The word purāṇa is not a title; it is a category, and the category has criteria. A Purāṇa is conventionally defined by five marks — pañca-lakṣaṇa — the five subjects a text must cover to qualify.
Debroy raises this in his introduction and then does something with it:
Take the Bhagavata Purana as an example. In Chapter 2(10), we are told about the ten characteristics of the Bhagavata Purana and these are sarga, visarga, sthana, poshana, uti, manvantara, isha-anukatha, nirodha, mukti and ashraya. ... Therefore, the text has much more than the five attributes cited.1
Ten, not five. The claim is not that the Bhāgavata satisfies the genre. It is that the genre's definition is inadequate to it.
The two counts do not have the same standing, and that is the first thing to get straight.
The five are external. They are the definition of the category purāṇa, supplied by the tradition of classification — a standard applied to texts from outside, by people sorting a library.
The ten are internal. They come from chapter 2(10), where Śuka lists them in the text's own voice.2 Nothing outside the book assigns them.
So this is not two classification schemes disagreeing. It is a text stating its own characteristics, and the count exceeding the count the category uses. The Bhāgavata is answering a question about genre membership that it was not asked, in its own words, with a larger number.
It would be neat if the ten contained the five and added five more. The relationship is looser than that.
The conventional five cover creation, secondary creation, genealogies of gods and patriarchs, the reigns of the Manus, and dynastic history. The Bhāgavata's ten are sarga and visarga — gross and subtle creation — plus preservation, sustenance, addiction to the senses, manvantaras, accounts of the lord, withdrawal, liberation, and the ultimate refuge.2
Creation appears on both, in two forms rather than one. Manvantara appears on both. Beyond that, the lists do different things. Dynastic history, which is a full fifth of the conventional definition, is not a category in the Bhāgavata's ten at all — though the volume contains a great deal of it, which would fall under accounts of the lord.
And the ten include four items the five have no room for: sustenance, addiction to the senses, liberation, and the refuge. Three of those four are soteriological, and none of them is the kind of thing a librarian's category would track.
Register: the conventional five are a genre definition summarised here from general knowledge; the ten are quoted from the text. The comparison above is not made by Debroy or by the passage, and this page does not claim it is.
Take the move seriously as a move and ask what changes when a text lists its own characteristics.
It converts membership into inadequacy. A text that merely satisfied five criteria would be a Purāṇa among Purāṇas. A text that lists ten has not failed the category; it has made the category too small to describe it, which is a claim about rank made without any claim about rank.
It is unanswerable in kind. The five are a definition, and definitions are contestable by argument. Ten self-declared characteristics are not a competing definition — they are a description, and there is nothing to argue with. Nobody can say the Bhāgavata does not have ten characteristics, because it is telling you what it has.
And it costs nothing. The text does not attack the five, deny them, or say other Purāṇas are lesser. It simply enumerates itself at a different length and moves on to the next sentence.
That is the same shape as the volume's other boundary operations — the worship catalogue that concedes rival efficacy, the avatāra roster that absorbs rival founders. Concede everything, contest nothing, and make the operative claim by enumeration rather than by argument.
There is a feature of the ten that the number obscures, and it is the more interesting claim.
The sentence immediately after the list says the first nine exist to establish the tenth.2 So the ten are not ten characteristics of equal standing. They are nine plus one.
Which means the comparison with the five is not really ten-against-five. It is nine instruments and a subject against five subjects. The Bhāgavata is not claiming to cover more ground; it is claiming to have a different structure — a body of material with a single thing at its centre, rather than a list of things a text should contain.
That is a much stronger claim than a larger number, and it is invisible if you quote the ten and stop, which is what the count's citation history has mostly done.3
Debroy raises a second definitional point in the same passage, and it belongs here:
But the text is not called Bhagavata Purana; it is in fact called Bhaagavata Purana. What does the word Bhaagavata mean? Taken as an adjective, the word means sacred ... However, taken as a noun, Bhaagavata means a devotee or worshipper of Vishnu — and bhaagavata dharma is the dharma of devotion towards Vishnu. Therefore, Bhaagavata Purana is the text that describes this bhaagavata dharma that Vishnu devotees follow.1
Two readings of the title, and they classify the book differently. Sacred Purāṇa is a description of status. The Purāṇa of the devotees' dharma is a description of subject matter and audience.
