fields (dish name, region and gates) are shown from the Japanese-edition data.
Upma is a semolina (rava) dish; the whole genre, so the thinking goes, was a modern food born of colonial-era semolina. That view overlooks the dish's older layer. An upma of dry-roasted broken rice and millet was in South Indian kitchens long before.
The three gates
- Ingredient-supply gate
- 米・在来穀は南アジア在来で恒常充足(外来セモリナを含まない)
- Cooking-technique gate
- 穀粒を乾煎りし湯で炊く炒り煮(テンパリング=タルカした穀の炒り煮)
- Venue gate
- 南インド家庭・ティファン(軽食)文化
Overview
Most people picture upma as the yellowish savory dish of wheat semolina tossed with mustard seeds and urad dal. But that 'semolina upma' is only one form of a much older roasted-and-simmered dish.
What concerns us here is that older layer—broken rice, coarsely milled rice, and millets, dry-roasted, then simmered in water after mustard seeds and urad dal were bloomed in oil (tempering). The mainstays are rice and native grains. These were long-established local foods, on South Indian tables as everyday fare from early on. Archaeobotanical study shows that rice cultivation in South India was already established by the Iron Age and Early Historic period.
The name itself does not point to wheat. It joins uppu (salt) and mavu (flour), and here mavu means grain flour in general—a plain name for a salted, roasted grain dish. Coastal Andhra still keeps a broken-rice upma, and Kerala and Andhra a flattened-rice (aval / atukulu) upma, preserving the older native-grain forms.
Making it calls only for the handwork of dry-roasting grain and simmering it in water with the aroma of the tempered oil. It was a dish made again and again at the family breakfast and at tiffin (the light meal), with whatever grain was on hand.
How we know
'Upma was originally a semolina (rava) dish, with no older layer to speak of—a modern food that appeared only once semolina spread in the colonial period.' This view runs deep, a natural inference from the fact that upma today is made almost entirely with wheat semolina.
But it confuses two separate questions. One is when 'the semolina form of upma' became possible to make. The other is how old 'the roasted grain dish called upma' itself is. The first is bound to the spread of imported wheat semolina; the second is far older.
Semolina became a mass product in South India only after roller milling of steel spread in the later nineteenth century (around the 1870s in Switzerland and Hungary), and its local uptake owed much to colonial trade and to the Second World War, when the British administration promoted semolina as a substitute grain amid a rice shortage. The semolina form, in other words, is the later style.
That the older layer was established well before the nineteenth century can be seen in a Tamil cookbook printed at Madras in January 1891, the '(Maharashtra, Karnataka, Andhra and Dravida) Hindu Pākaśāstra' by T. K. Ramachandra Rau (356 pages, more than 400 South Indian dishes), which includes a 'sada (plain) upma.' The work of a retired official, it is among the oldest surviving printed Tamil cookbooks, and went through a second edition (1900) and a coronation third edition (1912). Its author, year, and editions are separately confirmed in the local paper Chennai First. There is also an English work of the same name printed at Madras in 1836, 'Pakasastra... modern culinary receipts of the hindoos' by C. V. Ramasawmy, but that is a book of a different line.
The exact date of origin can be fixed, from the sources, only as 'before 1891,' and beyond that it remains open. Even so, the fact that the name is not limited to wheat, that native-grain upmas survive across regions, and that mass semolina is a much later technology all point the same way: the roasted dish of rice and native grain came first, and the semolina form was laid over it as the present style.
The full set of origin theories and the detailed verification log — traceable all the way to their sources — are available on the Japanese edition.
See the full origin theories and verification log (Japanese) →
Related dishes
Lineage Family, predecessor, variants
Shares the key ingredient
- BiryaniSouth Asia (Deccan / Mughal)C
- DosaSouth India (Tamil Nadu / Karnataka)C
- PaellaValencia, SpainC
- Risotto alla milaneseMilan (Lombardy, Italy)C
- Nasi gorengIndonesia (Java)C
- JambalayaNew Orleans (Louisiana, USA)C
- MujaddaraLevant (Syria, Lebanon, Palestine)B
- DoriaYokohama, JapanB
- IdliSouth IndiaC
- MaqlubaPalestine / LevantC
- White mixed rice (hondonban / goldongban — Bibimbap precursor)Korean PeninsulaC
- MandiHadhramaut (Yemen)C
- KiribathSri LankaD
- KabsaSaudi ArabiaC
- PlovUzbekistan (Samarkand / Fergana)C
- MachboosPersian Gulf (Bahrain / Kuwait)B
- Gallo pintoCosta Rica / NicaraguaC
- Dal bhatNepalB
- PilauEast Africa (Swahili Coast)C
- Bánh cuốnVietnamB
- MahshiEgyptB
- PolowIranC
- PulaoNorth IndiaC
- PilavAnatoliaC
- Kabuli palawAfghanistanC
- Pilaf (French)FranceC
- Pilaf (Japanese-style)JapanC
- TahdigIranC
- Arroz de mariscoAlgarve (Portugal)C
- CasadoCosta RicaC
- Nan gyi thokeMyanmarC
- Mont diMyanmarC
- AranciniSicily (Italy)C
- Mazbi (madhbi)Yemen (Hadhramaut)C