Appearance
1.4 — How a Story Grows
Take one episode and watch it change.
Version one, the oldest layer. Hanuman crosses the sea to Lanka. The text says he leaps from a mountain, that the force of his jump presses the mountain down, and that he flies over the water. It takes him most of a day. He is a formidable individual doing something extreme.
Version two, a few centuries later. He grows to an enormous size before jumping. Sea creatures rise to test him. A mountain offers him rest.
Version three, the devotional retellings from around 1500 CE. He is a form of Shiva. His leap is an act of pure devotion, and the distance is irrelevant because devotion has no distance.
Version four, the version most people carry today. He can fly whenever he wants, is immortal, and is present wherever the Ramayana is recited.
Nothing was ever removed. Every stage kept everything before it and added. That is the single most important fact about how these stories reached us, and once you can see it working, you can read backwards through it.
The six mechanisms
Stories do not inflate randomly. They inflate through a small number of specific processes, and all six are visible in this material.
1. The reciter's incentive. These texts were performed. A reciter in a village square or a royal court competed for attention with other reciters. A version in which the hero leaps further gets a better response than one in which he leaps far. Nobody is lying; the pressure is continuous, gentle, and always in one direction. Over forty generations of performance, one-directional pressure produces enormous drift.
2. Explanation of the unexplained. Every generation of listeners asks questions, and eventually somebody answers them inside the text. Why did Rama not simply refuse the exile? An answer gets supplied. How could Bhishma be unbeatable? A boon gets supplied. Why did Karna lose? A curse gets supplied. Curses and boons in these epics are very frequently retroactive explanations for something the older story simply reported. Once you know this, you can often reconstruct the older story by deleting the explanation and seeing whether the event still stands. It usually does.
3. Theological upgrading. As Rama and Krishna became objects of worship, passages were added identifying them as divine, and the passages cluster in the books that language-dating already flags as late. A hero becomes a god slowly, and the seams show. In the Ramayana, the seam is between the older books and books one and seven. In the Mahabharata, Krishna is a shrewd political operator through most of the war narrative, and a cosmic deity in specific inserted passages.
4. Moral cleaning. Later hands are uncomfortable with what the heroes do, and they soften it. Rama's killing of Vali from concealment gets a longer and longer justification. The dice game gets material added explaining that Yudhishthira was compelled by fate rather than by his own weakness. The awkward version is the older version, which is why 1.3's conclusion matters so much: the difficult reading is not modern cynicism, it is closer to the original.
5. Absorption of local traditions. As the epics spread, regional stories attached themselves. A local hill acquires a Ramayana episode. A local deity becomes a character. This is why there are so many Ramayanas — hundreds of versions across India and across Southeast Asia, with different events, different emphases, and in some versions different outcomes. The Jain Ramayanas make Ravana a tragic and admirable figure and have Lakshmana kill him rather than Rama. The Thai version has its own cast. None of them is the fake one. They are all what happens when a living story meets a living people.
6. Compression and merger. Multiple real people become one character. Multiple events become one event. A campaign of years becomes a battle of eighteen days. This is universal in oral tradition and it works in the opposite direction from the others — it makes the story tighter and more dramatic rather than bigger.
Reading backwards
The practical use of all this is that you can often run the process in reverse. Four tests, applied to any episode.
Strip the explanation and see if the event survives. Bhishma cannot be killed because of a boon, and he is killed anyway because of a technicality. Delete the boon: an old, immensely skilled commander is finally brought down by an opponent his code will not let him strike. The event survives perfectly well, and 3.13 argues that the second version is both older and far more interesting.
Ask whether the miracle is doing any work. In the Lanka bridge episode, the floating stones change nothing about the plot or the meaning. That is a strong sign of decoration. Where a supernatural element is genuinely load-bearing, it usually turns out to be a literary device that the text is quite open about.
Watch for the passage that argues. When a text spends fifty verses defending a character's action, that defence is almost always later than the action. Nobody defends what nobody has objected to. The length of the defence is a measure of how uncomfortable the episode had become.
And notice what a passage assumes. If a passage assumes temples, image worship, a developed caste system with rules that fit the medieval period, or vocabulary from a later philosophical school, it is from that period. The surrounding text may be a thousand years older.
The same thing happens to your own stories
This is not an exotic ancient process. It is running right now, on a shorter timescale, on everything you know.
A company's founding story. Ten years in, the garage is smaller, the odds were longer, the founder's insight is sharper and earlier than it actually was, and the six people who made it work have become one visionary.
A family story. Your grandfather's journey gets harder with each telling. Not because anyone lies, but because the version that gets a reaction is the version that gets repeated, and the version that gets repeated is the version that survives.
A national history. The same six mechanisms, with a state's resources behind them.
And your own memory of your own life, which Volume VI covers in detail: recall is reconstruction, and each act of remembering rewrites the memory slightly in the direction of the story you currently tell about yourself.
The Mahabharata is what happens when this ordinary process runs for two thousand five hundred years with the best storytellers of a civilisation working on it. It is not a strange object. It is a very old object.
What to carry from this page
Growth is one-directional, so the plain version is usually the old one. When you have two versions of anything and one is more dramatic, the less dramatic one is probably closer to the source. This is a reliable heuristic about ancient texts, family history, gossip and news.
A long defence marks an old embarrassment. Whenever a text, a person or an institution explains at length why something was fine, something was not fine.
And ask of every story you are told, including your own: what has this been optimised for? Attention, comfort, identity, or accuracy. Only one of those four is a truth-seeking pressure, and it is almost never the strongest one in the room.
Next: 1.5 takes on the question that decides how the whole of Parts 2 and 3 read — who the Devas, Asuras, Rakshasas, Vanaras, Nagas and Yakshas actually were, if they were not species.