Ancient poetry reveals India’s Deccan grasslands are not wastelands but timeless savannas

Researchers who analysed 700 years of Marathi literature argue that India’s plains are not human-made ruins.

In the film Dil Chahta Hai, the three lead actors agree to a pact overlooking the sea at Chapora Fort in Goa, with the dry scrub grassland landscape on the other side. The grassland is also depicted in the film’s closing scenes.

For a landscape ecologist, grasslands are striking even in a movie. The arid, rolling plains of the Deccan Plateau, which extend across the states of Maharashtra, Karnataka, Goa, Andhra Pradesh, and Telangana, are engaged in a quiet but intense scientific and cultural battle over the identity of the land itself. For decades, these vast, open spaces, which are dotted with thorny trees and thick grasses, have been officially maligned as “wastelands” or “degraded forests,” ecosystems that were supposedly destroyed by human activity like farming and colonial-era timber felling.

Following this flawed logic, the popular, modern solution has been to “fix” them by planting forests where, historically, none may have existed. However, a new study has unearthed a powerful line of evidence that overturns this misconception, as the truth has been hiding in centuries of poetry and folk songs.

Falling back on historical write-ups

The research was initiated during a conversation over coffee between two researchers, Ashish N Nerlekar and Digvijay Patil. Ashish Nerlekar is primarily an ecologist, specifically a plant scientist at Michigan State University, while Digvijay Paril is a PhD student in archaeology at the Indian Institute of Science, Education and Research (IISER) at Pune.

Patil’s archaeological work involves studying Marathi and Sanskrit texts for references to sacred sites. When Patil mentioned that he kept noticing peculiar plant names in old literary works, Nerlekar didn’t take long to recognise that these names referred to modern-day savanna species. Soon, both realised they had stumbled upon an untapped historical archive.

They meticulously reviewed 28 excerpts from Marathi literature spanning seven centuries from the 13th to the 20th, including biographies of saints, narrative poems, and folk songs called ovīs. Their goal was to identify georeferenced natural descriptions, thereby placing a scientific spotlight on locations mentioned by medieval bards and poets.

They ended up undertaking an ecological detective work using a literary time machine of sorts to dive into the past. They integrated rigorous ecological data with ancient Marathi literature, demonstrating that these tropical savannas are not ruined or degraded forests but rather ancient, natural ecosystems that have persisted for at least 700–800 years.

What they found was a consistently vivid picture of an open-canopy landscape dominated by grasses and specific thorny trees, a description that stands in stark contrast to the dense, closed-canopy forests people often romanticise as the past. As Nerlekar put it, “It’s fascinating that something hundreds of years old could so closely match what is around today and contrast so much with what people romanticise the past landscape to be.”

One of the earliest and most profound findings comes from a 13th-century biography of the religious figure Cakradhara, in which he identifies the hivara tree (Vachellia leucophloea). This iconic thorny acacia tree, recognisable by its pale yellow bark, is used as a powerful symbol for the cycle of death and reincarnation.

Critically, the hivara tree is a typical savanna species, a sun-loving species that thrives in open, grassy plains and would not grow under a closed forest canopy. The fact that it was prominent enough to be a cultural landmark in the 13th century strongly suggests that the landscape was already an open savanna at that time.

The researchers categorised 44 wild plant species mentioned in the literature and found that the majority (27 species) were savanna indicators, with only a small fraction being forest indicators.

They have tried deciphering the medieval Marathi and Sanskrit terms like vana or jangala, which are often translated simply as ‘forest’ or ‘jungle’ today, actually referred more broadly to wild, unsettled lands and places that included scrublands, grasslands, and savannas. The cowherders’ arrival in search of grass and the presence of species such as the sun-loving taraṭī (caper bush), mentioned in a 15th-century account, paint a consistent picture of an open, grass-filled biome.

Matching with evidence

The researchers used this literary data to triangulate with other lines of scientific evidence, notably the fossil record, pollen data, and genetics. They note that the remains of grazers such as grass-feeding hippos and antelope species such as the Blackbuck, which survive only on open plains, date back tens of thousands of years. And, fossilised pollen indicates that grasses, a reliable signature of savannas, have been continuously present in the region, despite natural shifts in tree communities over thousands of years.

Molecular phylogenies of endemic plants and lizards in the region confirm a rapid diversification event that coincided with global aridification millions of years ago, thereby establishing the antiquity of the savanna biome. Thus, every line of inquiry from the cultural memory stored in poems to the genetic code of its wildlife rejects the idea of a recent, human-caused ruin. It confirms that the savannas are not degraded forests, but a resilient, ancient ecosystem.

This finding carries enormous weight for conservation. When a savanna is mistaken for a degraded forest, the typical policy response is to plant trees, often using non-native species. This practice destroys the natural open habitat, threatens endangered species such as the Indian Wolf, and disrupts the ecosystem’s carbon storage capacity.

Savannas are excellent carbon sinks because their carbon is stored securely underground in their extensive root systems, rather than above ground in fragile trees. They also provide crucial fodder, indirectly supporting an estimated 20% of the global human population.

The good news is that human activities have not been as damaging as previously thought, resulting in the classification of wastelands or degraded lands. Therefore, our official classification system needs to correct the colonial outlook that sees these natural ecosystems as wastelands or degraded lands.

(The writer is with Research Matters)