Before it became a fixed written work, stories about a major dynastic conflict among the Bharatas likely circulated orally among bards and ritual specialists. This long oral phase helps explain why later written versions preserve multiple, sometimes conflicting, story layers.
The grammarian Pāṇini (often dated to around the 4th century BCE) includes rules that mention terms like “mahābhārata” and also names associated with the epic tradition. These references suggest that recognizable epic material (or at least well-known titles and characters) existed by this time, even if the full text was not yet fixed.
Another early Sanskrit source, the Āśvalāyana Gṛhyasūtra, is often cited as containing references to “bhārata.” Together with Pāṇini, this points to an epic tradition already known well enough to be referenced outside the epic itself.
Over the following centuries, the tradition grew from a core war story into a much larger work organized into multiple books (parvans). During this growth, genealogies, moral stories, and teachings were woven into the narrative, turning the epic into a wide-ranging cultural and ethical reference.
As redactors (editors working within a living tradition) expanded the poem, they inserted extended reflections on dharma—duty, right conduct, and social order. This shift made the Mahābhārata more than a war narrative: it became a framework for debating moral choices in family, politics, and religion.
An appendix-like text, the Harivaṃśa, developed with stories connected to Krishna and dynastic history. Its relationship to the Mahābhārata shows how the epic tradition remained open-ended, with new narrative blocks being added and stylistically adapted to fit the larger work.
By the early centuries CE, literary culture included authors who could allude to or quote material attributed to the Mahābhārata. Scholarship often notes that Aśvaghoṣa (commonly dated around the 1st–2nd century CE) is used as evidence that parts of the tradition, including material now found in the Harivaṃśa, were already being treated as “Mahābhārata” material.
As Sanskrit texts were increasingly written and recopied, scribes produced many local versions with small and large differences. This copying environment is crucial for the Mahābhārata: it created “recensions” (regional textual families) and set the stage for later scholars to compare manuscripts and reconstruct earlier forms.
By late antiquity, the Mahābhārata is commonly understood as a major, structured work divided into 18 books (parvans). While lines and episodes still varied across regions, the broad architecture and the epic’s role as a cultural “reference library” of stories and teachings became more stable.
Although the epic’s composition and redaction are usually placed centuries earlier, the oldest surviving physical manuscripts are much later due to the fragility of palm-leaf and birch-bark materials. This gap means scholars must infer earlier stages by comparing later manuscript families rather than relying on a single ancient “original.”
In 1919, the Bhandarkar Oriental Research Institute (Pune) began a long-term project to produce a Critical Edition by systematically comparing many manuscripts. The goal was not to print one regional version, but to reconstruct a carefully reasoned text with a record of variant readings.
The Critical Edition began appearing in published volumes, starting with the Ādiparvan edited by V. S. Sukthankar (published in 1933). Publishing in installments mattered because it made a transparent scholarly method available to other researchers, not just a finished “black box” text.
After decades of collation and editing, BORI announced completion of the Mahābhārata Critical Edition in 1966. This project compared a very large manuscript base and produced a reference text plus an apparatus (a structured record of variants), reshaping how modern scholarship discusses composition layers and regional traditions.
The Mahabharata (Indian epic: composition, redaction, and manuscript traditions, c. 400 BCE–400 CE)