Reconstruction
The Durga Devi famine is remembered as one of the medieval Deccan’s most destructive food crises. Later histories and regional traditions place its beginning around 1396 and commonly assign it a duration of eleven or twelve years. That chronology should not be read as a modern annual rainfall record: no contemporary series demonstrates twelve completely rainless monsoons, and surviving descriptions were written or collected long afterward. Nevertheless, independent regional gazetteer traditions consistently associate the period with severe scarcity, abandoned settlements and prolonged agrarian disruption between the Godavari and Krishna rivers.
The crisis struck a plateau where harvests depended heavily on variable monsoon rainfall. Sorghum, pearl millet and other millets were important dryland grains, while pulses such as horse gram and mung bean supplemented cereal diets; rice was cultivated or imported where water and purchasing power permitted. These foods are a reasonable reconstruction of the threatened food supply, not a documented famine ration. Repeated rainfall failure would have reduced grain, fodder and drinking water together. Seed stocks could be eaten, cattle and pack animals lost, and field labor interrupted, making recovery difficult even after rain returned.
Firishta, a Deccan historian writing approximately two centuries later, reports that the Bahmani ruler Mahmud Shah employed 10,000 bullocks to bring grain from Malwa and Gujarat and sell it cheaply. He also credits the ruler with establishing institutions for orphans. The numbers cannot be independently verified, but the account identifies a credible relief mechanism: pack-bullock transport linking the drought-stricken interior to western and northern grain markets. Without railways or modern roads, moving bulky grain hundreds of kilometres was slow, expensive and vulnerable to insecurity.
For ordinary households, famine adaptation probably meant reducing portions, replacing preferred grains with cheaper millets or gathered foods, thinning grain into gruel and migrating toward better-watered districts or towns. Such responses are reconstructions based on other South Asian famines; the sources do not preserve household menus from 1396–1407. Landless laborers, rural craftspeople and small cultivators would have been especially exposed because crop failure removed both food and employment. Later accounts describe depopulation, weakened revenue and villages remaining unsettled for decades, although their dramatic language resists precise demographic interpretation.
Recovery required more than the return of rain. Gazetteer accounts describe later efforts to recall cultivators, combine depopulated village territories and offer land on exceptionally light initial rents. These measures suggest losses of labor, livestock, seed and administrative control. The famine also endured in Vanjari community tradition, where Durga is remembered as a wealthy grain carrier and benefactor; that story is culturally significant but cannot be treated as a contemporary biography.
The event’s defensible modern relevance lies not in a single surviving dish but in the recurring food-security problems it exposes: dependence on uncertain rainfall, unequal access to grain, the importance of transport and storage, and the value—without invulnerability—of drought-adapted millets and pulses. Durga Devi remains both a historical famine tradition and a warning about how environmental stress becomes social catastrophe.
Historical context
Much of the northern and central Deccan belonged to the Bahmani Sultanate, whose capital was at Gulbarga, while the expanding Vijayanagara Empire controlled substantial territory farther south. Both states depended on agrarian revenue and contested strategic routes and frontier districts. Interior parts of the plateau were semi-arid, with rainfall concentrated in the monsoon and varying sharply between years and locations. Grain could move by pack bullock from Gujarat and Malwa over distances of several hundred kilometres, but transport remained costly and slow. Beyond the Deccan, Timur sacked Delhi in 1398, and the early Ming dynasty governed China; these contemporaneous events provide chronology but no demonstrated causal connection to the famine.
Evidence
Written sources
ModerateFirishta and later regional histories report severe famine, grain transport and relief under Mahmud Shah. Firishta wrote about two centuries after the event, while the gazetteers relied on earlier histories and regional traditions rather than contemporary administrative records.
Archaeological
LimitedRegional archaeology documents medieval settlement, agriculture, water management and transport landscapes, but no securely identified archaeological destruction or abandonment horizon has been directly attributed to this famine.
Dating
ProbableA start around 1396 is widely repeated, but sources variously end the famine in 1407 or 1408 and may compress a succession of poor harvests into a conventional twelve-year narrative.
Geographic attribution
ModerateThe tradition is consistently associated with Maharashtra and the wider Deccan, especially territory between the Godavari and Krishna. Claims that every region south of the Narmada was equally affected are not quantifiable.
Food identification
InterpretiveMillets, sorghum, rice and pulses are supported as crops and foods in the broader medieval Deccan and neighboring Vijayanagara world, but no surviving source provides a specific Durga Devi famine food list.
Preparation method
InterpretiveThe use of thin gruels, cheaper grains and gathered foods is a reasonable famine reconstruction based on comparative evidence, not a directly observed menu from 1396–1407.
Historical interpretation
LimitedDepopulation, disrupted cultivation and delayed resettlement are recurrent in later accounts, but mortality totals, annual rainfall conditions and the precise duration cannot be established.
Sources
- 1.Pankaj Tandon and Phillip B. Wagoner (2018). The Bahmani ‘Currency Reform’ of the Early Fifteenth Century in Light of the Akola Hoard. American Journal of Numismatics 29. hdl.handle.net/2144/35588Scientific literature
- 2.Muhammad Qasim Hindu Shah Astarabadi Firishta; translated by John Briggs (1829). History of the Rise of the Mahomedan Power in India, till the Year A.D. 1612, Volume II. Longman, Rees, Orme, Brown, and Green. www.cambridge.org/core/books/history-of-the-riseHistorical primary source
- 3.Bombay Presidency (1884). Gazetteer of the Bombay Presidency: Ahmadnagar, Volume XVII. Government Central Press, Bombay. gazetteers.maharashtra.gov.in/cultural.maharashtModern synthesis
- 4.Bombay Presidency (1885). Gazetteer of the Bombay Presidency: Satara, Volume XIX. Government Central Press, Bombay. gazetteers.maharashtra.gov.in/cultural.maharashtModern synthesis
- 5.Kathleen D. Morrison and Carla M. Sinopoli (1992). Economic Diversity and Integration in a Pre-Colonial Indian Empire. World Archaeology 23(3). doi:10.1080/00438243.1992.9980184Scientific literature
- 6.Kathleen D. Morrison (2000). Naturalizing Disaster: From Drought to Famine in South India. Environmental Disaster and the Archaeology of Human Response, Maxwell Museum of Anthropology, Anthropological Papers No. 7.Scientific literature
Limitations
- moderatedescription.evidence classification
Calling the evidence simply 'historical documentary' may imply contemporary documentation. The standard narrative depends substantially on Firishta, who wrote about two centuries later, colonial-era historical compilations and regional tradition.
- moderatedescription.duration
The claim of twelve continuously rainless years is not independently demonstrated by annual contemporary observations or securely matched palaeoclimate evidence. It may preserve a prolonged sequence of scarcity years rather than literal total rainfall failure every year.
- minordate end
The terminal year is not stable across sources: 1407 and 1408 both occur, partly because a famine beginning in 1396 is conventionally described as lasting twelve years. The entire interval should be treated as approximate.
