Mahabharata Explorer

About this site

A free, illustrated reference to the Mahabharata - what happens in it, who it happens to, and where.

What this site is

Mahabharata Explorer is a reader's companion to the Sanskrit epic. It holds 974 retold episodes in chronological order, 273 character records, 146 places plotted on an interactive map, and 363 events arranged on a timeline, together with the eighteen parvas, a family tree of the Kuru dynasty, a glossary of Sanskrit terms, thematic essays and guided reading paths.

It is free and it always has been. There is no account to create, no paywall, no newsletter gate and no advertising. Every page is rendered to static HTML at build time, so it loads quickly, works without JavaScript for reading, and can be archived or quoted like an ordinary web page. Most stories carry an audio narration; none of them require it.

The site exists because the Mahabharata is enormous and badly served by the two usual options: a two-hundred-page abridgement that leaves out the arguments, or a ten-volume translation that most people never finish. Breaking it into episodes that can be read one at a time, and cross-linking every episode to the people and places in it, is an attempt at a third option.

How the content is made

The retellings follow the Sanskrit text as it is represented in the complete prose translation of Kisari Mohan Ganguli (1883-1896), with the book and chapter structure of the Bhandarkar Oriental Research Institute's Critical Edition used for sequencing and for deciding which episodes belong where. They are retellings, not translations: they compress, they choose a point of view, and they render dialogue in modern English. Where an episode exists in more than one version, the page says so.

Drafts are prepared with AI assistance and are then edited by hand. Every draft is checked against the text for the facts it asserts - who is present, what is said, what follows - and anything that cannot be found in the source is cut rather than smoothed over. Chronological placement is decided by hand, because the epic's narrative order and its story-world order are not the same thing, and cross-references to characters, places and other episodes are added and then verified against the site's own data rather than generated.

Illustrations are generated images, reviewed before publication and captioned as original artwork rather than as historical depiction. Audio narration is synthesised from the final edited text, so a narrated page and its written page always say the same thing.

This is stated plainly because it matters for how the site should be used. It is a careful secondary source and a way into the epic. It is not a translation, and nothing here should be cited as the words of the Mahabharata.

The site is edited by one person, referred to throughout as the editor.

Sources and further reading

  • Kisari Mohan Ganguli, The Mahabharata of Krishna-Dwaipayana Vyasa, published by Pratap Chandra Roy, 1883-1896. Complete, public domain, and the primary reference for the retellings here.
  • The Critical Edition of the Mahabharata, Bhandarkar Oriental Research Institute, Pune, 1919-1966, begun under V. S. Sukthankar. Used for the parva and upaparva structure.
  • Bibek Debroy, The Mahabharata, 10 volumes (Penguin, 2010-2014). A modern unabridged translation of the Critical Edition.
  • J. A. B. van Buitenen and James L. Fitzgerald, The Mahabharata (University of Chicago Press, 1973 onwards). Incomplete, and the most heavily annotated English version.
  • John Brockington, The Sanskrit Epics (Brill, 1998), and Alf Hiltebeitel, Rethinking the Mahabharata (2001), for questions of composition and layering.
  • Irawati Karve, Yuganta: The End of an Epoch (1969), for character interpretation.

The guides list their own sources at the end of each essay, and say explicitly where scholarship is divided.

How the places are located

Coordinates on the map follow the standard scholarly identifications where those exist: Hastinapura with the site of that name in Meerut district, Kurukshetra with the region around Thanesar in Haryana, Indraprastha with a site in Delhi, Dwaraka with modern Dwarka in Gujarat. These identifications rest on a mixture of continuous local tradition, place-name survival and excavation, and they vary in strength.

A great many of the epic's places have no agreed location at all - forests, hermitages, mountain passes and sacred fords in particular - and several have two or three competing candidates. Where a location is uncertain or disputed, its page says so rather than presenting a pin as settled fact. Mythic geography, such as the realms of the nagas or Indra's city, is marked as mythic and placed only for the sake of the map.

Open the mapAll placesWhat archaeology can and cannot show

Corrections

Corrections are welcome and they get made. If a detail here contradicts the text - a name, a relationship, a sequence, a parva number, a coordinate - please report it, with the passage you are relying on if you have it to hand. Interpretation is arguable and the site tries to flag where it is interpreting; factual errors are simply errors.

Illustrations may be reused under the terms set out on the image licensing page: free for personal and educational use with attribution, permission required for commercial use.

Where to start