When Form Follows Death (SA 15-16 and SN 22.35-36)

While editing my translation of SA 15-16 today, I perused my copy of Bodhi’s translation as I often do. The closest Pali parallels are SN 22.35-36, but the SA version and the translations these Pali suttas are strangely incongruent.

The important parts of the two sutras:

SA 15-16 SN 22.36
「比丘!若隨使使者,即隨使死;若隨使死者,為取所縛。比丘!若不隨使使,則不隨使死;不隨使死者,則於取解脫。 “Yaṁ kho, bhikkhu, anuseti taṁ anumīyati; yaṁ anumīyati tena saṅkhaṁ gacchati. Yaṁ nānuseti na taṁ anumīyati; yaṁ nānumīyati na tena saṅkhaṁ gacchatī”ti.
Monk, if someone follows the tendencies, then they will follow death. If someone follows death, then they’re bound by grasping. Monk, if someone doesn’t follow the tendencies, then they won’t follow death. Not following death, they are freed from grasping. “Mendicant, you’re measured against what you have an underlying tendency for, and you’re defined by what you’re measured against. You’re not measured against what you have no underlying tendency for, and you’re not defined by what you’re not measured against.”

This uses the version in SA 15, which has a different conclusion than SA 16: bound/freed from grasping vs. increasing/not increasing in number. The two Pali suttas lack these conclusions. Instead, SN 22.35 is simpler, omitting the term anumīyati from the Buddha’s statement.

SN 22.35 SN 22.36
“Yaṁ kho, bhikkhu, anuseti, tena saṅkhaṁ gacchati; yaṁ nānuseti, na tena saṅkhaṁ gacchatī”ti. “Yaṁ kho, bhikkhu, anuseti taṁ anumīyati; yaṁ anumīyati tena saṅkhaṁ gacchati. Yaṁ nānuseti na taṁ anumīyati; yaṁ nānumīyati na tena saṅkhaṁ gacchatī”ti.

Now, this is where it gets interesting for me. When I checked the footnote on Bodhi’s reading of anumīyati in the later passage that applies this formula to the aggregates, he says:

Spk explains anumīyati as if it were equivalent to Skt anumṛyate, “to die along with”: “When the underlying tendency is dying, the form to which it tends dies along with it (anumarati!); for when the object is breaking up, the mental factors that it as object cannot persist.” This of course is ludicrous, for anumīyati is doubtlessly from anu + mā; CPD defines the verb as meaning “to be measured after,” which I follow here.

And, yet, here I sit, looking at SA 15-16 which do seem to have that Skt. word anumṛyate in them.

Huh! I wonder what has happened here? Has the Pali suffered some textual loss along the way, but the commentary remembers how it was read before? Or is this just a case of Western philologists getting Pali derivations wrong?


That’s interesting. I’m really not good in Magadhi linguistics, but vaguely I can imagine that the comparative lack of consonants in eastern Prakrit posed some problems for Sanskritization (leaving aside the issue if even before there was a process of Prakrit → Pali → Sanskrit).

So maybe some early Sanskrit translators got it right while Pali interpretors misinterpreted the term? Then it would be interesting of course if there are more cases like this where parallels feature diifererent interpretations of terms with similar orthography?

Hi Charles n all ,

I have read both the chinese SA15 /16 , sorry if i am being direct , bhante Analayo SA16 translation appears missing some terms .

Btw , what is the Pali term for
“underlying tendencies” ?

What makes this mind-bending is that we don’t know the original language SA was translated from. It could be some Central Asian language or Prakrit. Probably not straight Sanskrit in the early 5th century. So, I’m having to speak in equivalencies. Still, it’s spooky to see a reading in the Theravada commentary that agrees with SA.

I’ve seen a couple other cases of Pali readings not really making sense in parallel Agama passages. I noticed it with the expression dhamma-anudhamma not long ago working with MA.

Yes, Analayo struggled as much as I am with how to read these two sutras. I can tell by the translation. In SA 16, he decided that 增諸數 translates anumīyati, but the Pali commentary makes it clear to me that 隨使死 is equivalent to anumīyati.

My translation at the moment reads 隨使 as a verb and 使 and 死 as nouns (“follow the tendencies,” “follow death”), but that may be wrong, too. I’m thinking that 隨使使 should be 隨使, and 隨使死 should be 隨死, and they are both intended as verbs.

It would be the typical English translation of Skt. anuśaya. 隨使 and 使 were early Chinese translations; later on they switched to 隨眠 because 使 can mean “messenger” and other things.