He supplies both and adopts the second, with a stated reason. That is the apparatus doing what it does well — showing the fork rather than translating past it — and it is worth noting given how often in this volume the same hand identifies without hedging.
One further thing, and it is uncomfortable for the claim.
The ten are stated once, in one chapter, and then never used. No later passage says this is topic four, or we now turn to withdrawal, or refers back to the scheme in any way. The volume does not proceed through them in order and does not label its material by them.
So the ten characteristics function as a statement about the book rather than as a structure of it. They are a claim made in passing, in the middle of the Second Skandha, and abandoned.
Which places them alongside the volume's other enumerations that are stated once and never operated — the five characteristics of intelligence, the eight types of gods listed as nine. It is a text that reaches for counts readily and follows them rarely.
Your work is being assessed against criteria that do not fit it — a role's competency framework, a genre's conventions, a funder's categories, a review rubric written for something adjacent.
You have two moves and they are not equally available.
Qualify. Argue that you satisfy the criteria in a modified sense, or that some do not apply. This keeps you inside the framework and every argument is contestable by whoever owns it.
Enumerate. Describe what the work actually has, at whatever length that takes, without mentioning the criteria at all. This is what the Bhāgavata does, and its strength is that a description is not a competing standard and therefore has nothing to attack.
The second only works under two conditions. The enumeration has to be genuinely of the work — not the criteria renamed, or it reads as evasion. And it has to be longer or structurally different, because if it comes out the same size and shape as the standard, you have conceded the frame while appearing to resist it.
And know the cost: enumeration forfeits the standard's legitimating power. Satisfying five marks makes you a Purāṇa in a way everyone recognises. Listing ten characteristics makes you something nobody has a name for, and the tradition has to do extra work thereafter to place you — which, in this case, it did, and it did so by citing the ten and dropping the sentence that made them interesting.
The strongest evidence that the larger count is a positional claim rather than a taxonomic one is that it is stated once and never operated. A working classification gets used. This one appears in a single chapter, is not referred to again, and does not govern the order of anything.
Tension — the ten are cited as ten equal characteristics, and the text says nine of them exist to establish the tenth. The asymmetry is in the sentence immediately following the list and is routinely dropped in citation. Preserved.
Open question. Dynastic history is a fifth of the conventional definition and is not a category in the Bhāgavata's ten, though the volume is full of it. Is it absorbed under accounts of the lord, or is its absence from the list a claim about what this Purāṇa considers itself to be doing?
This vault's material on Purāṇic genre treats the pañca-lakṣaṇa as a descriptive convention that few surviving Purāṇas actually satisfy — a definition honoured more in classification than in content, with the extant texts sprawling well past it.
If that is right, then the Bhāgavata's ten are less exceptional than they are usually made to sound. Most Purāṇas exceed five marks; this one says so about itself. What distinguishes it is the declaration, not the excess.
The Fifth Veda and the People It Was Built For documents the same volume's other claim about its own status — a corpus built to route around a locked one. Both are self-descriptions, both are made in passing, and both position the text relative to a category rather than arguing about the category. Whatever else this book does, it is unusually willing to talk about what kind of thing it is.
With Word Weapons: Purr, Slur, and the Three Talks. That page works the way category labels do evaluative work while presenting as descriptive — a word that classifies also ranks, and the ranking travels without ever being asserted.
Purāṇa is a purr-word of exactly that kind: a classification that confers standing. The Bhāgavata's ten characteristics decline the classification's terms without declining the standing, which is a manoeuvre the purr/slur analysis has a place for and does not usually see performed this cleanly.
What the pairing produces: the word-weapons framing is built around the use of loaded categories — who applies them, to whom, and to what effect — and its defensive advice is to expose the evaluation hiding inside the description. This case shows a third option that the frame does not reach: instead of contesting the label or accepting it, the Bhāgavata out-describes it. It does not deny being a Purāṇa and does not argue about what a Purāṇa is; it simply publishes a longer specification of itself. Which suggests the vulnerability of a purr-word is not that its evaluation can be exposed — everyone already knows — but that its descriptive content is thin, and thin descriptions lose to detailed ones without any argument taking place. That is available to anyone being classified, and it does not require the classifier's cooperation.
Sharpest implication. The famous claim is not that this Purāṇa has ten characteristics rather than five. It is that it has nine and a subject — and the tradition has spent centuries citing the number while dropping the structure, which means the text's most distinctive statement about itself has been transmitted as a count.
Generative questions.